Absence of calretinin protein expression in malignant mesotheliomas from asbestos-exposed NF2+/− mice and mouse mesothelioma cell lines from various mouse strains

Absence of calretinin protein expression in malignant mesotheliomas from asbestos-exposed... Background: Calretinin is the most widespread positive marker for the immunohistochemical identification of malignant mesothelioma (MM) and was proposed to serve as a blood-based biomarker. Functionally, evidence has accumulated that calretinin might be implicated in MM tumorigenesis. We aimed to identify calretinin (CR; Calb2)in +/− murine MM and reactive mesothelial cells in granuloma from asbestos-exposed NF2 mice, a line heterozygous for the tumor suppressor merlin (NF2), used as a mouse MM model. Additionally, we sought to ascertain the presence of calretinin in MM cell lines from other mouse strains. We also intended to investigate the role of calretinin in +/− +/- −/− mesotheliomagenesis by comparing the survival of asbestos-exposed NF2 and NF2 CR mice. +/− +/- −/− Methods: NF2 and NF2 CR mice, both lines on a C57Bl/6J background, were exposed to asbestos following an +/− established protocol. Tumor histology and asbestos-induced mortality were assessed. MM and granuloma from NF2 mice were analyzed with immunohistochemical methods for calretinin expression. Levels of Calb2 mRNA and calretinin expression in tumors and MM cell lines of various mouse strains were determined by RT-qPCR and Western blot analysis, respectively. +/− +/− Results: No expression of calretinin at the protein level was detected, neither in MM from NF2 mice, NF2 MM- derived cell lines nor immortalized mesothelial cells of mouse origin. At the mRNA level we detected Calb2 expression +/− +/- −/− in MM cell lines from different mouse strains. Survival of NF2 and NF2 CR mice exposed to asbestos showed no significant difference in a log-rank (Kaplan-Meier) comparison. Conclusions: The concomitant determination of calretinin and mesothelin blood levels has been proposed for early detection of human MM. Mouse MM models based on asbestos exposure are assumed to yield helpful information on the time course of appearance of mesothelin and calretinin in the blood of asbestos-treated mice determining the +/− earliest time point for interventions. However, the observed absence of calretinin in MM from NF2 mice and derived cell lines, as well as from MM cells from Balb/c and C3H mice likely precludes the use of calretinin as a biomarker for mouse MM. The results also indicate possible species differences with respect to an involvement of calretinin in the formation of MM. +/− Keywords: Malignant mesothelioma, Calretinin, Calb2, C57Bl/6J, NF2 mice, Asbestos * Correspondence: beat.schwaller@unifr.ch Anatomy, Section of Medicine, University of Fribourg, Route Albert-Gockel 1, CH-1700 Fribourg, Switzerland Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Blum et al. Biomarker Research (2018) 6:19 Page 2 of 7 Background intraperitoneally (i.p.) with 400 μg of crocidolite Malignant mesothelioma (MM) is an extremely aggres- suspended in 500 μL PBS every 3 weeks for a total of eight sive neoplasm linked to asbestos exposure [1]. Median rounds of injection (i.e., a total of 3.2 mg of crocidolite per survival is very short (9–12 months), likely linked to mouse as described before [12]). Unio Internationale diagnosis at late stages of the disease. Correct patho- Contra Cancrum (UICC) grade crocidolite asbestos was logical diagnosis is of outmost importance and calretinin obtained from SPI Supplies (West Chester, PA). For has been described as one of the most sensitive and se- survival analysis, mice were kept for a maximum of lective positive markers for human MM [2, 3]. Calretinin 33 months. (CR; human gene symbol: CALB2) is a calcium-binding Four NOD/SCID gamma mice were injected i.p. with high protein of the EF-hand family with several reported a suspension (200 μl PBS) of FACS-sorted ZL55-SO functions, see [4]. Together with mesothelin, the detec- cells (100,000/mouse). These genetically-modified hu- tion of CR in blood samples by ELISA has been sug- man MM cells are identified by their increased endogen- gested as a potential predictor to detect MM at earlier ous expression of the stem cell markers SOX2 and stages [5, 6]. Other discussed MM biomarkers include OCT4 driving the reporter eGFP, the latter allowing for soluble mesothelin-related peptides, megakaryocyte the isolation of this small (< 5%) subpopulation [15]. high potentiating factor, osteopontin, fibulin-3 and high ZL55-SO cells show an increased tumor-initiating mobility group protein B1 (HMGB1) [7]. However, capacity in vivo compared to the parental ZL55 cells. the time course of appearance of these biomarkers Mice were sacrificed after 5 weeks and tissue was fixed during human MM pathogenesis is currently un- in 4% paraformaldehyde. Histopathological diagnosis known. Animal models (mostly mice) are widely used and analyses were performed on paraffin-embedded as models to address such questions. samples as described before [16]. All animal experiments Common genomic alterations and similar genomic were performed with the permission of the local animal profiles, e.g. loss of tumor suppressors (NF2, LATS2, care committee (Canton of Fribourg, Switzerland) and BAP1) in murine and human MM emphasize the rele- according to the present Swiss law and the European vance of mouse model systems to study mesothelioma- Communities Council Directive 86/609/EEC. genesis [8] and to possibly validate suggested MM biomarkers. Such genetic mouse MM models include Western Blotting mice heterozygous for NF2, one of the most often al- Proteins from murine MM cell lines and from C57- +/− tered tumor suppressor genes with up to 44% alterations NF2 tumor samples were prepared as described +/ in human MM cell lines [9, 10]. Asbestos-exposed NF2 before [16]. Briefly, protein extracts from MM cell lines mice were reported to be an important murine model (AB12, AK7, RN5 [16], RN29) were separated by SDS- recapitulating many molecular features of human MM PAGE (10% gel) and transferred onto a nitrocellulose and to be relevant for further investigations on MM membrane. The membrane was incubated with primary pathogenesis mechanisms, as well as for preclinical antibody against calretinin (rabbit polyclonal CR7696 testing of putative novel therapeutic approaches [11, 12]. [17]), previously available from SWANT, Marly, The aim of this study was to investigate the expression Switzerland at a dilution of 1:10,000. Secondary biotinyl- +/− of calretinin in I) murine MM originating from NF2 ated anti-rabbit IgG at a dilution of 1:10,000 for 2 h at mice; II) MM cell lines derived from these mice and III) RT was used. For signal detection the ABC system murine MM cell lines from strains other than C57Bl/6J, (Vectastain, Vector Laboratories, Burlingame, CA) was +/− the genetic background of the NF2 mice. In addition, applied and chemiluminescent signals were revealed with to investigate any potential role for CR in murine MM the HRP substrate (Millipore, Luminata Forte); signals were formation we assessed whether overall survival differed detected with the Western blot reader (FluorChem E +/- between asbestos-exposed NF2 mice with or without a System, Bucher Biotec, Basel, Switzerland) as previ- functional Calb2 gene. ously described [18]. Methods Histological analysis Animal studies Tumor samples were fixed in 4% paraformaldehyde in PBS, 129S2/SvPas Nf2+/− mice (strain name: B6;129S2- pH 7.4, dehydrated and embedded in paraffin followed by tm1Tyj Nf2 /J; JAX stock #008190) were backcrossed for ≥6 sectioning with a microtome. Tumor samples were patho- generations to wild-type C57Bl/6J (C57) mice to obtain logically evaluated using hematoxylin/eosin or Goldner- +/− a C57-NF2 mouse line. These mice were crossed with stained sections. For immunohistochemistry, sections were −/− C57-CR mice [13] to give rise to the genotype C57- deparaffinized and incubated in antigen-retrieval solu- +/- −/− NF2 CR ; animals were genotyped as described tion using sodium citrate, pH 6, and processed as previously [14]. Mice, 6–8 weeks of age, were injected described before [19]. Primary antibodies (CR7696, Blum et al. Biomarker Research (2018) 6:19 Page 3 of 7 CR7697 or CR7699/4; SWANT, Marly, Switzerland) were CR staining in the developing epithelial cells of the lung. used at a 1:500 dilution with overnight incubation at 4 °C. A control section stained without primary CR antibody was negative (Fig. 1h) including the single layer of meso- Semi-quantitative RT-PCR of Calb2 mRNA levels thelial cells (inset; arrows). The expression of murine Calb2 mRNA was detected by In order to confirm the IHC results we performed semi-quantitative RT-PCR. Murine mesothelioma cell lines Western blot analysis to detect CR expression in murine were cultured under standard conditions [16]and total MM cell lines. Even when loading ≥50 μgofprotein no RNA extracted with PeqGold Trifast (Axonlab, Le Mont- specific CR signal at approximately 30 kDa was detected sur-Lausanne, Switzerland) following the manufacturer’s in all mouse cell lines (Fig. 2a) and in immortalized meso- protocol. cDNA was prepared from 500 ng total RNA using thelial cells from a C57 mouse (iMeso-WT1 [16, 21]; not the QuantiTect Reverse Transcription kit (Qiagen, Stockach, shown), while a signal was evident in extracts from the Germany) by following manufacturer’sinstructions. As human MM cell line MSTO-211H, as well as in mouse housekeeping gene the ribosomal protein 13S was used and cerebellum, described to be rich in CR [22](Fig. 2a). A Calb2 expression was assessed using the following primers: protein extract from isolated primary mesothelial cells of a −/− Calb2 -FW: 5′-GAA ATG GGT ACA TTG AAG GTA-3′, C57-CR mouse served as negative control. Calb2 was Calb2 RV 5’-CCA TCT GAG TTC TTC TTA TCA TAC detected at the mRNA level in all tested murine MM cell 5′,RPS-13FW5’CGA AAG CAT CTT GAG AGG AAC A lines and in mesothelial iMeso-WT1 cells (Fig. 2b), but 3′,RPS-13RV5’-TCG AGC CAA ACG GTG AAT C-3′. not in mouse primary mesothelial cells maintained in cul- ture for 24 h, the latter as reported before [14]. Statistical analysis In human MM, calretinin is considered to be the most GraphPad Prism version 7 and R statistical software sensitive and selective marker for the diagnosis, in par- were used to perform the log-rank test (Mantel-Cox ticular of the epithelioid and the epithelioid part of the test) on the Kaplan-Meier survival plot for comparison. biphasic MM type [3]. Furthermore, its essential func- tion in vitro in human MM cell lines [18], as well as its Results function in protection from asbestos-induced cytotox- +/- The repetitive injection of crocidolite in C57-NF2 icity has been demonstrated before [23]. While the clin- mice resulted in approximately 10% of confirmed ical relevance of calretinin IHC in the diagnosis of murine mesothelioma cases and in a multitude of human MM is undisputed, the putative role of calretinin granuloma formation in treated animals [12]. in the development of MM in vivo has remained elusive. Granuloma were described to be rich in reactive Thus, mouse survival after asbestos exposure was tested mesothelial cells and human reactive mesothelial cells in a well-studied MM mouse model, i.e. in mice hetero- were previously shown to be strongly positive for CR zygous for the tumor suppressor NF2 [11] either wild- +/− +/- −/− [20]. Human ZL55 MM cells injected i.p. into NOD/ type (C57-NF2 ) or null-mutant (C57-NF2 CR ) SCID gamma mice were used as a positive control. for the Calb2 gene. Long-term survival data were ZL55-derived tumors showed strong calretinin IHC analyzed by the standard Kaplan-Meyer method and staining, clearly a differentiating marker for the tumor differences between groups were assessed using a log- nodules located in connective tissue often consisting of rank test. Mice (48 in total) of the two genotypes were +/- +/+ +/- −/− CR-negative nodules of adipose tissue (Fig. 1a-c). None treated with asbestos (NF2 CR , n = 18; NF2 CR ; +/− of the C57-NF2 mice-derived MM cells and the react- n = 30). A total of 17 (94%) mice (2 with diagnosed MM) +/- +/+ ive mesothelial cells on the peritoneal side of the dia- died in the NF2 CR group (censored, n = 1) and 28 +/- −/− phragm from the same mice showed CR expression in (93%; 3 with diagnosed MM) died in the NF2 CR the IHC analysis, representative examples are shown in group (censored, n = 2) until the end of the long-term Fig. 1d-e. As previously described [14], the developing follow up in the 2 cohorts. Kaplan-Meier curves were lung of the murine embryo shows staining for CR in the similar in both groups. Median survival was 14. visceral mesothelial layer and in a fraction of mesenchy- 22 months (95% confidence interval: 9.23–19.07) in mice +/- +/+ mal cells (Fig. 1f), even though staining is weaker com- with a functional Calb2 gene (NF2 CR ) and 14. pared to human mesothelioma [14]. Also in the 63 months (95% confidence interval 13.4–17.3) in mice +/- −/− developing human lung, cells of the visceral pleura, i.e. with a non-functional Calb2 gene (NF2 CR ). A log- the outermost mesothelial cell layer, as well as a subpop- rank test indicated no significant difference in survival +/- +/+ ulation of stromal (mesenchymal) cells stained positive (p = 0.7345). The log-rank hazard ratio (NF2 CR for CR, in particular evident as brown-stained cytoplasm used as reference) was 0.90 (95% confidence interval 0. in mesothelial cells in Fig. 1g (inset; arrows). Few stro- 49–1.66). This indicates that mice with a functional mal cells in the mesenchyme showed very strong stain- Calb2 gene have a marginally higher, however ing (Fig. 1g, arrowheads). Note the absence of specific insignificant, risk of death after asbestos exposure. Blum et al. Biomarker Research (2018) 6:19 Page 4 of 7 high Fig. 1 Immunohistochemical analysis of human and mouse MM and embryonic mouse and human lung. ZL55-SO cells were injected i.p. into NOD/SCID gamma mice and tumor cells show strong expression of CR (b; higher magnification inset in c), arrow pointing towards a strong CR-positive tumor nodule. The negative control (parallel section) is shown in (a). d) Representative example of a mouse MM showing absence of CR expression, as well as in another example of an asbestos-induced lymphocytic and granulomatous inflammation (e) of the parietal pleura. f) In the developing lung of a mouse, CR protein is already expressed in the visceral pleura at day 10 (arrows) as reported before [14]. g) Expression of CR in the mesothelial cell layer (arrows) and in scattered stromal (mesenchymal) cells of the developing human lung (arrowheads). h) Negative control on parallel section incubated without primary CR antibody. The inset shows CR-negative mesothelial cells (arrows) at 2-fold higher magnification. The few round lightly brownish-stained cells are erythrocytes, where the endogenous peroxidase activity was not completely blocked. Abbreviations: A adipose tissue; T tumor; S stroma; G granulomatous inflammation; D pleura parietalis adjacent to the diaphragm; L lung. Scale bars: A and B: 2.5 mm, C: 500 μm. D-F: 100 μm; G,H: 50 μm; insets with layer of mesothelial cells: 2-fold magnification compared to G,H Discussion (e.g. ZL34, ONE58), as well as biphasic MM (e.g. Our results indicate that MM derived from asbestos- MSTO-211H, Mero83) show a clear and in few cases +/− exposed C57-NF2 mice do not express calretinin relatively strong CR expression evidenced by Western protein levels detectable by either IHC or Western blot blot analysis [18, 25]. Our negative CR IHC results analysis. No positive Western blot signal was observed are in contrast to previous reports showing positive in RN5 and RN29 MM cells derived from MM tissue of CR IHC staining in murine pleural MM samples. In +/− the same asbestos-exposed C57-NF2 mouse strain; toxicology studies, Yokohira et al. reported positive yet, RN5 and RN29 cells express Calb2 mRNA. CR immunostaining of the lung surface in female A/J Likewise, well-established murine MM cell lines AK7 mice exposed to 4-(methylnitrosamino)-1-(3-pyridyl)- [24] also from C57 mice, and AB12 cells from Balb/c 1-butanone (NNK) [26] and to potassium octatitanate mice were negative for CR protein expression and (TISMO) fibers [27]. In a later study, also mice from add- positive for Calb2 mRNA. Based on the previous obser- itional strains (C3H/HeN, ICR and C57BL/6), exposed to vation of a very strong positive correlation between CR TISMO showed a thickening of the visceral pleura due to protein and CALB2 mRNA levels in 13 human MM cell mesothelial cell infiltration with part of the cells showing lines, i.e. regulation mostly at the transcriptional level strong CR immunoreactivity (Fig. 3 in [28]). The differ- (see Fig. 1c in [25]), we assume levels of Calb2 mRNA ences between previous results and the current ones with to be low in mouse MM cell lines and MM tissue with respect to CR IHC might be several-fold. IHC results no detectable CR protein levels. However, species differ- might be related to differences in I) the materials used to ences in CR regulation may not be entirely excluded. induce changes in the tunica serosa (NNK and TIMSO fi- Thus, in contrast to human MM, where calretinin is bers vs. crocidolite), II) the site of injection (pleura vs. considered as a positive marker, we did not detect calre- peritoneum) and III) the CR antibodies used. With respect tinin at the protein level in MM and derived cell lines of to strain differences, Yokohira et al. also reported positive mouse origin. Even human sarcomatoid MM cell lines CR immunoreactivity in C57Bl/6 mice (Fig. 3 in [28]), Blum et al. Biomarker Research (2018) 6:19 Page 5 of 7 Fig. 2 Western Blot analysis for calretinin and RT-PCR for the detection of Calb2 mRNA. a) Western blot signals (upper panel) and the corresponding Ponceau Red-stained membrane (lower panel). As positive control, 10 ng of purified human recombinant CR was used. Sizes (molecular masses) of the marker proteins in the right lane are (from top to bottom): 130, 100, 75 (red), 63, 48, 35, 28 (green), and 17 kDa. The specific signal for CR runs just slightly above the green marker protein at approximately 30 kDa. In MSTO-211H cell extracts, a strong signal at the same position as purified CR indicates the presence of CR in these cells. All mouse samples from the MM cell lines RN29, RN5 and AK7 (all from C57Bl/6 J −/− mice), AB12 (Balb/c) are negative, as well as the negative control prMC from a CR mouse (prMC−/−). Mouse cerebellum extract (most right) shows a strong signal for CR. b) Positive signals for Calb2 mRNA were detected in RN29, RN5, AB12, AK7 and iMeso-WT cells; no signal was present in samples from freshly isolated C57 mouse prMC and in the negative control (H O). RPS13 was used as housekeeping gene while our tissue samples and cell lines (RN5, RN29, AK7) derived from C57Bl/6 mice were negative for CR. Of note, in the same study by Yokohira et al., weak CR positivity was also detectable in the healthy lung tissue, especially in C57Bl/6 J mice [28], while lung tissue was always com- pletely negative in our study, in line with many studies reporting that normal lung tissue is considered CR- negative. In addition, the specificity of the CR antibody to unequivocally detect mouse CR was not reported in the previous studies, while we have additionally confirmed the specificity of our CR antibodies by Western blotting. A role for calretinin in the development of human MM had been suggested based on in vitro studies. Induced expression of SV40 early gene products augments +/- +/+ Fig. 3 Kaplan-Meier survival analysis. C57-NF2 CR (blue line; n = 18) CR expression in the human mesothelial cell line MeT- +/- −/− and C57-NF2 CR (red line; n = 30) mice were injected with 8 × 5A, which leads to increased resistance to asbestos- 400 μg of crocidolite. The survival curves of the 2 groups (all mice) are mediated acute cytotoxicity, an effect mediated by the statistically not different (log-rank test, p =0.7345) AKT/PI3K pathway [23]. Inversely, down-regulation of Blum et al. Biomarker Research (2018) 6:19 Page 6 of 7 CR decreases the viability and proliferation of human MM selectively in human MM might point towards different cells in vitro, most evident in epithelioid and biphasic cell (possibly none) functions in the process of murine lines, but to a lesser extent also in MM cell lines with mesotheliomagenesis compared to the formation of MM prevalent sarcomatoid morphology [18]. Moreover, pri- in humans. Of note, the presumed role of CR in human −/− mary mesothelial cells from CR mice are characterized MM development is currently based on in vitro studies by a lower proliferation rate and are slower to close a with human MM cell lines. III) The low CR levels con- scratch evidenced in a 2D-wound assay compared to pri- ceivably present in mouse MM (below the detection mary mesothelial cells from wild-type mice [14]. This limit of IHC or Western blot analysis used in this study) −/− phenotype disappears, if CR is overexpressed in CR still might be sufficient to exert a role, e.g. as a signaling prMC. Based on these results we hypothesized to possibly molecule during mouse MM development. IV) Various observe a genotype-dependent effect in the survival of in Calb2 mRNA transcripts, in part derived from alterna- +/- −/− vivo asbestos-exposed C57-NF2 CR compared to tive splicing (so far detected only in human MM), might +/− mice with a functional Calb2 gene (C57-NF2 mice). have a function on their own as recently proposed [30] However, survival of mice in both groups was not and thus Calb2 transcripts might possibly be implicated different. Of note, in our survival analysis all mice were in murine MM tumorigenesis. All of the above points included, not only the ones with a diagnosed bona need to be considered, when using mouse MM models fide MM. Given the low penetrance in our model, as with the aim to possibly find strategies that might be +/− reported before (≈10% in C57-NF2 mice [12]), i.e. applied for developing novel human therapeutic ap- +/- −/− 3/30 in the C57-NF2 CR mouse group and 2/18 proaches. This is particularly the case when using mouse +/- +/+ in the C57-NF2 CR mouse group in this study, models aimed at exploiting the putative role of CR in the small number of confined MM cases doesn’t the pathogenesis of human MM. In such a case, xeno- allow to directly link Calb2 genotype to MM-caused graft models with human CR-positive MM cells might mortality. Thus, including all mice might have be considered as the current method of choice. With masked some subtle Calb2 genotype-dependent differ- respect to validating CR as a biomarker for early de- ences that are only observable in mice with a diag- tection of human MM (currently less than 5% are di- nosed MM. However, the deaths in both groups were agnosed at an early stage 1A [31]), one needs to be asbestos exposure-dependent, since unexposed mice aware that CR is unlikely of any predictive value in of both genotypes had a much higher median survival mouse MM models. Even if CR is expressed at very +/- +/+ (control C57-NF2 CR mice) of 29.13 months (95% low protein levels in mouse MM, CR plasma levels confidence interval: 5.80–32.93, log-rank test: p = 0.0008). are most probably below the detection limit of even This leads to several different hypotheses. I) The heavy most sensitive ELISA techniques. asbestos exposure by direct intraperitoneal fiber injec- tion resulting in relatively short latency for MM devel- Conclusions opment in mice (around 1 year in mice compared to Our results indicate the absence of CR (at least below 20–40 years in humans) leads to other routes of MM the threshold of Western blot detection) in mouse MM formation in human and mice. This is partially sup- cell lines, in line with negative IHC staining in asbestos- ported by the observation of mostly sarcomatoid and bi- induced mouse MM tissue. These findings essentially phasic MM in mice compared to human, where the rule out the use of CR as a putative biomarker for early largest number of MM cases is of the epithelioid type. detection of MM in mouse models, while this may well However, frequency of gene loci deletions containing be appropriate for human MM. Together with other di- genes likely involved in MM formation were reported to vergent results in human and mouse MM, e.g. with re- be similar in samples from human MM and asbestos- spect of MM histotype or commonly mutated genes, +/− exposed NF2 mice including CDKN2A, CDKN2B, it requires careful consideration, when translating find- CHD5, TP53, and NF2 [8]. In contrast, 14/15 cell lines ings obtained in one species to another. derived from asbestos-injected mice of several mouse Abbreviations strains showed homologous deletions of the CDKN2A C57: C57Bl/6J; CR: Calretinin; HMGB1: High mobility group B1 protein; locus, but no chromosomal losses in either the BAP1 MM: Malignant mesothelioma; NF2: Neurofibromatosis 2 and NF2 regions and moreover no mutations in these 2 Acknowledgements genes as well as in LATS2, commonly mutated in human We would like to thank L. Wu, University of Toronto for the help in the initial MM [29]; this might again hint towards species differ- collecting of the tumor samples. The excellent technical work of S. Eichenberger, A. Oberson and M. Sanchez, University of Fribourg is highly appreciated. ences in the pathways leading to MM II). Although CR expression is prevalent during both mouse [14] and hu- Funding man lung embryonic development (Fig. 1g, h), CR re- The project was supported by the Swiss National Science Foundation (SNSF appearance at levels detectable by Western blot analysis Sinergia grant CRSII3 #147697) and intramural funding from the University of Blum et al. Biomarker Research (2018) 6:19 Page 7 of 7 Fribourg. The funding body was not involved in the design of the study, the 11. Altomare DA, et al. A mouse model recapitulating molecular features of collection, analysis, and interpretation of data and in writing of the manuscript. human mesothelioma. Cancer Res. 2005;65:8090–5. 12. Rehrauer H, Wu L, Blum W, Pecze L, Henzi T, Serre-Beinier V, et al. How asbestos drives the tissue towards tumors: YAP activation, macrophage and Availability of data and materials mesothelial precursor recruitment, RNA editing, and somatic mutations. All data generated or analyzed during this study are included in this published Oncogene. 2018; doi.org/10.1038/s41388-018-0153-z article. A limited amount of paraffin sections from the used mouse tissue 13. Schurmans S, Schiffmann SN, Gurden H, Lemaire M, Lipp HP, Schwam V, et al. samples are available upon request. Impaired long-term potentiation induction in dentate gyrus of calretinin- deficient mice. Proc Natl Acad Sci U S A. 1997;94:10415–20. Authors’ contributions 14. Blum W, Pecze L, Felley-Bosco E, Schwaller B. Overexpression or absence of BS and WB conceived the study. WB, TH, HECS, LP, JWR and BV performed calretinin in mouse primary mesothelial cells inversely affects proliferation the experiments. WB, TH, and BS performed data analysis, BV assessed and cell migration. Respir Res. 2015;16:153. histopathology. WB, TH and JWR wrote the paper. All authors read and 15. Blum W, Pecze L, Felley-Bosco E, Wu L, de Perrot M, Schwaller B. Stem cell approved the final manuscript. factors-based identification and functional properties of in vitro- selected subpopulations of malignant mesothelioma cells. Stem Cell Reports. 2017;8: Ethics approval and consent to participate 1005–17. Human embryonic lung tissue came from the specimen collection of the 16. Blum W, Pecze L, Felley-Bosco E, Worthmüller-Rodriguez J, Wu L, Vrugt B, et Unit of Anatomy, Section of Medicine, University of Fribourg, where donators al. Establishment of immortalized murine mesothelial cells and a novel had approved the use of tissue for teaching and research purposes. All mesothelioma cell line. In Vitro Cell Dev Biol Anim. 2015;1:714–21. experiments requiring mouse tissue (embryos) were performed with permission 17. Schwaller B, Buchwald P, Blümcke I, Celio MR, Hunziker W. Characterization of the local animal care committee (Canton of Fribourg, Switzerland) and of a polyclonal antiserum against the purified human recombinant calcium according to the present Swiss law and the European Communities Council binding protein calretinin. Cell Calcium. 1993;14:639–48. Directive of 24 November 1986 (86/609/EEC). 18. Blum W, Schwaller B. Calretinin is essential for mesothelioma cell growth/ survival in vitro: a potential new target for malignant mesothelioma therapy? Competing interests Int J Cancer. 2013;133:2077–88. The authors declare that they have no competing interests. 19. Frei C, Opitz I, Soltermann A, Fischer B, Moura U, Rehrauer H, et al. Pleural mesothelioma side populations have a precursor phenotype. Carcinogenesis. 2011;32:1324–32. Publisher’sNote 20. Gotzos V, Wintergerst ES, Musy JP, Spichtin HP, Genton CY. Selective Springer Nature remains neutral with regard to jurisdictional claims in distribution of calretinin in adenocarcinomas of the human colon and published maps and institutional affiliations. adjacent tissues. Am J Surg Pathol. 1999;23:701–11. 21. Lutz S, Perret-Gentil S, Pecze L, Henzi T, Dechézelles J-F, Schwaller B, et al. Author details Mouse mesothelium–derived cell lines: models to assess cytotoxic effects of Anatomy, Section of Medicine, University of Fribourg, Route Albert-Gockel 1, novel nanomaterials in vitro and to ultimately investigating carcinogenesis CH-1700 Fribourg, Switzerland. Institute of Pathology and Molecular in vivo. Toxicol Int. 2016;23:178–88. Pathology, University Hospital Zurich, Zurich, Switzerland. 2+ 22. Girard F, Venail J, Schwaller B, Celio MR. The EF-hand Ca -binding protein super-family: a genome-wide analysis of gene expression patterns in the Received: 24 January 2018 Accepted: 7 May 2018 adult mouse brain. Neuroscience. 2015;294:116–55. 23. Henzi T, Blum W-V, Pfefferli M, Kawecki TJ, Salicio V, Schwaller B. SV40- induced expression of Calretinin protects mesothelial cells from asbestos References cytotoxicity and may be a key factor contributing to mesothelioma pathogenesis. 1. Carbone M, Ly BH, Dodson RF, Pagano I, Morris PT, Dogan UA, et al. Am J Pathol. 2009;174:2324–36. Malignant mesothelioma: facts, myths, and hypotheses. J Cell Physiol. 24. Cordier Kellerman L, Valeyrie L, Fernandez N, Opolon P, Sabourin J-C, 2011;227:44–58. Maubec E, et al. Regression of AK7 malignant mesothelioma established 2. Gotzos V, Vogt P, Celio MR. The calcium binding protein calretinin is a in immunocompetent mice following intratumoral gene transfer of interferon selective marker for malignant pleural mesotheliomas of the epithelial gamma. Cancer Gene Ther. 2003;10:481–90. type. Pathol Res Pract. 1996;192:137–47. 25. Kresoja-Rakic J, Kapaklikaya E, Ziltener G, Dalcher D, Santoro R, Christensen 3. Husain AN, Colby T, Ordonez N, Krausz T, Attanoos R, Beasley MB, et al. BC, et al. Identification of cis- and trans-acting elements regulating calretinin Guidelines for pathologic diagnosis of malignant mesothelioma: 2012 expression in mesothelioma cells. Oncotarget. 2016;7:21272–86. update of the consensus statement from the international mesothelioma 26. Yokohira M, Hashimoto N, Yamakawa K, Suzuki S, Saoo K, Kuno T, et al. interest group. Arch Pathol Lab Med. 2013;137:647–67. Potassium octatitanate fibers (TISMO) induce pleural mesothelial cell reactions 4. Schwaller B. Calretinin: from a “simple” Ca2+ buffer to a multifunctional with iron accumulation in female a/J mice. Oncol Lett. 2010;1:589–94. protein implicated in many biological processes. Front Neuroanat. 2014;8:3. 27. Yokohira M, Hashimoto N, Yamakawa K, Saoo K, Kuno T, Imaida K. Lack of 5. Casjens S, Weber DG, Johnen G, Raiko I, Taeger D, Meinig C, et al. promoting effects from physical pulmonary collapse in a female a/J mouse Assessment of potential predictors of calretinin and mesothelin to improve lung tumor initiated with 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone the diagnostic performance to detect malignant mesothelioma: results from (NNK) with remarkable mesothelial cell reactions in the thoracic cavity by a population-based cohort study. BMJ Open. 2017;7:e017104. the polymer. Exp Toxicol Pathol. 2011;63:181–5. 6. Johnen G, Gawrych K, Raiko I, Casjens S, Pesch B, Weber DG, et al. Calretinin 28. Yokohira M, Nakano Y, Yamakawa K, Kishi S, Ninomiya F, Saoo K, et al. Strain as a blood-based biomarker for mesothelioma. BMC Cancer. 2017;17:386. differences in pleural mesothelial cell reactions induced by potassium 7. Chen Z, Gaudino G, Pass HI, Carbone M, Yang H. Diagnostic and prognostic octatitanate fibers (TISMO) infused directly into the thoracic cavity. Exp biomarkers for malignant mesothelioma: an update. Transl Lung Cancer Res. Toxicol Pathol. 2013;65:925–32. 2017;6:259–69. 29. Sneddon S, Patch A-M, Dick IM, Kazakoff S, Pearson JV, Waddell N, et al. 8. Jean D, Thomas E, Manié E, Renier A, de Reynies A, Lecomte C, et al. Syntenic Whole exome sequencing of an asbestos-induced wild-type murine model relationships between genomic profiles of Fiber-induced murine and human ofmalignant mesothelioma. BMC Cancer. 2017;17:396. malignant mesothelioma. Am J Pathol. 2011;178:881–94. 30. Kresoja-Rakic J, Sulemani M, Kirschner MB, Ronner M, Reid G, Kao S, et al. 9. Bianchi AB, Mitsunaga SI, Cheng JQ, Klein WM, Jhanwar SC, Seizinger B, et Posttranscriptional regulation controls Calretinin expression in malignant al. High frequency of inactivating mutations in the neurofibromatosis type 2 pleural mesothelioma. Front Genet. 2017;8:223. gene (NF2) in primary malignant mesotheliomas. Proc Natl Acad Sci U S A. 31. Carbone M, Kanodia S, Chao A, Miller A, Wali A, Weissman D, et al. Consensus 1995;92:10854–8. report of the 2015 Weinman international conference on mesothelioma. 10. Sekido Y, Pass HI, Bader S, Mew DJ, Christman MF, Gazdar AF, et al. J Thorac Oncol. 2016;11:1246–62. Neurofibromatosis type 2 (NF2) gene is somatically mutated in mesothelioma but not in lung cancer. Cancer Res. 1995;55:1227–31. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Biomarker Research Springer Journals

Absence of calretinin protein expression in malignant mesotheliomas from asbestos-exposed NF2+/− mice and mouse mesothelioma cell lines from various mouse strains

Free
7 pages
Loading next page...
 
/lp/springer_journal/absence-of-calretinin-protein-expression-in-malignant-mesotheliomas-VSE5ReJAaA
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Biomedicine, general; Cancer Research; Neurosciences
eISSN
2050-7771
D.O.I.
10.1186/s40364-018-0132-0
Publisher site
See Article on Publisher Site

Abstract

Background: Calretinin is the most widespread positive marker for the immunohistochemical identification of malignant mesothelioma (MM) and was proposed to serve as a blood-based biomarker. Functionally, evidence has accumulated that calretinin might be implicated in MM tumorigenesis. We aimed to identify calretinin (CR; Calb2)in +/− murine MM and reactive mesothelial cells in granuloma from asbestos-exposed NF2 mice, a line heterozygous for the tumor suppressor merlin (NF2), used as a mouse MM model. Additionally, we sought to ascertain the presence of calretinin in MM cell lines from other mouse strains. We also intended to investigate the role of calretinin in +/− +/- −/− mesotheliomagenesis by comparing the survival of asbestos-exposed NF2 and NF2 CR mice. +/− +/- −/− Methods: NF2 and NF2 CR mice, both lines on a C57Bl/6J background, were exposed to asbestos following an +/− established protocol. Tumor histology and asbestos-induced mortality were assessed. MM and granuloma from NF2 mice were analyzed with immunohistochemical methods for calretinin expression. Levels of Calb2 mRNA and calretinin expression in tumors and MM cell lines of various mouse strains were determined by RT-qPCR and Western blot analysis, respectively. +/− +/− Results: No expression of calretinin at the protein level was detected, neither in MM from NF2 mice, NF2 MM- derived cell lines nor immortalized mesothelial cells of mouse origin. At the mRNA level we detected Calb2 expression +/− +/- −/− in MM cell lines from different mouse strains. Survival of NF2 and NF2 CR mice exposed to asbestos showed no significant difference in a log-rank (Kaplan-Meier) comparison. Conclusions: The concomitant determination of calretinin and mesothelin blood levels has been proposed for early detection of human MM. Mouse MM models based on asbestos exposure are assumed to yield helpful information on the time course of appearance of mesothelin and calretinin in the blood of asbestos-treated mice determining the +/− earliest time point for interventions. However, the observed absence of calretinin in MM from NF2 mice and derived cell lines, as well as from MM cells from Balb/c and C3H mice likely precludes the use of calretinin as a biomarker for mouse MM. The results also indicate possible species differences with respect to an involvement of calretinin in the formation of MM. +/− Keywords: Malignant mesothelioma, Calretinin, Calb2, C57Bl/6J, NF2 mice, Asbestos * Correspondence: beat.schwaller@unifr.ch Anatomy, Section of Medicine, University of Fribourg, Route Albert-Gockel 1, CH-1700 Fribourg, Switzerland Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Blum et al. Biomarker Research (2018) 6:19 Page 2 of 7 Background intraperitoneally (i.p.) with 400 μg of crocidolite Malignant mesothelioma (MM) is an extremely aggres- suspended in 500 μL PBS every 3 weeks for a total of eight sive neoplasm linked to asbestos exposure [1]. Median rounds of injection (i.e., a total of 3.2 mg of crocidolite per survival is very short (9–12 months), likely linked to mouse as described before [12]). Unio Internationale diagnosis at late stages of the disease. Correct patho- Contra Cancrum (UICC) grade crocidolite asbestos was logical diagnosis is of outmost importance and calretinin obtained from SPI Supplies (West Chester, PA). For has been described as one of the most sensitive and se- survival analysis, mice were kept for a maximum of lective positive markers for human MM [2, 3]. Calretinin 33 months. (CR; human gene symbol: CALB2) is a calcium-binding Four NOD/SCID gamma mice were injected i.p. with high protein of the EF-hand family with several reported a suspension (200 μl PBS) of FACS-sorted ZL55-SO functions, see [4]. Together with mesothelin, the detec- cells (100,000/mouse). These genetically-modified hu- tion of CR in blood samples by ELISA has been sug- man MM cells are identified by their increased endogen- gested as a potential predictor to detect MM at earlier ous expression of the stem cell markers SOX2 and stages [5, 6]. Other discussed MM biomarkers include OCT4 driving the reporter eGFP, the latter allowing for soluble mesothelin-related peptides, megakaryocyte the isolation of this small (< 5%) subpopulation [15]. high potentiating factor, osteopontin, fibulin-3 and high ZL55-SO cells show an increased tumor-initiating mobility group protein B1 (HMGB1) [7]. However, capacity in vivo compared to the parental ZL55 cells. the time course of appearance of these biomarkers Mice were sacrificed after 5 weeks and tissue was fixed during human MM pathogenesis is currently un- in 4% paraformaldehyde. Histopathological diagnosis known. Animal models (mostly mice) are widely used and analyses were performed on paraffin-embedded as models to address such questions. samples as described before [16]. All animal experiments Common genomic alterations and similar genomic were performed with the permission of the local animal profiles, e.g. loss of tumor suppressors (NF2, LATS2, care committee (Canton of Fribourg, Switzerland) and BAP1) in murine and human MM emphasize the rele- according to the present Swiss law and the European vance of mouse model systems to study mesothelioma- Communities Council Directive 86/609/EEC. genesis [8] and to possibly validate suggested MM biomarkers. Such genetic mouse MM models include Western Blotting mice heterozygous for NF2, one of the most often al- Proteins from murine MM cell lines and from C57- +/− tered tumor suppressor genes with up to 44% alterations NF2 tumor samples were prepared as described +/ in human MM cell lines [9, 10]. Asbestos-exposed NF2 before [16]. Briefly, protein extracts from MM cell lines mice were reported to be an important murine model (AB12, AK7, RN5 [16], RN29) were separated by SDS- recapitulating many molecular features of human MM PAGE (10% gel) and transferred onto a nitrocellulose and to be relevant for further investigations on MM membrane. The membrane was incubated with primary pathogenesis mechanisms, as well as for preclinical antibody against calretinin (rabbit polyclonal CR7696 testing of putative novel therapeutic approaches [11, 12]. [17]), previously available from SWANT, Marly, The aim of this study was to investigate the expression Switzerland at a dilution of 1:10,000. Secondary biotinyl- +/− of calretinin in I) murine MM originating from NF2 ated anti-rabbit IgG at a dilution of 1:10,000 for 2 h at mice; II) MM cell lines derived from these mice and III) RT was used. For signal detection the ABC system murine MM cell lines from strains other than C57Bl/6J, (Vectastain, Vector Laboratories, Burlingame, CA) was +/− the genetic background of the NF2 mice. In addition, applied and chemiluminescent signals were revealed with to investigate any potential role for CR in murine MM the HRP substrate (Millipore, Luminata Forte); signals were formation we assessed whether overall survival differed detected with the Western blot reader (FluorChem E +/- between asbestos-exposed NF2 mice with or without a System, Bucher Biotec, Basel, Switzerland) as previ- functional Calb2 gene. ously described [18]. Methods Histological analysis Animal studies Tumor samples were fixed in 4% paraformaldehyde in PBS, 129S2/SvPas Nf2+/− mice (strain name: B6;129S2- pH 7.4, dehydrated and embedded in paraffin followed by tm1Tyj Nf2 /J; JAX stock #008190) were backcrossed for ≥6 sectioning with a microtome. Tumor samples were patho- generations to wild-type C57Bl/6J (C57) mice to obtain logically evaluated using hematoxylin/eosin or Goldner- +/− a C57-NF2 mouse line. These mice were crossed with stained sections. For immunohistochemistry, sections were −/− C57-CR mice [13] to give rise to the genotype C57- deparaffinized and incubated in antigen-retrieval solu- +/- −/− NF2 CR ; animals were genotyped as described tion using sodium citrate, pH 6, and processed as previously [14]. Mice, 6–8 weeks of age, were injected described before [19]. Primary antibodies (CR7696, Blum et al. Biomarker Research (2018) 6:19 Page 3 of 7 CR7697 or CR7699/4; SWANT, Marly, Switzerland) were CR staining in the developing epithelial cells of the lung. used at a 1:500 dilution with overnight incubation at 4 °C. A control section stained without primary CR antibody was negative (Fig. 1h) including the single layer of meso- Semi-quantitative RT-PCR of Calb2 mRNA levels thelial cells (inset; arrows). The expression of murine Calb2 mRNA was detected by In order to confirm the IHC results we performed semi-quantitative RT-PCR. Murine mesothelioma cell lines Western blot analysis to detect CR expression in murine were cultured under standard conditions [16]and total MM cell lines. Even when loading ≥50 μgofprotein no RNA extracted with PeqGold Trifast (Axonlab, Le Mont- specific CR signal at approximately 30 kDa was detected sur-Lausanne, Switzerland) following the manufacturer’s in all mouse cell lines (Fig. 2a) and in immortalized meso- protocol. cDNA was prepared from 500 ng total RNA using thelial cells from a C57 mouse (iMeso-WT1 [16, 21]; not the QuantiTect Reverse Transcription kit (Qiagen, Stockach, shown), while a signal was evident in extracts from the Germany) by following manufacturer’sinstructions. As human MM cell line MSTO-211H, as well as in mouse housekeeping gene the ribosomal protein 13S was used and cerebellum, described to be rich in CR [22](Fig. 2a). A Calb2 expression was assessed using the following primers: protein extract from isolated primary mesothelial cells of a −/− Calb2 -FW: 5′-GAA ATG GGT ACA TTG AAG GTA-3′, C57-CR mouse served as negative control. Calb2 was Calb2 RV 5’-CCA TCT GAG TTC TTC TTA TCA TAC detected at the mRNA level in all tested murine MM cell 5′,RPS-13FW5’CGA AAG CAT CTT GAG AGG AAC A lines and in mesothelial iMeso-WT1 cells (Fig. 2b), but 3′,RPS-13RV5’-TCG AGC CAA ACG GTG AAT C-3′. not in mouse primary mesothelial cells maintained in cul- ture for 24 h, the latter as reported before [14]. Statistical analysis In human MM, calretinin is considered to be the most GraphPad Prism version 7 and R statistical software sensitive and selective marker for the diagnosis, in par- were used to perform the log-rank test (Mantel-Cox ticular of the epithelioid and the epithelioid part of the test) on the Kaplan-Meier survival plot for comparison. biphasic MM type [3]. Furthermore, its essential func- tion in vitro in human MM cell lines [18], as well as its Results function in protection from asbestos-induced cytotox- +/- The repetitive injection of crocidolite in C57-NF2 icity has been demonstrated before [23]. While the clin- mice resulted in approximately 10% of confirmed ical relevance of calretinin IHC in the diagnosis of murine mesothelioma cases and in a multitude of human MM is undisputed, the putative role of calretinin granuloma formation in treated animals [12]. in the development of MM in vivo has remained elusive. Granuloma were described to be rich in reactive Thus, mouse survival after asbestos exposure was tested mesothelial cells and human reactive mesothelial cells in a well-studied MM mouse model, i.e. in mice hetero- were previously shown to be strongly positive for CR zygous for the tumor suppressor NF2 [11] either wild- +/− +/- −/− [20]. Human ZL55 MM cells injected i.p. into NOD/ type (C57-NF2 ) or null-mutant (C57-NF2 CR ) SCID gamma mice were used as a positive control. for the Calb2 gene. Long-term survival data were ZL55-derived tumors showed strong calretinin IHC analyzed by the standard Kaplan-Meyer method and staining, clearly a differentiating marker for the tumor differences between groups were assessed using a log- nodules located in connective tissue often consisting of rank test. Mice (48 in total) of the two genotypes were +/- +/+ +/- −/− CR-negative nodules of adipose tissue (Fig. 1a-c). None treated with asbestos (NF2 CR , n = 18; NF2 CR ; +/− of the C57-NF2 mice-derived MM cells and the react- n = 30). A total of 17 (94%) mice (2 with diagnosed MM) +/- +/+ ive mesothelial cells on the peritoneal side of the dia- died in the NF2 CR group (censored, n = 1) and 28 +/- −/− phragm from the same mice showed CR expression in (93%; 3 with diagnosed MM) died in the NF2 CR the IHC analysis, representative examples are shown in group (censored, n = 2) until the end of the long-term Fig. 1d-e. As previously described [14], the developing follow up in the 2 cohorts. Kaplan-Meier curves were lung of the murine embryo shows staining for CR in the similar in both groups. Median survival was 14. visceral mesothelial layer and in a fraction of mesenchy- 22 months (95% confidence interval: 9.23–19.07) in mice +/- +/+ mal cells (Fig. 1f), even though staining is weaker com- with a functional Calb2 gene (NF2 CR ) and 14. pared to human mesothelioma [14]. Also in the 63 months (95% confidence interval 13.4–17.3) in mice +/- −/− developing human lung, cells of the visceral pleura, i.e. with a non-functional Calb2 gene (NF2 CR ). A log- the outermost mesothelial cell layer, as well as a subpop- rank test indicated no significant difference in survival +/- +/+ ulation of stromal (mesenchymal) cells stained positive (p = 0.7345). The log-rank hazard ratio (NF2 CR for CR, in particular evident as brown-stained cytoplasm used as reference) was 0.90 (95% confidence interval 0. in mesothelial cells in Fig. 1g (inset; arrows). Few stro- 49–1.66). This indicates that mice with a functional mal cells in the mesenchyme showed very strong stain- Calb2 gene have a marginally higher, however ing (Fig. 1g, arrowheads). Note the absence of specific insignificant, risk of death after asbestos exposure. Blum et al. Biomarker Research (2018) 6:19 Page 4 of 7 high Fig. 1 Immunohistochemical analysis of human and mouse MM and embryonic mouse and human lung. ZL55-SO cells were injected i.p. into NOD/SCID gamma mice and tumor cells show strong expression of CR (b; higher magnification inset in c), arrow pointing towards a strong CR-positive tumor nodule. The negative control (parallel section) is shown in (a). d) Representative example of a mouse MM showing absence of CR expression, as well as in another example of an asbestos-induced lymphocytic and granulomatous inflammation (e) of the parietal pleura. f) In the developing lung of a mouse, CR protein is already expressed in the visceral pleura at day 10 (arrows) as reported before [14]. g) Expression of CR in the mesothelial cell layer (arrows) and in scattered stromal (mesenchymal) cells of the developing human lung (arrowheads). h) Negative control on parallel section incubated without primary CR antibody. The inset shows CR-negative mesothelial cells (arrows) at 2-fold higher magnification. The few round lightly brownish-stained cells are erythrocytes, where the endogenous peroxidase activity was not completely blocked. Abbreviations: A adipose tissue; T tumor; S stroma; G granulomatous inflammation; D pleura parietalis adjacent to the diaphragm; L lung. Scale bars: A and B: 2.5 mm, C: 500 μm. D-F: 100 μm; G,H: 50 μm; insets with layer of mesothelial cells: 2-fold magnification compared to G,H Discussion (e.g. ZL34, ONE58), as well as biphasic MM (e.g. Our results indicate that MM derived from asbestos- MSTO-211H, Mero83) show a clear and in few cases +/− exposed C57-NF2 mice do not express calretinin relatively strong CR expression evidenced by Western protein levels detectable by either IHC or Western blot blot analysis [18, 25]. Our negative CR IHC results analysis. No positive Western blot signal was observed are in contrast to previous reports showing positive in RN5 and RN29 MM cells derived from MM tissue of CR IHC staining in murine pleural MM samples. In +/− the same asbestos-exposed C57-NF2 mouse strain; toxicology studies, Yokohira et al. reported positive yet, RN5 and RN29 cells express Calb2 mRNA. CR immunostaining of the lung surface in female A/J Likewise, well-established murine MM cell lines AK7 mice exposed to 4-(methylnitrosamino)-1-(3-pyridyl)- [24] also from C57 mice, and AB12 cells from Balb/c 1-butanone (NNK) [26] and to potassium octatitanate mice were negative for CR protein expression and (TISMO) fibers [27]. In a later study, also mice from add- positive for Calb2 mRNA. Based on the previous obser- itional strains (C3H/HeN, ICR and C57BL/6), exposed to vation of a very strong positive correlation between CR TISMO showed a thickening of the visceral pleura due to protein and CALB2 mRNA levels in 13 human MM cell mesothelial cell infiltration with part of the cells showing lines, i.e. regulation mostly at the transcriptional level strong CR immunoreactivity (Fig. 3 in [28]). The differ- (see Fig. 1c in [25]), we assume levels of Calb2 mRNA ences between previous results and the current ones with to be low in mouse MM cell lines and MM tissue with respect to CR IHC might be several-fold. IHC results no detectable CR protein levels. However, species differ- might be related to differences in I) the materials used to ences in CR regulation may not be entirely excluded. induce changes in the tunica serosa (NNK and TIMSO fi- Thus, in contrast to human MM, where calretinin is bers vs. crocidolite), II) the site of injection (pleura vs. considered as a positive marker, we did not detect calre- peritoneum) and III) the CR antibodies used. With respect tinin at the protein level in MM and derived cell lines of to strain differences, Yokohira et al. also reported positive mouse origin. Even human sarcomatoid MM cell lines CR immunoreactivity in C57Bl/6 mice (Fig. 3 in [28]), Blum et al. Biomarker Research (2018) 6:19 Page 5 of 7 Fig. 2 Western Blot analysis for calretinin and RT-PCR for the detection of Calb2 mRNA. a) Western blot signals (upper panel) and the corresponding Ponceau Red-stained membrane (lower panel). As positive control, 10 ng of purified human recombinant CR was used. Sizes (molecular masses) of the marker proteins in the right lane are (from top to bottom): 130, 100, 75 (red), 63, 48, 35, 28 (green), and 17 kDa. The specific signal for CR runs just slightly above the green marker protein at approximately 30 kDa. In MSTO-211H cell extracts, a strong signal at the same position as purified CR indicates the presence of CR in these cells. All mouse samples from the MM cell lines RN29, RN5 and AK7 (all from C57Bl/6 J −/− mice), AB12 (Balb/c) are negative, as well as the negative control prMC from a CR mouse (prMC−/−). Mouse cerebellum extract (most right) shows a strong signal for CR. b) Positive signals for Calb2 mRNA were detected in RN29, RN5, AB12, AK7 and iMeso-WT cells; no signal was present in samples from freshly isolated C57 mouse prMC and in the negative control (H O). RPS13 was used as housekeeping gene while our tissue samples and cell lines (RN5, RN29, AK7) derived from C57Bl/6 mice were negative for CR. Of note, in the same study by Yokohira et al., weak CR positivity was also detectable in the healthy lung tissue, especially in C57Bl/6 J mice [28], while lung tissue was always com- pletely negative in our study, in line with many studies reporting that normal lung tissue is considered CR- negative. In addition, the specificity of the CR antibody to unequivocally detect mouse CR was not reported in the previous studies, while we have additionally confirmed the specificity of our CR antibodies by Western blotting. A role for calretinin in the development of human MM had been suggested based on in vitro studies. Induced expression of SV40 early gene products augments +/- +/+ Fig. 3 Kaplan-Meier survival analysis. C57-NF2 CR (blue line; n = 18) CR expression in the human mesothelial cell line MeT- +/- −/− and C57-NF2 CR (red line; n = 30) mice were injected with 8 × 5A, which leads to increased resistance to asbestos- 400 μg of crocidolite. The survival curves of the 2 groups (all mice) are mediated acute cytotoxicity, an effect mediated by the statistically not different (log-rank test, p =0.7345) AKT/PI3K pathway [23]. Inversely, down-regulation of Blum et al. Biomarker Research (2018) 6:19 Page 6 of 7 CR decreases the viability and proliferation of human MM selectively in human MM might point towards different cells in vitro, most evident in epithelioid and biphasic cell (possibly none) functions in the process of murine lines, but to a lesser extent also in MM cell lines with mesotheliomagenesis compared to the formation of MM prevalent sarcomatoid morphology [18]. Moreover, pri- in humans. Of note, the presumed role of CR in human −/− mary mesothelial cells from CR mice are characterized MM development is currently based on in vitro studies by a lower proliferation rate and are slower to close a with human MM cell lines. III) The low CR levels con- scratch evidenced in a 2D-wound assay compared to pri- ceivably present in mouse MM (below the detection mary mesothelial cells from wild-type mice [14]. This limit of IHC or Western blot analysis used in this study) −/− phenotype disappears, if CR is overexpressed in CR still might be sufficient to exert a role, e.g. as a signaling prMC. Based on these results we hypothesized to possibly molecule during mouse MM development. IV) Various observe a genotype-dependent effect in the survival of in Calb2 mRNA transcripts, in part derived from alterna- +/- −/− vivo asbestos-exposed C57-NF2 CR compared to tive splicing (so far detected only in human MM), might +/− mice with a functional Calb2 gene (C57-NF2 mice). have a function on their own as recently proposed [30] However, survival of mice in both groups was not and thus Calb2 transcripts might possibly be implicated different. Of note, in our survival analysis all mice were in murine MM tumorigenesis. All of the above points included, not only the ones with a diagnosed bona need to be considered, when using mouse MM models fide MM. Given the low penetrance in our model, as with the aim to possibly find strategies that might be +/− reported before (≈10% in C57-NF2 mice [12]), i.e. applied for developing novel human therapeutic ap- +/- −/− 3/30 in the C57-NF2 CR mouse group and 2/18 proaches. This is particularly the case when using mouse +/- +/+ in the C57-NF2 CR mouse group in this study, models aimed at exploiting the putative role of CR in the small number of confined MM cases doesn’t the pathogenesis of human MM. In such a case, xeno- allow to directly link Calb2 genotype to MM-caused graft models with human CR-positive MM cells might mortality. Thus, including all mice might have be considered as the current method of choice. With masked some subtle Calb2 genotype-dependent differ- respect to validating CR as a biomarker for early de- ences that are only observable in mice with a diag- tection of human MM (currently less than 5% are di- nosed MM. However, the deaths in both groups were agnosed at an early stage 1A [31]), one needs to be asbestos exposure-dependent, since unexposed mice aware that CR is unlikely of any predictive value in of both genotypes had a much higher median survival mouse MM models. Even if CR is expressed at very +/- +/+ (control C57-NF2 CR mice) of 29.13 months (95% low protein levels in mouse MM, CR plasma levels confidence interval: 5.80–32.93, log-rank test: p = 0.0008). are most probably below the detection limit of even This leads to several different hypotheses. I) The heavy most sensitive ELISA techniques. asbestos exposure by direct intraperitoneal fiber injec- tion resulting in relatively short latency for MM devel- Conclusions opment in mice (around 1 year in mice compared to Our results indicate the absence of CR (at least below 20–40 years in humans) leads to other routes of MM the threshold of Western blot detection) in mouse MM formation in human and mice. This is partially sup- cell lines, in line with negative IHC staining in asbestos- ported by the observation of mostly sarcomatoid and bi- induced mouse MM tissue. These findings essentially phasic MM in mice compared to human, where the rule out the use of CR as a putative biomarker for early largest number of MM cases is of the epithelioid type. detection of MM in mouse models, while this may well However, frequency of gene loci deletions containing be appropriate for human MM. Together with other di- genes likely involved in MM formation were reported to vergent results in human and mouse MM, e.g. with re- be similar in samples from human MM and asbestos- spect of MM histotype or commonly mutated genes, +/− exposed NF2 mice including CDKN2A, CDKN2B, it requires careful consideration, when translating find- CHD5, TP53, and NF2 [8]. In contrast, 14/15 cell lines ings obtained in one species to another. derived from asbestos-injected mice of several mouse Abbreviations strains showed homologous deletions of the CDKN2A C57: C57Bl/6J; CR: Calretinin; HMGB1: High mobility group B1 protein; locus, but no chromosomal losses in either the BAP1 MM: Malignant mesothelioma; NF2: Neurofibromatosis 2 and NF2 regions and moreover no mutations in these 2 Acknowledgements genes as well as in LATS2, commonly mutated in human We would like to thank L. Wu, University of Toronto for the help in the initial MM [29]; this might again hint towards species differ- collecting of the tumor samples. The excellent technical work of S. Eichenberger, A. Oberson and M. Sanchez, University of Fribourg is highly appreciated. ences in the pathways leading to MM II). Although CR expression is prevalent during both mouse [14] and hu- Funding man lung embryonic development (Fig. 1g, h), CR re- The project was supported by the Swiss National Science Foundation (SNSF appearance at levels detectable by Western blot analysis Sinergia grant CRSII3 #147697) and intramural funding from the University of Blum et al. Biomarker Research (2018) 6:19 Page 7 of 7 Fribourg. The funding body was not involved in the design of the study, the 11. Altomare DA, et al. A mouse model recapitulating molecular features of collection, analysis, and interpretation of data and in writing of the manuscript. human mesothelioma. Cancer Res. 2005;65:8090–5. 12. Rehrauer H, Wu L, Blum W, Pecze L, Henzi T, Serre-Beinier V, et al. How asbestos drives the tissue towards tumors: YAP activation, macrophage and Availability of data and materials mesothelial precursor recruitment, RNA editing, and somatic mutations. All data generated or analyzed during this study are included in this published Oncogene. 2018; doi.org/10.1038/s41388-018-0153-z article. A limited amount of paraffin sections from the used mouse tissue 13. Schurmans S, Schiffmann SN, Gurden H, Lemaire M, Lipp HP, Schwam V, et al. samples are available upon request. Impaired long-term potentiation induction in dentate gyrus of calretinin- deficient mice. Proc Natl Acad Sci U S A. 1997;94:10415–20. Authors’ contributions 14. Blum W, Pecze L, Felley-Bosco E, Schwaller B. Overexpression or absence of BS and WB conceived the study. WB, TH, HECS, LP, JWR and BV performed calretinin in mouse primary mesothelial cells inversely affects proliferation the experiments. WB, TH, and BS performed data analysis, BV assessed and cell migration. Respir Res. 2015;16:153. histopathology. WB, TH and JWR wrote the paper. All authors read and 15. Blum W, Pecze L, Felley-Bosco E, Wu L, de Perrot M, Schwaller B. Stem cell approved the final manuscript. factors-based identification and functional properties of in vitro- selected subpopulations of malignant mesothelioma cells. Stem Cell Reports. 2017;8: Ethics approval and consent to participate 1005–17. Human embryonic lung tissue came from the specimen collection of the 16. Blum W, Pecze L, Felley-Bosco E, Worthmüller-Rodriguez J, Wu L, Vrugt B, et Unit of Anatomy, Section of Medicine, University of Fribourg, where donators al. Establishment of immortalized murine mesothelial cells and a novel had approved the use of tissue for teaching and research purposes. All mesothelioma cell line. In Vitro Cell Dev Biol Anim. 2015;1:714–21. experiments requiring mouse tissue (embryos) were performed with permission 17. Schwaller B, Buchwald P, Blümcke I, Celio MR, Hunziker W. Characterization of the local animal care committee (Canton of Fribourg, Switzerland) and of a polyclonal antiserum against the purified human recombinant calcium according to the present Swiss law and the European Communities Council binding protein calretinin. Cell Calcium. 1993;14:639–48. Directive of 24 November 1986 (86/609/EEC). 18. Blum W, Schwaller B. Calretinin is essential for mesothelioma cell growth/ survival in vitro: a potential new target for malignant mesothelioma therapy? Competing interests Int J Cancer. 2013;133:2077–88. The authors declare that they have no competing interests. 19. Frei C, Opitz I, Soltermann A, Fischer B, Moura U, Rehrauer H, et al. Pleural mesothelioma side populations have a precursor phenotype. Carcinogenesis. 2011;32:1324–32. Publisher’sNote 20. Gotzos V, Wintergerst ES, Musy JP, Spichtin HP, Genton CY. Selective Springer Nature remains neutral with regard to jurisdictional claims in distribution of calretinin in adenocarcinomas of the human colon and published maps and institutional affiliations. adjacent tissues. Am J Surg Pathol. 1999;23:701–11. 21. Lutz S, Perret-Gentil S, Pecze L, Henzi T, Dechézelles J-F, Schwaller B, et al. Author details Mouse mesothelium–derived cell lines: models to assess cytotoxic effects of Anatomy, Section of Medicine, University of Fribourg, Route Albert-Gockel 1, novel nanomaterials in vitro and to ultimately investigating carcinogenesis CH-1700 Fribourg, Switzerland. Institute of Pathology and Molecular in vivo. Toxicol Int. 2016;23:178–88. Pathology, University Hospital Zurich, Zurich, Switzerland. 2+ 22. Girard F, Venail J, Schwaller B, Celio MR. The EF-hand Ca -binding protein super-family: a genome-wide analysis of gene expression patterns in the Received: 24 January 2018 Accepted: 7 May 2018 adult mouse brain. Neuroscience. 2015;294:116–55. 23. Henzi T, Blum W-V, Pfefferli M, Kawecki TJ, Salicio V, Schwaller B. SV40- induced expression of Calretinin protects mesothelial cells from asbestos References cytotoxicity and may be a key factor contributing to mesothelioma pathogenesis. 1. Carbone M, Ly BH, Dodson RF, Pagano I, Morris PT, Dogan UA, et al. Am J Pathol. 2009;174:2324–36. Malignant mesothelioma: facts, myths, and hypotheses. J Cell Physiol. 24. Cordier Kellerman L, Valeyrie L, Fernandez N, Opolon P, Sabourin J-C, 2011;227:44–58. Maubec E, et al. Regression of AK7 malignant mesothelioma established 2. Gotzos V, Vogt P, Celio MR. The calcium binding protein calretinin is a in immunocompetent mice following intratumoral gene transfer of interferon selective marker for malignant pleural mesotheliomas of the epithelial gamma. Cancer Gene Ther. 2003;10:481–90. type. Pathol Res Pract. 1996;192:137–47. 25. Kresoja-Rakic J, Kapaklikaya E, Ziltener G, Dalcher D, Santoro R, Christensen 3. Husain AN, Colby T, Ordonez N, Krausz T, Attanoos R, Beasley MB, et al. BC, et al. Identification of cis- and trans-acting elements regulating calretinin Guidelines for pathologic diagnosis of malignant mesothelioma: 2012 expression in mesothelioma cells. Oncotarget. 2016;7:21272–86. update of the consensus statement from the international mesothelioma 26. Yokohira M, Hashimoto N, Yamakawa K, Suzuki S, Saoo K, Kuno T, et al. interest group. Arch Pathol Lab Med. 2013;137:647–67. Potassium octatitanate fibers (TISMO) induce pleural mesothelial cell reactions 4. Schwaller B. Calretinin: from a “simple” Ca2+ buffer to a multifunctional with iron accumulation in female a/J mice. Oncol Lett. 2010;1:589–94. protein implicated in many biological processes. Front Neuroanat. 2014;8:3. 27. Yokohira M, Hashimoto N, Yamakawa K, Saoo K, Kuno T, Imaida K. Lack of 5. Casjens S, Weber DG, Johnen G, Raiko I, Taeger D, Meinig C, et al. promoting effects from physical pulmonary collapse in a female a/J mouse Assessment of potential predictors of calretinin and mesothelin to improve lung tumor initiated with 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone the diagnostic performance to detect malignant mesothelioma: results from (NNK) with remarkable mesothelial cell reactions in the thoracic cavity by a population-based cohort study. BMJ Open. 2017;7:e017104. the polymer. Exp Toxicol Pathol. 2011;63:181–5. 6. Johnen G, Gawrych K, Raiko I, Casjens S, Pesch B, Weber DG, et al. Calretinin 28. Yokohira M, Nakano Y, Yamakawa K, Kishi S, Ninomiya F, Saoo K, et al. Strain as a blood-based biomarker for mesothelioma. BMC Cancer. 2017;17:386. differences in pleural mesothelial cell reactions induced by potassium 7. Chen Z, Gaudino G, Pass HI, Carbone M, Yang H. Diagnostic and prognostic octatitanate fibers (TISMO) infused directly into the thoracic cavity. Exp biomarkers for malignant mesothelioma: an update. Transl Lung Cancer Res. Toxicol Pathol. 2013;65:925–32. 2017;6:259–69. 29. Sneddon S, Patch A-M, Dick IM, Kazakoff S, Pearson JV, Waddell N, et al. 8. Jean D, Thomas E, Manié E, Renier A, de Reynies A, Lecomte C, et al. Syntenic Whole exome sequencing of an asbestos-induced wild-type murine model relationships between genomic profiles of Fiber-induced murine and human ofmalignant mesothelioma. BMC Cancer. 2017;17:396. malignant mesothelioma. Am J Pathol. 2011;178:881–94. 30. Kresoja-Rakic J, Sulemani M, Kirschner MB, Ronner M, Reid G, Kao S, et al. 9. Bianchi AB, Mitsunaga SI, Cheng JQ, Klein WM, Jhanwar SC, Seizinger B, et Posttranscriptional regulation controls Calretinin expression in malignant al. High frequency of inactivating mutations in the neurofibromatosis type 2 pleural mesothelioma. Front Genet. 2017;8:223. gene (NF2) in primary malignant mesotheliomas. Proc Natl Acad Sci U S A. 31. Carbone M, Kanodia S, Chao A, Miller A, Wali A, Weissman D, et al. Consensus 1995;92:10854–8. report of the 2015 Weinman international conference on mesothelioma. 10. Sekido Y, Pass HI, Bader S, Mew DJ, Christman MF, Gazdar AF, et al. J Thorac Oncol. 2016;11:1246–62. Neurofibromatosis type 2 (NF2) gene is somatically mutated in mesothelioma but not in lung cancer. Cancer Res. 1995;55:1227–31.

Journal

Biomarker ResearchSpringer Journals

Published: Jun 6, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off