SEARCH

SEARCH BY CITATION

Keywords:

  • carcinoma;
  • stroma;
  • myofibroblast;
  • actin;
  • lectin

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Tumor stroma is an active part influencing the biological properties of malignancies via molecular cross-talk. Cancer-associated fibroblasts play a significant role in this interaction. These cells frequently express smooth muscle actin and can be classified as myofibroblasts. The adhesion/growth-regulatory lectin galectin-1 is an effector for their generation. In our study, we set the presence of smooth muscle actin-positive cancer-associated fibroblasts in relation to this endogenous lectin and an in vivo competitor (galectin-3). In squamous cell carcinomas of head and neck, upregulation of galectin-1 presence was highly significantly correlated to presence of smooth muscle actin-positive cancer-associated fibroblasts in the tumor (p = 4 × 10−8). To pinpoint further correlations on the molecular level, we applied microarray analyses to the transcription profiles of the corresponding tumors. Significant correlations of several transcripts were detected with the protein level of galectin-1 in the cancer-associated fibroblasts. These activated genes (MAP3K2, TRIM23, PTPLAD1, FUSIP1, SLC25A40 and SPIN1) are related to known squamous-cell-carcinoma poor-prognosis factors, NF-κB upregulation and splicing downregulation. These results provide new insights into the significance of presence of myofibroblasts in squamous cell carcinoma.

Increasing attention is being turned to the stroma part in carcinomas. It is formed by fibroblasts producing the extracellular matrix, macrophages and other inflammatory cells as well as blood/lymphatic capillaries.1 Recent progress in cancer and skin cell biology has markedly changed our view on the functional significance of the stroma. Classically, the tumor stroma has been considered as a milieu suitable for growth of capillaries that guarantee the supply with oxygen and nutrients for the cancer epithelium. With the new data being accrued, the stroma can be described as an active microenvironment, which modulates the biology of the tumor including cancer stem (initiating) cells by mechanisms similar to the function of the niche in the case of tissue stem cells.2 As a consequence, a focus of this research is given to the study of mutual epithelial-mesenchymal interactions. This interplay is known to be crucial for embryonic development, and a major role in the route of tumor progression is likely.2, 3

The fibroblasts of the stroma, that is, cancer-associated fibroblasts (CAF), frequently contain fibers of smooth muscle α-actin (SMA and ACTA2), showing notable similarities to myofibroblasts responsible for the contraction of a wound in the course of a healing process.4 The origin of CAF is not well established: they can derive from the local mesenchyme by influence of the paracrine activity of cancer cells, by transition from other cell types present in the tumor site such as the tumor epithelium, pericytes, macrophages, endothelial cells5 or from mesenchymal stem cells, which have entered the tumor site.6 CAF are known to exert strong biological activities: fibroblasts isolated from tumors of squamous cell epithelia such as basal cell carcinoma of the skin or squamous cell carcinoma of head and neck are able to alter characteristics of the phenotype of normal epithelial cells and change their differentiation status into the direction of epidermal stem cells or cancer cells.7, 8 For example, CAF differ from normal fibroblasts in the gene expression profiles, with deviations detected for 560 genes, and in the secretion of cytokines strongly acting on epithelial cells.9 In addition to these well-known effectors, CAF also produce and secrete other classes of potent modulators of cells.

An emerging class of proteins relevant in this context is endogenous lectins from the galectin family. They share β-sandwich folding and affinity to a galactoside core and are capable to elicit regulation of adhesion and growth via carbohydrate/protein-protein interactions.10 The immunohistochemical monitoring of galectin presence in tumors has indicated cell type/differentiation-dependent profiles, which have given respective fingerprinting diagnostic and prognostic value.11–13

Previous studies have indicated that tumor stroma and also granulation tissue of wounds are frequently very rich for the homodimeric proto-type galectin-1 (Gal-1 and LGALS1).14, 15 At both sites, presence of myofibroblasts is documented, and SMA-positive CAF express Gal-1.16 Functionally important, an epithelial-stromal cross-talk has been inferred in pancreatic carcinoma involving a dual autocrine-paracrine signaling loop with Gal-1 and tissue plasminogen activator in the transition zone, contribution to cell invasion, migration and proliferation.17 Thus, Gal-1 secreted from CAF and tumor cells can be a mechanistically versatile constituent of this microenvironment. Secretion of Gal-1 and active cross-talk is also known from immune regulation exerted by regulatory T cells, with relevance for autoimmune diseases.18, 19 An influence on angiogenesis has been described, as also a pivotal role in the activity spectrum of the tumor suppressor p16INK4a.20, 21 In laryngeal tumors, it is a negative prognostic indicator.22 Sodium butyrate treatment, which induces cells to acquire the squamous phenotype, results in increased Gal-1 expression in the tumor, while retinoic acid treatment, which inhibits establishment of this phenotype, decreases Gal-1 presence.23

Galectin-3 (Gal-3 and LGALS3) is of interest in this context, because it can block Gal-1 activity on tumor cells.24, 25 Gal-3 presence is documented in squamous epithelia under physiological conditions and in cancer tissue.26, 27 The level of its expression increases with the differentiation status of head and neck squamous cell carcinomas (HNSCC) and correlates with keratinization.23, 28 These data on presence and functional divergence from Gal-1 prompted us to add monitoring for Gal-3 to our study.

Based on this body of knowledge, we analyzed occurrence of SMA-positive CAF (myofibroblasts) set in relation to both mentioned galectins in HNSCC, in normal tissue and in surgical margins (SM) using immunofluorescence. Because a relationship between biological properties of HNSCC and human papilloma virus (HPV) positivity has been established,29 occurrence of correlation between Gal-1, SMA-positive CAF and HPV was also tested. A strong association between SMA and Gal-1 but not Gal-3 was revealed. This result led us to define expression differences caused by the presence of Gal-1-positive stromal myofibroblasts on the transcription profile of the tumor.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Sample collection

The specimens were obtained from patients suffering from HNSCC after their informed consent in full agreement with the local ethical committee based on the Declaration of Helsinki. Samples from squamous cell carcinoma (SCC), SM of the resecate and normal epithelium (non-neoplastic epithelium) (NE) were collected from 31 patients. The tissue was protected by RNA-Later (Ambion, Austin, TX), deeply frozen in liquid nitrogen and stored at −85°C to prevent activity of endogenous RNases. Frozen sections (Cryocut-E, Reichert-Jung, Vienna, Austria) were used for RNA isolation and for immunocytochemical detection of SMA, Gal-1 and Gal-3.

Immunohistochemical detection of SMA, Gal-1, Gal-3, CD31 and MAP3K2

Cryocut sections were washed and rehydrated by phosphate-buffered saline (PBS, pH 7.2), fixed by paraformaldehyde in PBS for 5 min and, after extensive washing, subjected to processing to detect the proteins listed above. The mouse monoclonal antibody for the detection of SMA was purchased from DAKO (Glostrup, Denmark), the rabbit polyclonal antibody against MAP3K2 from LifeSpan BioSciences (Seattle, WA) and anti-CD31 from (Abcam, Cambridge, UK). Primary antibodies were diluted according to the recommendation of the supplier. Homemade rabbit antibodies against Gal-1 and Gal-3 were used in dilution of 1:50. These antibody fractions had been systematically tested for specificity and lack of cross-reactivity among the galectin family, with affinity depletion being performed by affinity chromatography in each positive case followed by another round of controls by ELISA.11, 30, 31 Fluorescein isothiocyanate-labeled swine anti-rabbit antibody (DAKO, Glostrup, Denmark) and tetramethylrhodamine isothiocyanate-labeled goat antimouse antibody (Sigma, Prague, Czech Republic), both diluted as recommended by supplier, were used as the second-step reagents. Nuclear DNA was visualized by 4′,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich, Prague, Czech Republic). Controls of specificity had been performed by omission/replacement of specific antibodies by irrelevant isotypic antibodies of preimmune serum. Specimens were mounted to Vectashield (Vector Laboratories, Burlingame, CA), then inspected by an Eclipse 90i fluorescence microscope (Nikon, Prague, Czech Republic) equipped with filterblocks for FITC, TRITC and DAPI and a Cool-1300Q CCD camera (Vosskühler, Osnabrück, Germany); images were analyzed by a LUCIA 5.1 computer-assisted image analysis system (Laboratory Imaging, Prague, Czech Republic) to evaluate the levels of fluorescence intensity in defined areas of specimens (epithelium, mesenchymal component) in arbitrary units [FI (AU)]. Background levels were estimated from a negative control and galectin-negative areas of experimental specimens.

Detection of HPV

Detection of HPV with high oncogenic potential (high-risk HPV group: HPV16, 19, 31, 33 and 45) was performed by OneStep reverse transcription RT-PCR kit (Qiagen, Hilden, Germany) on cRNA templates. The cRNA was prepared by linear amplification of mRNA and thus only transcriptionally active HPVs were detected. Primers specific for the selected HPV subtypes were designed to target the viral genome in the area encoding the E6 and E7 oncogenes and used as mixture in multiplexed PCR (see Supporting Information Table 1 for primer sequences and product lengths). The sensitivity of the procedure was tested on cRNA from HeLa cells (HPV-positive). Input amount of 50 ng of cRNA was estimated to be sufficient for the analysis. One-step RT-PCR was performed according to manufacturer's instructions using the following program: 50°C/30 min, 95°C/15 min, 40 cycles of (94°C/1 min, 50°C/1 min and 72°C/1 min) and 72°C/10 min. Possibility for cross-hybridization of primers was tested and was not observed under the given conditions.

Microarray analysis

The microarray analyses were performed on material from a subset of 21 patients (Table 1). Total RNA was isolated from the cryostat sections using RNeasy Micro Kit (Qiagen), checked for integrity, amplified and hybridized on an Illumina HumanWG-6 v3 Expression BeadChip (Illumina). The raw data were analyzed and processed using the beadarray package of the Bioconductor, as described previously.32 In short, the transcription profiles were background corrected using the normal-exponential model, quantile normalized and variance stabilized using base-2 logarithmic transformation. Transcriptionally active genes correlated to Gal-1 presence (detected by immunohistochemistry, IHC) were identified using the standard function (cor.test) of the R statistical environment (R Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org/). As we could not expect a linear correlation a priori, we used Spearman rank correlation as correlation method within the test. To remove spuriously correlated transcripts, an at least two-fold change difference in the gene's transcriptional activity over the range of Gal-1 presence was required. Storey's q-value33 was used to adjust p-values on multiple testing. Transcripts were considered correlated when q < 0.1. Detected transcripts were annotated using the beadchip manifest (HumanWG-6_V3_0_R2_11282955_A.bgx, Illumina). The transcription data are MIAME-compliant and deposited in the ArrayExpress database (accession #: E-MTAB-850).

Table 1. Characterization of the patient cohort
inline image

Statistical analysis

We have used Fisher's exact test for the univariate association studies between clinical markers and presence of SMA-positive CAF. Disease-free survival curves have been compared using Mantel-Cox log-rank test. Statistical significance was considered at the level of p = 0.05. Statistical analysis of the microarray data was performed as described above.

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Clinical characterization of the samples

The characteristics of the tumor samples used for immunohistochemical detection are listed in Table 1. It demonstrates that, among the 31 tumors, the majority was of the clinical stage T2 (45%) and histological grade G2 (48%). The lymph node stage was distributed approximately evenly among N0 (42%), N1 (23%) and N2 (35%). All patients were of the stage M0, that is, without distant organ metastases. Concerning the localization, tonsillar carcinomas prevailed (32%).

SMA-positive CAF were present in 61% of the SCC samples

When we evaluated NE and the SM of the resecate, we observed no SMA-positive myofibroblasts (Figs. 1a, 1b, 1e and 1f). The apparent signal came from SMA-positive smooth muscle cells in the wall of vessels, which can be considered as an internal control for the specificity of the histochemical reaction. Within the tumors, SMA-positive CAF were present in >60% of the tested SCC (19 patients of the total of 31; Table 1, Figs. 1d and 1g). The remaining tumors apparently did not contain these cells (12 patients of 31, Fig. 2c). These specimens exhibited a strong signal of SMA in the wall of vessels, only, with an exception of a few round cells per view field (e.g., three positive cells in Fig. 1b). The colocalization experiments, in which SMA and CD31 endothelial marker were detected simultaneously, verified that the SMA signal was in the stroma attributed dominantly to myofibroblasts and not to myoblasts of the wall of the small vessels (Supporting Information Fig. 1).

thumbnail image

Figure 1. Normal epithelium (NE) (a, e), the tissue area of the surgical margin of tumors (SM) (b, f) and minority of tumor tissues (c) contained no smooth muscle actin (SMA) (red signal)-positive myofibroblasts (MFB). The positive signal was exclusively observed in smooth muscle cells of vessels (white arrow). Majority (61%) of tumors contained CAF positive for SMA (d, g). No presence of Gal-1 (Gal-1, green signal) was observed in normal and marginal epithelial cells (ad). The connective tissue component of normal epithelium (a), the SM (b) and stroma with SMA-negative CAF (c) presented a very low but specific signal intensity for Gal-1 in contrast to tumors with SMA-positive CAF in the stroma (d). Weak but specific signal for presence of Gal-3, with intensity well above the background level, was detected in normal epithelial cells (e) and cells from the epithelium of the SM (f). Malignant epithelial cells were negative (g). No reactivity was observed in connective tissue of normal epithelium (e), the SM (f) and the tumor stroma (g). When the extent of presence of Gal-1 (black column) and Gal-3 (gray column) was compared in connective tissue of epithelium and tumor stroma (positive or negative for MFB) by measurements of fluorescence intensity (h), the stroma of tumors with SMA-positive CAF was stained significantly stronger than normal tissue or the SM (asterisk; paired t-test). The stromal signal for Gal-3 presence was very close to the background level (dashed line). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

thumbnail image

Figure 2. Lack of a statistically significant association between presence of SMA-positive CAF and TNM classification (a, b), histological grade (c), disease-free survival (d) and transcriptional activity of HPV in the tumor (e). The tumors with SMA-positive CAF tended to localize in palatile tonsils, while tumors with SMA-negative CAF localized to the oral cavity, pharyngeal wall and soft palate (f). p-Values of the Fisher's exact test (ac, e and f) and Mantel-Cox log-rank test (d) are given.

Download figure to PowerPoint

Association of the clinical data with SMA positivity of CAF

We observed no association between the clinical stage of the tumors or their histological grade with SMA positivity (Figs. 2a2c). We, however, observed indications, yet not of a statistically significant level, for an association of disease-free two-year survival among patients with SMA-positive CAF, in contrast to published data34 (Fig. 2d), possibly attributed to the low coverage of the survival data (Table 1). No statistically significant association was detected between HPV and SMA positivity of CAF (Fig. 2e). A statistically significant (p = 0.04, Fisher's exact test) majority of the tumors with SMA-positive CAF was present in SCC of palatine tonsils (Fig. 2f), while tumors with SMA-negative CAF tended to localize in the oral cavity, pharyngeal wall and soft palate.

Presence of Gal-1 correlated with presence of SMA-positive CAF and absence of Gal-3

Gal-1 was not detectable in normal and malignant epithelium (Figs. 1a1c). In contrast, a much stronger signal was seen in the stroma of all tumors with presence of SMA-positive CAF (Figs. 1d and 1h). The stroma of the tumors lacking SMA-positive CAF exhibited a very weak signal for Gal-1 (Figs. 1c and 1h) with fluorescence intensity levels comparable to the signal detected in normal tissues (Figs. 1a, 1b and 1h). When probing the Gal-3 presence, epithelial cells of NE and SM exhibited a distinct signal (Figs. 1e and 1f), while cancer cells were negative or they exhibited distinct, but a very weak signal just above the background level (Fig. 1g). No signal of Gal-3 was obtained in connective tissue of all three types of studied specimens including the tumors with SMA-positive CAF in their stroma (Figs. 1e1h).

Transcription data of the bulk tumor samples differed from the immunohistochemical signal intensities in a predicted way

The results of immunohistochemical inspection were compared to mRNA microarray analysis on a subset of 21 selected patients (Table 1). The transcription profiles were obtained from partially homogenized bulk samples, which included epithelial parts and other types of tissue. Overall, there are two sites, where SMA and Gal-1 genes are transcribed: smooth muscle cells of vessels20, 35 and stromal myofibroblasts.4, 16, 36 When evaluating correlations of the two factors, normal tissue with no stromal fibroblasts and hence only vascular production was expected to present such a correlation of transcriptional intensity for these two genes. Indeed, transcriptional activity of Gal-1 and SMA genes correlated very well in NE (p-value = 0.0002, Fig. 3a). On the contrary, in tumors, both vessels and stromal myofibroblasts were present, thus the correlation was lost (p = 0.4; Fig. 3a). The same held also true for other markers of smooth muscle cells and myofibroblasts (MYH11 and CDH11; Supporting Information Figs. 2a and 2b). Angiogenesis, as represented by VEGF and EDN1, was highly correlated (p = 0.001, 0.01, respectively) with level of Gal-1 transcription in tumors with SMA-negative CAF, but not in tumors with SMA-positive CAF (p = 0.9, 0.2, respectively, Figs. 3b and 3c). That implies that while in a tumor with SMA-negative CAF Gal-1 was produced only or predominantly in vessels, in tumors with SMA-positive CAF, another source of Gal-1 production was present, that is, the myofibroblasts. Transcription profiles of the SMA-positive CAF and their influence on the transcription profile of the tumors were apparently masked by the transcription profile of smooth muscle cells of vessels and could not be assessed directly. This apparent complexity could, however, be dissected by correlation analyses.

thumbnail image

Figure 3. Two sites of Gal-1 production in SMA-positive tumors. Correlation of the transcriptional activity for Gal-1 (LGALS1) and SMA (ACTA2) revealed differences between normal and tumor tissues. While the Pearson correlation of the log-intensities of the two transcripts in the normal tissue was highly significant (p = 0.0002), the correlation was lost in tumors (p = 0.4) (a). Transcriptional activity of vascular smooth muscle markers VEGFA (b) and EDN1 (c) correlated well with that of Gal-1 (LGALS1) in tumors with SMA-negative CAF but not in tumors with SMA-positive CAF. Statistical significance of the Pearson correlation of the log-intensities was evaluated, and the p-values are given. In the case of VEGFA, an outlier case (marked) had been removed from the statistical analysis. Considering this case, the p-value was 0.09.

Download figure to PowerPoint

Transcription of several genes in tumors was highly correlated to the presence of Gal-1 in myofibroblasts

To delineate changes of the tumor transcription profile caused by myofibroblasts only, we identified transcripts associated with the presence of SMA-positive CAF by correlation analysis of the microarray data with the detection of CAF-related Gal-1 at the protein level, as estimated by semi-quantitative immunohistochemical staining. Thus, only transcripts present in or induced by SMA-positive CAF were detected. Toward this end, we used data on 16 patients, for whom both microarray and immunohistochemical data (IHC) were available. We identified six transcripts significantly correlated with Gal-1 presence (as estimated by IHC). They were significantly upregulated in high-Gal-1 tissue samples with transcription varying at least two-fold over the range of Gal-1 intensities (see Table 2 for the list of the detected transcripts as well as Supporting Information Figs. 3 and 4 for the correlation plots). The measured upregulation of the production of the MAP3K2 transcript was ascertained immunohistochemically. A marked difference between tumors with SMA-positive and SMA-negative CAF was observed (Fig. 4).

thumbnail image

Figure 4. Detection of MAP3K2 (green signal) in tumors containing SMA (red signal) in CAF of the tumor stroma (a) and in tumors containing SMA exclusively in smooth muscle cells of vascular wall (b). A signal for MAP3K2 was confined to cancer cells of tumors characterized by the presence of SMA-expressing CAF. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Table 2. Transcripts significantly correlated with expression of Gal-1 in CAF
inline image

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

The presence of myofibroblasts, SMA-producing CAF, in the stroma of SCC has been associated with poor prognosis.5 In view of the emerging role of Gal-1 as effector in autocrine and paracrine communication between stroma and tumor parts in the local microenvironment we have focused our study on Gal-1. Here, we have detected several transcripts associated with Gal-1 production by the stromal myofibroblasts in head and neck SCC (Table 2). The majority of these detected transcripts directs synthesis of proteins involved in processes that have been associated with poor prognosis. Thus, this regulation within SMA-positive CAF is suggested to contribute to their activity at the clinical level.

Presence of myofibroblasts is associated with tumor localization

We have observed that the SMA-positive CAF-containing tumors are more often localized to palatine tonsils, while the SMA-negative CAF containing tumors localize more likely to the oral cavity, pharyngeal wall and soft palate (p = 0.04). No significant differences in TMN parameters, grading or HPV classification were seen.

Presence of Gal-1 correlates within the presence of myofibroblasts

The results demonstrate a distinct relationship between Gal-1 presence in the tumor stroma and presence of SMA-expressing CAF in tumors. Of note, we previously reported that exogenous Gal-1 has a stimulatory effect on the transition of fibroblasts to myofibroblasts under in vitro condition. The extent of this effect of Gal-1 is lower than the activity of both TGF-β1 and TGF-β3, but it has a strong additive effect when combined with these proteins.37 We have not observed any association of Gal-3 presence with the SMA-positive CAF.

Several cellular processes worsening disease prognosis are correlated with Gal-1 production by myofibroblasts

A small number of transcripts was detected, which significantly correlated with the presence of Gal-1 in myofibroblasts. These genes cover aspects of several important cancer-related processes, that is, downregulation of splicing (FUSIP1), NF-κB activation (TRIM23, PTPLAD1 and MAP3K2), ubiquitinylation (TRIM23), cell cycle control (SPIN1), autophagy (PTPLAD1), protection against mitochondrial oxidative stress (SLC25A40) and MAPK signaling (MAP3K2).

Downregulation of splicing

Chung et al.38 observed that splicing is negatively affected in poor-prognosis SCC of head and neck. We have detected upregulation of the gene for FUS interacting protein (FUSIP1) with Gal-1 presence. FUSIP1 is, despite its typical SR protein structure, a splicing repressor activated by dephosphorylation during the M phase and after a heat shock.39 Its principal function is to repress, rather than activate, splicing. FUSIP1 contains seven putative high-stringency 14-3-3 interacting domains, and it binds to ectopic 14-3-3σ (stratifin SFN).40 It then protects its binding partner from dephosphorylation and activation.41 Stratifin has been shown to be a poor-prognosis predictor in SCC, the mechanism of its action being yet unknown.38

NF-κB activation

Activation of nuclear factor κ-light-chain-enhancer of activated B cells (NF-κB) has recently been linked to poor prognosis in head and neck SCC.38 We have detected three transcripts (TRIM23, PTPLAD1 and MAP3K2) correlated with Gal-1 presence by SMA-positive CAF and related to NF-κB activation. As these transcripts are all produced in cardiac myocytes and smooth muscle cells,35, 42 they may be generated also directly by SMA-positive myofibroblasts and can cause the difference between the tumors with SMA-positive CAF and SMA-negative CAF by activation of NF-κB. That would link poor prognosis of tumors with activated NF-κB and the observation of association of poor-prognosis and SMA-positive CAF in SCC.5

Tripartite motif-containing 23 (TRIM23) is an E3 ubiquitin ligase, which adds ubiquitin to NF-κB essential modulator (NEMO). Consistently with this NEMO modification, TRIM23 expression considerably and dose-dependently regulates the NF-κB promoter activity.43 Mitogen-activated protein kinase kinase kinase 2 (MAP3K2 and MEKK2) is a MEK kinase for the ERK5 and JNK pathways,44 which is activated by EGF and stress stimuli. It regulates delayed activation of NF-κB in response to cytokine stimulation (TNF-α and IL-1α), being a candidate NEMO interactor as identified by protein microarray screening.45 MAP3K2 is also critical for both B-cell and T-cell development and interacts with FUS, an association partner of FUSIP1.46 A large-scale RNAi screen revealed that MAP3K2 regulates mitochondrial abundance and function.47 NF-κB-related factor is the protein tyrosine phosphatase-like A domain-containing protein (PTPLAD1). It has been described to belong to the set of butyrate-induced genes and is also known as B-ind1. Transfection experiments have shown that PTPLAD1 acts downstream of the small GTPase Rac1 in the pathway leading to NF-κB activation and potentiates JNK activation48 and is a member of the autophagy network.49

Other deregulated transcripts

Solute carrier family 25 member 40 (SLC25A40) is a putative manganese-containing superoxide dismutase, playing a critical role in protection against mitochondria oxidative stress. Thus, it is essential for cell survival50 and may play a role in tumor proliferation. Spindlin 1 (SPIN1) is an important mitotic spindle component. Its overexpression may disrupt cell cycle progression and can lead to tumorigenesis.51

Conclusions

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Distinct galectins and myofibroblasts appear to play pivotal roles in three distinct pathological processes: wound healing, fibrosis and carcinomas.34 On the cellular level, these processes share the stages of tissue damage, repair, remodeling and cell proliferation. It is reasonable to assume that tumors misuse pathways of tissue healing for their proliferation as predicted by Dvorak.52 To extend our knowledge on the potential of Gal-1 in this respect, we describe its expression profile and identify several transcripts that are associated with Gal-1 production by SMA-positive CAF. Of note, they are associated with biological processes proven to worsen disease prognosis. As Gal-1, directly, these gene products may qualify as targets to interfere with disease progression.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Authors are grateful to Iva Burdová, Vít Hajdúch, the service laboratory at IMG and especially to Martina Chmelíková for excellent technical assistance.

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information
  • 1
    Polyak K, Haviv I, Campbell IG. Co-evolution of tumor cells and their microenvironment. Trends Genet 2009; 25: 308.
  • 2
    Plzák J, Lacina L, Chovanec M, Dvořánková B, Szabo P, Čada Z, Smetana K, Jr. Epithelial–stromal interaction in squamous cell epithelium-derived tumors: an important new player in the control of tumor biological properties. Anticancer Res 2010; 30: 45562.
  • 3
    Josson S, Sharp S, Sung SY, Johnstone PA, Aneja R, Wang R, Gururajan M, Turner T, Chung LW, Yates C. Tumor-stromal interactions influence radiation sensitivity in epithelial- versus mesenchymal-like prostate cancer cells. J Oncol, 2010;Article ID 232831, 10 pages (doi:10.1155/2010/232831).
  • 4
    Franco OE, Shaw AK, Strand DW, Hayward SW. Cancer associated fibroblasts in cancer pathogenesis. Semin Cell Dev Biol 2010; 21: 339.
  • 5
    De Wever O, Demetter P, Mareel M, Bradle M. Stromal myofibroblasts are drivers of invasive cancer growth. Int J Cancer 2008; 123: 222938.
  • 6
    Mishra PJ, Mishra PJ, Humeniuk R, Medina DJ, Alexe G, Mesirov JP, Ganesan S, Glod JW, Banerjee D. Carcinoma-associated fibroblast-like differentiation of human mesenchymal stem sells. Cancer Res 2008; 68: 43319.
  • 7
    Lacina L, Dvořánková B, Smetana K, Jr., Chovanec M, Plzák J, Tachezy R, Kideryová L, Kučerová L, Čada Z, Bouček J, Kodet R, André S, et al. Marker profiling of normal keratinocytes identifies the stroma from squamous cell carcinoma of the oral cavity as a modulatory microenvironment in co-culture. Int J Radiat Biol 2007; 83: 83748.
  • 8
    Lacina L, Smetana K, Jr., Dvořánková B, Pytlík R, Kideryová L, Kučerová L, Plzáková Z, Štork J, Gabius H-J, André S. Stromal fibroblasts from basal cell carcinoma affect phenotype of normal keratinocytes. Br J Dermatol 2007; 156: 81929.
  • 9
    Strnad H, Lacina L, Kolář M, Čada Z, Vlček Č, Dvořánková B, Betka J, Plzák J, Chovanec M, Šáchová J, Valach J, Urbanová M, et al. Head and neck squamous cancer fibroblasts produce growth factors influencing phenotype of normal human keratinocytes. Histochem Cell Biol 2010; 133: 20111.
  • 10
    Kaltner H, Gabius H-J. A toolbox of lectins for translating the sugar code: the galectin network in phylogenesis and tumors. Histol Histopathol 2012; 27: 397416.
  • 11
    Čada Z, Smetana K, Jr., Lacina L, Plzáková Z, Štork J, Kaltner H, Russwurm R, Lensch M, André S, Gabius H-J. Immunohistochemical fingerprinting of the network of seven adhesion/growth-regulatory lectins in human skin and detection of distinct tumour-associated alterations. Folia Biol (Praha) 2009; 55: 14552.
  • 12
    Saussez S, de Leval L, Decaestecker C, Sirtaine N, Cludts S, Duray A, Chevalier D, André S, Gabius H-J, Remmelink M, Leroy X. Galectin fingerprinting in Warthin's tumors: lectin-based approach to trace its origin? Histol Histopathol 2010; 25: 54150.
  • 13
    Remmelink M, de Leval L, Decaestecker C, Duray A, Crompot E, Sirtaine N, André S, Kaltner H, Leroy X, Gabius H-J, Saussez S. Quantitative immunohistochemical fingerprinting of adhesion/growth-regulatory galectins in salivary gland tumours: divergent profiles with diagnostic potential. Histopathology 2011; 58: 54356.
  • 14
    Klíma J, Lacina L, Dvořánková B, Herrmann D, Carnwath JW, Niemann H, Kaltner H, André S, Motlík J, Gabius H-J, Smetana K, Jr. Differential regulation of galectin expression/reactivity during wound healing in porcine skin and in cultures of epidermal cells with functional impact on migration. Physiol Res 2009; 58: 87384.
  • 15
    Saussez S, Decaestecker C, Cludts S, Ernoux P, Chevalier D, Smetana K, Jr., André S, Leroy X, Gabius H-J. Adhesion/growth-regulatory tissue lectin galectin-1 in relation to angiogenesis/lymphocyte infiltration and prognostic relevance of stromal up-regulation in laryngeal carcinomas. Anticancer Res 2009; 29: 5966.
  • 16
    Wu M-H, Hong H-C, Hong T-M, Chiang W-F, Jin Y-T, Chen Y-L. Targeting galectin-1 in carcinoma-associated fibroblasts inhibits oral squamous cell carcinoma metastasis by downregulating MCP-1/CCL2 expression. Clin Cancer Res 2011; 17: 130816.
  • 17
    Roda O, Ortiz-Zapater E, Martínez-Bosch N, Guitiérrez-Gallego R, Villa-Perelló M, Ampurdanés C, Gabius H-J, André S, Andreu D, Real FX, Navarro P. Galectin-1 is a novel functional receptor for tissue plasminogen activator in pancreatic cancer. Gastroenterology 2009; 136: 137990.
  • 18
    Wang J, Lu Z-H, Gabius H-J, Rohowsky-Kochan C, Ledeen RW, Wu G. Crosslinking of GM1 ganglioside by galectin-1 mediates regulatory T cell activity involving TRPC5 channel activation: possible role in suppressing experimental autoimmune encephalomyelitis. J Immunol 2009; 182: 403645.
  • 19
    Wu G, Lu Z-H, Gabius H-J, Ledeen RW, Bleich D. Ganglioside GM1 deficiency in effector T cells from NOD mice induces resistance to regulatory T-cell supression. Diabetes 2011; 60: 23419.
  • 20
    Thijssen VLJL, Postel R, Brandwijk RJMGE, Dings RPM, Nesmelova I, Satijn S, Verhofstad N, Nakabeppu Y, Baum LG, Bakkers J, Mayo KH, Poirier F, et al. Galectin-1 is essential in tumor angiogenesis and is a target for antiangiogenesis therapy. Proc Natl Acad Sci USA 2006; 103: 1597580.
  • 21
    André S, Sanchez-Ruderisch H, Nakagawa H, Buchholz M, Kopitz J, Forberich P, Kemmner W, Böck C, Deguchi K, Detjen KM, Wiedenmann B, von Knebel Doeberitz M, et al. Tumor suppressor p16INK4a: modulator of glycomic profile and galectin-1 expression to increase susceptibility to carbohydrate-dependent induction of anoikis in pancreatic carcinoma cells. FEBS J 2007; 274: 323356.
  • 22
    Saussez S, Decaestecker C, Lorfevre F, Cucu D-R, Mortuaire G, Chevalier D, Wacreniez A, Kaltner H, André S, Toubeau G, Camby I, Gabius H-J, et al. High level of galectin-1 expression is a negative prognostic predictor of recurrence in laryngeal squamous cell carcinomas. Int J Oncol 2007; 30: 110917.
  • 23
    Gillenwater A, Xu X-C, El-Naggar AK, Clayman GL, Lotan R. Expression of galectins in head and neck squamous cell carcinoma. Head Neck 1996; 18: 42232.
  • 24
    Kopitz J, von Reitzenstein C, André S, Kaltner H, Uhl J, Ehemann V, Cantz M, Gabius H-J. Negative regulation of neuroblastoma cell growth by carbohydrate-dependent surface binding of galectin-1 and functional divergence from galectin-3. J Biol Chem 2001; 276: 3591723.
  • 25
    Sanchez-Ruderisch H, Fischer C, Detjen KM, Welzel M, Wimmel A, Manning JC, André S, Gabius H-J. Tumor suppressor p16INK4a: downregulation of galectin-3, an endogenous competitor of the pro-anoikis effector galectin-1, in a pancreatic carcinoma model. FEBS J 2010; 277: 355263.
  • 26
    Smetana K, Jr., Dvořánková B, Chovanec M, Bouček J, Klíma J, Motlík J, Lensch M, Kaltner H, André S, Gabius H-J. Nuclear presence of adhesion-/growth-regulatory galectins in normal/malignant cells of squamous epithelial origin. Histochem Cell Biol 2006; 125: 17182.
  • 27
    Saussez S, Decaestecker C, Mahillon V, Cludts S, Capouillez A, Chevalier D, Kaltner H, André S, Toubeau G, Leroy X, Gabius H-J. Galectin-3 upregulation during tumor progression in head and neck cancer. Laryngoscope 2008; 118: 158390.
  • 28
    Choufani G, Nagy N, Saussez S, Marchant H, Bisschop P, Burchert M, Danguy A, Louryan S, Salmon I, Gabius H-J, Kiss R, Hassid S. The levels of expression of galectin-1, galectin-3, and the Thomsen–Friedenreich antigen and their binding sites decrease as clinical aggressiveness increases in head and neck cancers. Cancer 1999; 86: 235363.
  • 29
    Rotnáglová E, Tachezy R, Saláková M, Procházka B, Košl'abová E, Veselá E, Ludvíková V, Hamšíková E, Klozar J. HPV involvement in tonsillar cancer: prognostic significance and clinically relevant markers. Int J Cancer 2011; 129: 10110.
  • 30
    Kaltner H, Seyrek K, Heck A, Sinowatz F, Gabius H-J. Galectin-1 and galectin-3 in fetal development of bovine respiratory and digestive tracts. Comparison of cell type specific expression profiles and subcellular localization. Cell Tissue Res 2002; 307: 3546.
  • 31
    Lohr M, Lensch M, André S, Kaltner H, Siebert HC, Smetana K, Jr., Sinowatz F, Gabius H-J. Murine homodimeric adhesion/growth-regulatory galectins-1, -2 and -7: comparative profiling of gene/promoter sequences by database mining, of expression by RT-PCR/immunohistochemistry and of contact sites for carbohydrate ligands by computational chemistry. Folia Biol (Praha) 2007; 53: 10928.
  • 32
    Szabo P, Kolář M, Dvořánková B, Lacina L, Štork J, Vlček Č, Strnad H, Tvrdek M, Smetana K, Jr. Mouse 3T3 fibroblasts under the influence of fibroblasts isolated from stroma of human basal cell carcinoma acquire properties of multipotent stem cells. Biol Cell 2011; 103: 23348.
  • 33
    Storey JD. The positive false discovery rate: a Bayesian interpretation and the q-value. Ann Stat 2003; 31: 201335.
  • 34
    Kalluri R, Zeisberg M. Fibroblasts in cancer. Nat Rev Cancer 2006; 6: 392401.
  • 35
    Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, Block D, Zhang J, Soden R, Hayakawa M, Kreiman G, Cooke MP, Walker JR, et al. A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci USA 2004; 101: 60627.
  • 36
    Grinnell F. Fibroblasts, myofibroblasts, and wound contraction. J Cell Biol 1994; 124: 4014.
  • 37
    Dvořánková B, Szabo P, Lacina L, Gal P, Uhrová J, Zíma T, Kaltner H, André S, Gabius H-J, Syková E, Smetana K, Jr. Human galectins induce conversion of dermal fibroblasts into myofibroblasts and production of extracellular matrix: potential application in tissue engineering and wound repair. Cells Tissues Organs 2011;194:46980.
  • 38
    Chung CH, Parker JS, Ely K, Carter J, Yi Y, Murphy BA, Kian Ang K, El-Naggar AK, Zanation AM, Cmelak AJ, Levy S, Slebos RJ, et al. Experimental therapeutics, molecular targets, and chemical biology: gene expression profiles identify epithelial-to-mesenchymal transition and activation of nuclear factor-κB signaling as characteristics of a high-risk head and neck squamous cell carcinoma. Cancer Res 2006; 66: 82108.
  • 39
    Shin C, Feng Y, Manley JL. Dephosphorylated SRp38 acts as a splicing repressor in response to heat shock. Nature 2004; 427: 5538.
  • 40
    Benzinger A, Muster N, Koch HB, Yates JR, Hermeking H. Targeted proteomic analysis of 14–3-3*, a p53 effector commonly silenced in cancer. Mol Cell Proteomics 2005; 4: 78595.
  • 41
    Shi Y, Manley JL. A complex signaling pathway regulates SRp38 phosphorylation and pre-mRNA splicing in response to heat shock. Mol Cell 2007; 28: 7990.
  • 42
    Wu C, Orozco C, Boyer J, Leglise M, Goodale J, Batalov S, Hodge C, Haase J, Janes J, Huss J, Su A. BioGPS: an extensible and customizable portal for querying and organizing gene annotation resources. Genome Biol 2009; 10: R130.
  • 43
    Arimoto K, Funami K, Saeki Y, Tanaka K, Okawa K, Takeuchi O Akira S, Murakami Y, Shimotohno K. Polyubiquitin conjugation to NEMO by triparite motif protein 23 (TRIM23) is critical in antiviral defense. Proc Natl Acad Sci USA 2010; 107: 1585661.
  • 44
    Sun W, Wei X, Kesavan K, Garrington TP, Fan R, Mei J, Anderson SM, Gelfand EW, Johnson GL. MEK kinase 2 and the adaptor protein Lad regulate extracellular signal-regulated kinase 5 activation by epidermal growth factor via Src. Mol Cell Biol 2003; 23: 2298308.
  • 45
    Fenner BJ, Scannell M, Prehn JHM. Expanding the substantial interactome of NEMO using protein microarrays. PLoS One 2010; 5: e8799.
  • 46
    Bandyopadhyay S, Chiang C-Y, Srivastava J, Gersten M, White S, Bell R, Kurschner C, Martin CH, Smoot M, Sahasrabudhe S, Barber DL, Chanda SK, et al. A human MAP kinase interactome. Nat Methods 2010; 7: 8015.
  • 47
    Yoon JC, Ng A, Kim BH, Bianco A, Xavier RJ, Elledge SJ. Wnt signaling regulates mitochondrial physiology and insulin sensitivity. Genes Dev 2010; 24: 150718.
  • 48
    Courilleau D, Chastre E, Sabbah M, Redeuilh G, Atfi A, Mester J. B-ind1, a novel mediator of Rac1 signaling cloned from sodium butyrate-treated fibroblasts. J Biol Chem 2000; 275: 173448.
  • 49
    Behrends C, Sowa M, Gygi SP, Harper JW. Network organization of the human autophagy system. Nature 2010; 466: 6876.
  • 50
    Haitina T, Lindblom J, Renström T, Fredriksson R. Fourteen novel human members of mitochondrial solute carrier family 25 (SLC25) widely expressed in the central nervous system. Genomics 2006; 88: 77990.
  • 51
    Yuan H, Zhang P, Qin L, Chen L, Shi S, Lu Y, Yan F, Bai C, Nan X, Liu D, Li Y, Yue W, et al. Overexpression of SPINDLIN1 induces cellular senescence, multinucleation and apoptosis. Gene 2008; 410: 6774.
  • 52
    Dvorak HF. Tumours: wounds that do not heal. N Engl J Med 1986; 315: 16509.

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_27550_sm_SuppFigures.pdf2900KSupporting Information Figures
IJC_27550_sm_SuppTab1.pdf28KSupporting Information Table 1

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.