- Email: [email protected]
Stem cell marker expression in small cell lung carcinoma and developing lung tissue
Human Pathology (2008) 39, 1597–1605
www.elsevier.com/locate/humpath
Original contribution
Stem cell marker expression in small cell lung carcinoma and developing lung tissue Lin-Kristin Koch a , Hui Zhou MD a , Jörg Ellinger MD b , Katharina Biermann MD a , Tobias Höller c , Alexander von Rücker MD a , Reinhard Büttner MD a , Ines Gütgemann MD a,⁎ a
Institut of Pathology, University of Bonn, D-53127 Bonn, Germany Department of Urology, University of Bonn, D-53127 Bonn, Germany c Institute for Biometry, Informatics and Epidemiology, University of Bonn, D-53127 Bonn, Germany b
Received 19 November 2007; revised 1 February 2008; accepted 1 March 2008
Keywords: Stem cells; Small cell lung cancer; Podocalyxin like protein-1; Polycomb group protein; Bmi-1; Immunhistochemistry
Summary Histopathologic and clinical findings suggest that small cell lung cancer is derived from a multipotent proximal airway epithelial cell. In order to investigate the histogenetic origin of small cell lung cancer, we compared stem cell marker expression in human fetal lung tissue, human adult bronchial tissue, and a cohort of 64 small cell lung cancers. Supporting derivation of a multipotent precursor cell, 87.5% (56/64) of small cell lung cancers showed a dot-like expression of podocalyxinlike protein 1 (PODXL-1), a marker of embryonic and hematopoetic stem cells. Of small cell lung cancers, 98.4% (63/64) ubiquitously expressed Bmi-1, a key player in self-renewal of stem cells. Oct4 and AP2γ were not expressed. Although podocalyxin-like protein 1 did not correlate with p53 or Wilms tumor suppressor 1, known regulators of podocalyxin-like protein 1, we could demonstrate demethylated CpG islands in the podocalyxin-like protein 1 promoter in small cell lung cancer, indicating epigenetic regulation. During fetal lung development and within adult bronchial mucosa, Bmi-1 was expressed ubiquitously. In contrast, podocalyxin-like protein 1 was detected in few stromal cells during the pseudoglandular phase (n = 7) and, importantly, in clustered epithelial cells within proximal bronchi and the trachea during the canalicular phase (n = 10). Interestingly, podocalyxin-like protein 1 was not expressed in normal or metaplastic adult bronchial epithelium (n = 36) but was expressed in sparse epithelial cells in half of the cases of normal tumor adjacent bronchial mucosa (20/40). Taken together, we show that small cell lung cancers and clustered epithelial cells in developing proximal bronchi share the expression of stem cell markers, suggesting a possible histogenetic link. © 2008 Elsevier Inc. All rights reserved.
1. Introduction
⁎ Corresponding author. Tel.: +49 228 287 16968; fax: +49 228 287 15030. E-mail address: [email protected] (I. Gütgemann). 0046-8177/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2008.03.008
Small cell lung cancer (SCLC) is a high-grade malignant epithelial tumor typically arising in a proximal and peribronchial location. Morphologically, SCLC resembles stem cells as tumor cells are small and round to fusiform, have a high nuclear to cytoplasmatic ratio, finely
1598 granular chromatin, absent or inconspicuous nucleoli, and frequent mitoses [1]. The subtype of combined small cell carcinoma demonstrates multipotent differentiation of tumor cells, as combined small cell carcinomas can consist of areas of pure SCLC with components of squamous cell carcinoma, adenocarcinoma, neuroendocrine carcinoma, spindle cell carcinoma, or giant cell carcinoma. Often, SCLC is unresectable because of its discohesive growth pattern within the lung and early systemic spread. Some patients with SCLC respond favorably to first-line, multimodal chemotherapy but often relapse within a short time [2]. Histopathologic and clinical findings suggest that SCLC is derived from cells that bear characteristics of multipotent proximal airway epithelial cells [3]. In this study, we further explored this concept by investigating stem cell marker expression in SCLC as well as in adult and fetal respiratory epithelium. As identifying markers of human proximal airway stem cells are currently unknown, we chose podocalyxin-like protein 1 (PODXL-1), a marker of hematopoetic and embryonic stem cells [4,5]; Bmi-1, a key regulator in stem cell homeostasis [6]; a putative embryonic stem cell marker, AP2γ [7]; and a well-characterized marker of pluripotency, Oct4 [8] for our study. Furthermore, we analyzed the methylation status of the PODXL-1 promoter in
L. -K. Koch et al. SCLC as well as expression of Wilms tumor suppressor 1, a positive transcriptional regulator and p53, a negative regulator of PODXL-1 expression [9].
2. Methods 2.1. Case selection Formalin-fixed, paraffin-embedded bronchial biopsies (n = 94), including SCLC (n = 64), normal bronchial epithelium located adjacent to tumor (n = 40), normal respiratory epithelium from lung biopsies without any metaplastic disease (n = 27), and bronchial epithelium with squamous cell metaplasia (n = 3), were retrieved from the archives of the Institute of Pathology, University of Bonn (Bonn, Germany). Furthermore, proximal bronchi from pulmonary lobectomy specimens (n = 6) with normal respiratory epithelium were assessed as well as autopsy material of normal fetal lungs (n = 21). Standard paraffin tissue sections were stained with hematoxylin-eosin (H&E) for conventional histology. The diagnosis of SCLC was verified by pankeratin, TTF-1, and CD56 immunhistochemical staining.
Fig. 1 Immunhistochemical analysis in SCLC showing dot-like (A) or membranous (A, inset) staining with anti-PODXL-1 and strong nuclear Bmi-1 expression in tumor cells (B). Scattered, sparse PODXL-1 positive cells are found in the tumor adjacent bronchial epithelium (C), whereas Bmi-1 staining is ubiquitously expressed in most epithelial cells (D) (scale bar = 50 μm).
Stem cell markers in SCLC
1599
2.2. Immunohistochemistry
2.4. DNA isolation, bisulphite modification, and methylation-specific polymerase chain reaction
Sections (2-3 μm) were deparaffinized by routine techniques, placed in 200 ml of target retrieval solution (citrate buffer of pH 6.0) and heated for 30 minutes at boiling temperature (microwave/600 W). After cooling for 20 minutes, sections were washed with Tris-buffered saline. Immunohistochemical staining was performed by an automated immunohistochemical staining stainer (“Tech Mate 500” Autostainer, DAKO Cytomation, Glostrup, Denmark) using the streptavidine-biotin-peroxidase complex technique. Endogenous peroxidase activity was blocked by treatment with 3% H2O2 for 5 minutes. Slides were developed with EnVisionTM (DAKO). Primary antibodies used were PODXL-1 (R&D Systems, Wiesbaden-Nordenstadt, Germany; clone: AF 1658, 1:50, overnight), Bmi-1 (Chemicon, Hofheim, Germany; clone: F6, 1:50, overnight), Oct4 (Santa Cruz, Biotechnology, Santa Cruz, CA; clone: C-10, 1:50, overnight), AP2γ (Santa Cruz, clone: H-77, 1:500, overnight), p53 (DAKO, Hamburg, Germany; clone: DO-7, 1:100, 30 minutes), Wilms tumor suppressor 1 (WT1) (DCS, clone: WLM04, 1:400, overnight), CD31 (DAKO, clone: JC/70A, 1:100), and p63 (DAKO, clone: 4A4, 1:100, overnight). Endothelial cells served as internal staining controls for PODXL-1 and normal respiratory epithelium for Bmi-1.
2.3. Scoring Standard sections were evaluated for morphology and immunostaining by two independent investigators. PODXL-1 expression was found in a characteristic dot like or membranous pattern in SCLCs (Fig. 1A). Immunolabeling in less than 5% of tumor cells was scored as negative, and staining in greater than 5% of tumor cells, as positive, as described before [10]. All tumor cells within SCLC biopsies were scored. As in SCLCs, different staining intensities of PODXL-1 were found; strong staining in more than 5% of tumor cells was distinguished from weak staining. Nuclear Bmi-1 staining in SCLCs was scored in a 3-tiered manner as described above for PODXL-1. Strong nuclear p53 immunolabeling in more than 10% of tumor cell nuclei and nuclear p63 staining were interpreted as positive [11,12]. In addition, frequency of p53 immunoreactivity was assessed in SCLCs by subdividing the cases into three groups: nonimmunoreactive and immunoreactive in 50% or less or more than 50% of nuclei [13].
Table 1
Positive Negative
Four 5-μm slices of formalin-fixed, paraffin-embedded tissue biopsies were deparaffinized using xylol and ethanol. The Qiagen Robot M48 (Hilden, Germany) was used to isolate genomic DNA using the MagAttract DNA Mini Kit (Quiagen; final elution volume: 100 μL). As positive and negative controls for polymerase chain reaction (PCR) analysis, testes embryonal carcinoma (Tera-1) cell line– derived DNA and healthy volunteers' white blood cell (WBC) DNA was isolated. Universal methylated DNA was prepared by treating of WBC-DNA with SssI CpGMethylase (New England Biolabs, Frankfurt, Germany). We performed methylation (MSP)/unmethylation-specific (USP) PCR of the promoter region of PODXL-1 (MSP: forward 5′-TTT TTT AAG GCG CGG AGG TC-3′ and reverse 5′-CGC TAA ACC GTA AAC AAT AAA CA-3′; USP: forward 5′-TTA AGG TGT GGA GGT TGT TGG-3′ and reverse 5′- CCA AAA CCA AAA CTA AAC AAA CAC-3′). These primer sets targeted the CpGs (CpG no. −23 to −27 relative to transcription start site) previously shown to be differentially methylated in Tera-1 (unmethylated, PODXL-1 expression) and Meg01 cells (human megakaryoblastic cells; hypermethylated and PODXL-1 not expressed) [14]. All PCR experiments were carried out in a volume of 10 μL with 384-well plates and an ABIPrism 7900HT (Perkin Elmer, Foster City, CA). Samples (1.5-μL bisulphite-treated DNA) were analyzed in triplicate containing 5-μL Power SYBR Green Master Mix (Applied Biosystems) and 200 nmol/L of each forward and reverse primer. PCR amplification was performed as follows: 95°C for 10 minutes, followed by 40 cycles at 95°C for 15 seconds and 59°C for 60 seconds. Following the last cycle, a dissociation curve analysis was performed.
2.5. Statistical analysis Comparative analysis of ordinal data, such as p53 expression in SCLC compared with either PODXL-1 or Bmi-1 (subdivided by intensity in “negative,” “weak,” and “strong”), was performed using the Cochran-Armitage test. For dichotomous data such as sex and age groups (b63 years versus ≥63 years) the Fisher exact test was used. Statistical significance was determined when the 2-sided exact P value of a test was less than .05. All analyses were performed using SPSS version 14.0.2 (SPSS, Chicago, IL).
Immunhistochemical expression of stem cell marker PODXL-1, Bmi-1, and p53 in SCLC (n = 64) Intensity
PODXL-1–reactive
Bmi-1–reactive
p53-reactive
Strong Weak Negative
73.4% (47/64) 14.1% (9/64) 12.5% (8/64)
92.2% (59/64) 6.3% (4/64) 1.5% (1/64)
65.6% (42/64) 34.4% (22/64)
1600
3. Results 3.1. Stem cell marker and p53 expression in SCLC Overall, 87.5% (56/64) of SCLCs showed coexpression of PODXL-1 and Bmi-1 (Fig. 1). Of the PODXL-1 immunoreactive SCLCs, 73.4% (47/64) could be subdivided in a strong 14.1% (9/64) and a weak dot-like (Fig. 1A) or membranous staining pattern (Fig. 1A insert). Of SCLCs, 12.5% (8/64) were PODXL-1 negative; 92.2% (59/64) cases of SCLC strongly expressed Bmi-1 in all examined tumor nuclei, 6.3% (4/64) stained weakly, and 1.5% (1/64) were negative for Bmi-1 (Table 1). In contrast, putative embryonic stem cell markers AP2γ (0/37), Oct4 (0/37), and WT1 (0/37) were not expressed in SCLCs. Likewise, SCLCs were negative for p63 and CD31 (n = 11, data not shown). Nuclear p53 expression was observed in 65.6% (42/64) of SCLCs and no p53 immunostaining was observed in 34.4% (22/64) of cases. No significant correlation was seen when comparing p53 expression in SCLC with either PODXL-1 or Bmi-1 (P = 1.000 and P = .344, respectively). When p53 frequencies
L. -K. Koch et al. were taken into account by subdividing SCLCs into groups with 50% or less or more than 50% of nuclear p53 reactivity, no significant correlation with PODXL-1 (P = .877) nor Bmi-1 immunostaining was observed (P = .721). No correlation was found when comparing age (b63 years: n = 24, ≥63 years: n = 40) and sex (male: n = 43, female: n = 21) of patients suffering from SCLC with PODXL-1 and Bmi-1 expression in tumor sections.
3.2. PODXL-1 promoter methylation of CpG islands in SCLC Next, we assessed the methylation status of CpG-rich islands within the PODXL-1 promoter, which has been shown to correlate with transcriptional activity [14]. In 8 of 9 SCLCs, we detected unmethylated PODXL-1 products with the USP primer set (peak in dissociation curve at ∼72°C). Here, no signal was observed using the MSP primer set (Fig. 2). The Tera-1 cell line DNA was unmethylated at the investigated CpG sites (Fig. 2), as has been described [14]. Universal methylated WBC-DNA demonstrated a peak 75.7°C using the MSP primers.
Fig. 2 Methylation status of CpG island of the PODXL-1 promoter in SCLC and control cell lines and peripheral blood leucocytes (WBC DNA), identified by dissociation curve analysis using unmethylation specific (USP) (left) and methylation specific primer sets (MSP) (right). A peak at 72.8°C using the USP-primer set indicates unmethylated DNA (Tera-1 cell line, SCLC). A peak using MSP primers at 75.7°C is seen with universal methylated control DNA prepared by treating WBC-DNA with SssI CpG-Methylase (Sssl). Eight of nine SCLC samples showed similar peaks at 72.8°C (SCLC) indicating unmethylated CpG islands. During this procedure, the double-stranded PCR product is denatured according to its unique melting Temperature. A peak at 75.7°C indicated a methylated DNA in the MSP, and a product at 72.8°C indicated unmethylated DNA in the USP. Abbreviations: NTC, nontissue control.
Stem cell markers in SCLC
1601
Fig. 3 Immunhistochemical analysis of fetal lung tissue. Pseudoglandular phase: H&E (A), PODXL-1 dot-like expression in very few stromal cells (B), strong ubiquitous Bmi-1 staining (C), and p63 expression limited to the proximal bronchial tree (D). Canalicular phase: H&E (E), clustered epithelial cells show PODXL-1 expression in particular in deep pockets of proximal bronchial and tracheal epithelium (F), ubiquitous nuclear Bmi-1 expression in epithelial and stromal cells (G), and p63 staining in basal cells (H) (scale bar = 25 μm).
1602 Table 2
L. -K. Koch et al. Immunhistochemical expression of PODXL-1, Bmi-1, and p63 in fetal (n = 21) and normal adult bronchial tissue (n = 33)
Phase of lung development
PODXL-1
Bmi-1
p63
Pseudoglandular phase (5-17 wk) Canalicular phase (18-26 wk)
1% of stromal cells, dot like
Ubiquitous in stromal and epithelial cells Ubiquitous in stromal and epithelial cells
Basal cells, proximal bronchial tree Basal cells, proximal bronchial tree
Saccular phase (26-38 wk)
Normal bronchial tissue
Clustered epithelial cells, especially deep pockets, proximal bronchi and trachea, dot like Clustered epithelial cells, especially deep pockets, proximal bronchi and trachea, dot-like No expression in normal bronchial epithelium
3.3. Stem cell marker expression in fetal lung Analysis of Bmi-1 and PODXL-1 expression in fetal lung tissue (n = 21) was subdivided into three periods of lung formation: pseudoglandular phase (5-17 weeks) (n = 7), canalicular phase (18-26 weeks) (n = 10), and saccular phase (26-38 weeks) (n = 3). During the pseudoglandular phase, there were no morphologically detectable clustered epithelial cells. Few stromal cells in fetal lung tissue during this phase showed a weak dot-like PODXL-1 expression (Fig. 3B). However, CD31 staining was observed in single positive cells and stromal capillary vessels (data not shown) indicating that at least some of the PODXL-1 positive cells may correspond to sprouting endothelial cells. In contrast, we observed PODXL-1 staining during the canalicular phase in sparse clustered epithelial cells in major bronchi (Fig. 3F). Here, PODXL-1–positive cells were frequently observed in deep pockets of major bronchi or lobar bronchi in between cartilaginous nests. p63 highlighted basal cells underlining respiratory epithelial cells within major bronchi or lobar bronchi. These p63-positive cells were located lower within the epithelium than PODXL-1–positive cells and were found at high abundance within the trachea and major bronchi (Fig. 3D and H, Table 2). Beginning with the saccular phase, the frequency of PODXL-1 stained cells decreased until birth, where only very few epithelial cells showed PODXL-1 expression (data not shown). Bmi-1 was expressed in stromal as well as epithelial cells from the fifth week until maturity (n = 4, Fig. 3C and G, Table 2).
3.4. Stem cell marker expression in normal bronchial, reactive respiratory epithelium in adult lung, and tumor-adjacent normal bronchial epithelium Bronchial biopsies showing normal respiratory epithelium (n = 27), metaplastic lesions (squamous cell metaplasia
Ubiquitous in stromal and epithelial cells Normal bronchial epithelial cells and lymphocytes
Basal cells, proximal bronchial tree
n = 3), and proximal bronchi from pulmonary lobectomy specimens (n = 6) with normal respiratory epithelium showed no PODXL-1 expression (Fig. 4B and F). In contrast, half of the cases of normal bronchial epithelium located adjacent to tumor (20/40) showed focal PODXL-1 expression (Fig. 1C). Normal respiratory epithelial cells of major bronchi, metaplastic epithelium, and bronchial epithelium adjacent to tumor cells demonstrated a ubiquitous nuclear Bmi-1 staining pattern within most of the nuclei. p63 highlighted basal cells within the respiratory epithelium above the basement membrane in adult bronchi and was also detected in squamous metaplasia (Fig. 4D and H).
4. Discussion Here, we show that PODXL-1 and Bmi-1, but not AP2γ, Oct4, and WT1, are expressed in SCLC (Fig. 1). The expression of stem cell markers such as PODXL-1 and Bmi1 in tumor cells underscores the possible link between cancer stem cells and normal stem cells [15]. This hypothesis is supported by tumor cell transplantation studies using CD133 as a marker of cancer stem cells in lung tumors and observations that CD133 expressing cells are increased in regenerating lung tissue [16]. PODXL-1 is a heavily glycosylated sialomucin expressed in hematopoietic, embryonic, and mesenchymal stem cells [4,5,17]. Like other stem cell markers, it has also been described in human malignancies, in particular, in pancreatic [10] and breast carcinoma [18] as well as in leukaemias [19]. Inappropriate expression or function of PODXL-1 may be caused by hypermethylation of the PODXL-1 promoter [14]; gene amplification, which has been proposed for prostate cancer [20]; or inappropriate expression of PODXL-1 due to changes in the WT1 and/or p53 activity [9]. In comparing the expression of PODXL-1 and p53, no significant correlation was found indicating that protein levels of PODXL-1 in
Fig. 4 Immunhistochemical staining of normal adult bronchial epithelium (A-D), and squamous metaplasia (E-H): H&E (A and E), lack of PODXL-1 expression (B and F), ubiquitous Bmi-1 expression within the majority of epithelial nuclei (C and G). p63 highlights cells of the respiratory epithelium above the basement membrane within the lower level of bronchial epithelium (D and H) (scale bar = 50 μm).
Stem cell markers in SCLC
1603
1604 SCLC may be influenced by other factors. Instead, we observed unmethylated DNA in CpG-rich islands of the promoter region, suggesting that DNA demethylation may be involved in PODXL-1 regulation in SCLC (Fig. 2). Once airway epithelial stem cells have been identified unambiguously, it will be interesting to compare the epigenetic regulation of the PODXL-1 gene in these cells and SCLCs. The cell of origin of SCLC is still enigmatic. SCLCs, because of the expression of neuroendocrine markers such as chromogranin, neuron-specific enolase [21], and calcitonin gene–related peptide have been proposed to be derived from neuroendocrine cells (NECs) [22]. On the other hand, SCLC also shares characteristics of proximal bronchial precursor cells that give rise to respiratory epithelial cells during development and repair and, like these cells, depend on the ligand-dependent Sonic Hedgehog (Shh) activation pathway [23]. Although it is unclear, at the present time, whether these cells truly fulfill stem cell characteristics including long-term cell division, Shh-dependent epithelial airway regeneration occurs from these cells, followed by neuroendocrine differentiation [23]. In order to identify a corresponding counterpart of SCLC, we investigated PODXL-1 expression in developing lung and bronchial tissue, where these cells might be more frequent than in adult bronchiopulmonary biopsies. There are many different cell types described within the developing respiratory epithelium, including ciliated and nonciliated (Clara) cells, basal cells, and NECs. The latter can not only be found as scattered single cells within the proximal bronchial tree but also in groups related with nerve fibers called neuroepithelial bodies [24,25]. Within the unipotent precursor population of NECs within neuroepithelial bodies, variant Clara cells with multipotent stem cell features have been identified in mice [25]. Here, we find that PODXL-1 is expressed in a small subset of cuboidal epithelial cells within proximal airways (Fig. 3F) during the canalicular phase (1826 weeks) of human lung development. Whether these cells are similar to variant Clara cells in adult bronchi cannot be investigated at the present time, as defining human markers are lacking. Colocalization and functional studies are needed to address this possibility. During the pseudoglandular phase of lung morphogenesis (5-17 weeks), PODXL-1 was expressed in developing vessels and within single stromal cells. Here, because of the high abundance of sprouting endothelial cells during the pseudoglandular phase highlighted by CD31 staining, single PODXL-1–positive cells may correspond to sprouting endothelial cells rather than precursor cells. Interestingly, however, the distribution of PODXL-1–positive cells in human fetal lung is similar to Shh-activated cells in developing murine lung: Shh-activated single cells are observed within the stroma during the pseudoglandular phase. During the canalicular phase, clustered epithelial calcitonin gene–related peptide and patched double-positive cells are found within proximal bronchi next to Shh expressing epithelial cells [23].
L. -K. Koch et al. Basal cells have been described in mice as multipotent reserve cells, which can differentiate into each of the mature cell types of the bronchial epithelium [26]. It is unclear whether, in our study, some p63 positive cells in fetal lungs show coexpression of PODXL-1 expression. However, PODXL-1 staining cells were found predominantly in deep folds of bronchial epithelium, whereas p63 stained a cell layer underlying the cuboidal epithelial cells of major bronchi during the canalicular phase (Fig. 3H). This observation is in line with previous evidence, supporting that squamous carcinoma but not SCLC is derived from basal cells [3]. Bmi-1 expression was observed in 98.4% (63/64) cases of SCLC in all examined tumor nuclei. Bmi-1 is a member of the Polycomb group family of proteins, which represent epigenetic chromatin modifiers [27,28]. It is described in numerous other human cancers [29]. Dysregulation of normal self-renewal pathways such as the Shh pathway and the products of Bmi-1 and other Polycomb Group genes may explain tumorigenesis as because of dysregulated stem cell proliferation [6]. Our study shows that Bmi-1 in the lung is not a phenotypic marker for stem cells but, instead, stains self-renewing cells in normal bronchial epithelium, proliferating cells in SCLC, and proliferating cells during fetal lung development. Interestingly, Bmi-1 is a downstream effector of the extracellular signaling molecule Shh. As mentioned before, this pathway plays a pivotal role during normal lung development, airway repair, and in SCLC proliferation [23]. Therefore, it is likely, that Bmi-1 overexpression in SCLC is a result of juxtacrine Shh signaling. In summary, the present study is the first to demonstrate differential expression of PODXL-1 and Bmi-1 in SCLC, human fetal lung tissue, and adult bronchial epithelium. The localization of PODXL-1 positive cells during lung morphogenesis underscores the hypothesis that there is a possible histogenetic link between SCLC and multipotent airway epithelial cells in the proximal bronchial tree that have retained expression of stem cell markers. In the future, it will be interesting to examine whether PODXL-1–expressing cells during the canalicular phase of human lung development bear functional stem cell characteristics and how epigenetic mechanisms govern SCLC tumorigenesis.
Acknowledgments We thank S Reynolds and H Schorle for helpful discussions and C Esch for technical assistance.
References [1] Hirsch FR, Matthews MJ, Aisner S, et al. Histopathologic classification of small cell lung cancer. Changing concepts and terminology. Cancer 1988;62:973-7.
Stem cell markers in SCLC [2] Zochbauer-Muller S, Gazdar AF, Minna JD. Molecular pathogenesis of lung cancer. Annu Rev Physiol 2002;64:681-708. [3] Giangreco A, Groot KR, Janes SM. Lung cancer and lung stem cells: strange bedfellows? Am J Respir Crit Care Med 2007;175: 547-53. [4] Doyonnas R, Nielsen JS, Chelliah S, et al. Podocalyxin is a CD34related marker of murine hematopoietic stem cells and embryonic erythroid cells. Blood 2005;105:4170-8. [5] Brandenberger R, Wei H, Zhang S, et al. Transcriptome characterization elucidates signaling networks that control human ES cell growth and differentiation. Nat Biotechnol 2004;22:707-16. [6] Burkert J, Wright NA, Alison MR. Stem cells and cancer: an intimate relationship. J Pathol 2006;209:287-97. [7] Kim CG, Lee JJ, Jung DY, et al. Profiling of differentially expressed genes in human stem cells by cDNA microarray. Mol Cells 2006;21: 343-55. [8] Nichols J, Zevnik B, Anastassiadis K, et al. Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 1998;95:379-91. [9] Stanhope-Baker P, Kessler PM, Li W, Agarwal ML, Williams BR. The Wilms tumor suppressor-1 target gene podocalyxin is transcriptionally repressed by p53. J Biol Chem 2004;279:33575-85. [10] Ney JT, Zhou H, Sipos B, et al. Podocalyxin-like protein 1 expression is useful to differentiate pancreatic ductal adenocarcinomas from adenocarcinomas of the biliary and gastrointestinal tracts. HUM PATHOL 2007;38:359-64. [11] Han H, Landreneau RJ, Santucci TS, et al. Prognostic value of immunohistochemical expressions of p53, HER-2/neu, and bcl-2 in stage I non–small-cell lung cancer. HUM PATHOL 2002;33: 105-10. [12] Sheikh HA, Fuhrer K, Cieply K, Yousem S. p63 expression in assessment of bronchioloalveolar proliferations of the lung. Mod Pathol 2004;17:1134-40. [13] Lee JS, Yoon A, Kalapurakal SK, Ro JY, et al. Expression of p53 oncoprotein in non-small-cell lung cancer: a favorable prognostic factor. J Clin Oncol 1995;13:1893-903. [14] Butta N, Larrucea S, Alonso S, et al. Role of transcription factor Sp1 and CpG methylation on the regulation of the human podocalyxin gene promoter. BMC Mol Biol 2006;7:17. [15] Jordan CT, Guzman ML, Noble M. Cancer stem cells. N Engl J Med 2006;355:1253-61.
1605 [16] Eramo A, Lotti F, Sette G, et al. Identification and expansion of the tumorigenic lung cancer stem cell population. Cell Death Differ 2007. [17] Schopperle WM, DeWolf WC. The TRA-1-60 and TRA-1-81 human pluripotent stem cell markers are expressed on podocalyxin in embryonal carcinoma. Stem Cells 2007;25:723-30. [18] Somasiri A, Nielsen JS, Makretsov N, et al. Overexpression of the antiadhesin podocalyxin is an independent predictor of breast cancer progression. Cancer Res 2004;64:5068-73. [19] Kelley TW, Huntsman D, McNagny KM, Roskelley CD, Hsi ED. Podocalyxin: a marker of blasts in acute leukemia. Am J Clin Pathol 2005;124:134-42. [20] Casey G, Neville PJ, Liu X, et al. Podocalyxin variants and risk of prostate cancer and tumor aggressiveness. Hum Mol Genet 2006;15: 735-41. [21] Wistuba II. Molecular pathology: applications of molecular biology in pathological anatomy. Rev Med Chil 2001;129:791-804. [22] Mason JF, Murray JF, Courtney Broaddus V, Nadel JA, editors. Textbook of respiratory medicine. Philadelphia: Elsevier Saunders; 1994. [23] Watkins DN, Berman DM, Burkholder SG, Wang B, Beachy PA, Baylin SB. Hedgehog signalling within airway epithelial progenitors and in small-cell lung cancer. Nature 2003;422:313-7. [24] Khoor A, Gray ME, Singh G, Stahlman MT. Ontogeny of Clara cell– specific protein and its mRNA: their association with neuroepithelial bodies in human fetal lung and in bronchopulmonary dysplasia. J Histochem Cytochem 1996;44:1429-38. [25] Reynolds SD, Giangreco A, Power JH, Stripp BR. Neuroepithelial bodies of pulmonary airways serve as a reservoir of progenitor cells capable of epithelial regeneration. Am J Pathol 2000;156:269-78. [26] Hong KU, Reynolds SD, Watkins S, Fuchs E, Stripp BR. Basal cells are a multipotent progenitor capable of renewing the bronchial epithelium. Am J Pathol 2004;164:577-88. [27] Sewalt RG, Lachner M, Vargas M, et al. Selective interactions between vertebrate Polycomb homologs and the SUV39H1 histone lysine methyltransferase suggest that histone H3-K9 methylation contributes to chromosomal targeting of Polycomb group proteins. Mol Cell Biol 2002;22:5539-53. [28] Satijn DP, Otte AP. Polycomb group protein complexes: do different complexes regulate distinct target genes? Biochim Biophys Acta 1999; 1447:1-16. [29] Gil J, Bernard D, Peters G. Role of Polycomb group proteins in stem cell self-renewal and cancer. DNA Cell Biol 2005;24:117-25.
www.elsevier.com/locate/humpath
Original contribution
Stem cell marker expression in small cell lung carcinoma and developing lung tissue Lin-Kristin Koch a , Hui Zhou MD a , Jörg Ellinger MD b , Katharina Biermann MD a , Tobias Höller c , Alexander von Rücker MD a , Reinhard Büttner MD a , Ines Gütgemann MD a,⁎ a
Institut of Pathology, University of Bonn, D-53127 Bonn, Germany Department of Urology, University of Bonn, D-53127 Bonn, Germany c Institute for Biometry, Informatics and Epidemiology, University of Bonn, D-53127 Bonn, Germany b
Received 19 November 2007; revised 1 February 2008; accepted 1 March 2008
Keywords: Stem cells; Small cell lung cancer; Podocalyxin like protein-1; Polycomb group protein; Bmi-1; Immunhistochemistry
Summary Histopathologic and clinical findings suggest that small cell lung cancer is derived from a multipotent proximal airway epithelial cell. In order to investigate the histogenetic origin of small cell lung cancer, we compared stem cell marker expression in human fetal lung tissue, human adult bronchial tissue, and a cohort of 64 small cell lung cancers. Supporting derivation of a multipotent precursor cell, 87.5% (56/64) of small cell lung cancers showed a dot-like expression of podocalyxinlike protein 1 (PODXL-1), a marker of embryonic and hematopoetic stem cells. Of small cell lung cancers, 98.4% (63/64) ubiquitously expressed Bmi-1, a key player in self-renewal of stem cells. Oct4 and AP2γ were not expressed. Although podocalyxin-like protein 1 did not correlate with p53 or Wilms tumor suppressor 1, known regulators of podocalyxin-like protein 1, we could demonstrate demethylated CpG islands in the podocalyxin-like protein 1 promoter in small cell lung cancer, indicating epigenetic regulation. During fetal lung development and within adult bronchial mucosa, Bmi-1 was expressed ubiquitously. In contrast, podocalyxin-like protein 1 was detected in few stromal cells during the pseudoglandular phase (n = 7) and, importantly, in clustered epithelial cells within proximal bronchi and the trachea during the canalicular phase (n = 10). Interestingly, podocalyxin-like protein 1 was not expressed in normal or metaplastic adult bronchial epithelium (n = 36) but was expressed in sparse epithelial cells in half of the cases of normal tumor adjacent bronchial mucosa (20/40). Taken together, we show that small cell lung cancers and clustered epithelial cells in developing proximal bronchi share the expression of stem cell markers, suggesting a possible histogenetic link. © 2008 Elsevier Inc. All rights reserved.
1. Introduction
⁎ Corresponding author. Tel.: +49 228 287 16968; fax: +49 228 287 15030. E-mail address: [email protected] (I. Gütgemann). 0046-8177/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2008.03.008
Small cell lung cancer (SCLC) is a high-grade malignant epithelial tumor typically arising in a proximal and peribronchial location. Morphologically, SCLC resembles stem cells as tumor cells are small and round to fusiform, have a high nuclear to cytoplasmatic ratio, finely
1598 granular chromatin, absent or inconspicuous nucleoli, and frequent mitoses [1]. The subtype of combined small cell carcinoma demonstrates multipotent differentiation of tumor cells, as combined small cell carcinomas can consist of areas of pure SCLC with components of squamous cell carcinoma, adenocarcinoma, neuroendocrine carcinoma, spindle cell carcinoma, or giant cell carcinoma. Often, SCLC is unresectable because of its discohesive growth pattern within the lung and early systemic spread. Some patients with SCLC respond favorably to first-line, multimodal chemotherapy but often relapse within a short time [2]. Histopathologic and clinical findings suggest that SCLC is derived from cells that bear characteristics of multipotent proximal airway epithelial cells [3]. In this study, we further explored this concept by investigating stem cell marker expression in SCLC as well as in adult and fetal respiratory epithelium. As identifying markers of human proximal airway stem cells are currently unknown, we chose podocalyxin-like protein 1 (PODXL-1), a marker of hematopoetic and embryonic stem cells [4,5]; Bmi-1, a key regulator in stem cell homeostasis [6]; a putative embryonic stem cell marker, AP2γ [7]; and a well-characterized marker of pluripotency, Oct4 [8] for our study. Furthermore, we analyzed the methylation status of the PODXL-1 promoter in
L. -K. Koch et al. SCLC as well as expression of Wilms tumor suppressor 1, a positive transcriptional regulator and p53, a negative regulator of PODXL-1 expression [9].
2. Methods 2.1. Case selection Formalin-fixed, paraffin-embedded bronchial biopsies (n = 94), including SCLC (n = 64), normal bronchial epithelium located adjacent to tumor (n = 40), normal respiratory epithelium from lung biopsies without any metaplastic disease (n = 27), and bronchial epithelium with squamous cell metaplasia (n = 3), were retrieved from the archives of the Institute of Pathology, University of Bonn (Bonn, Germany). Furthermore, proximal bronchi from pulmonary lobectomy specimens (n = 6) with normal respiratory epithelium were assessed as well as autopsy material of normal fetal lungs (n = 21). Standard paraffin tissue sections were stained with hematoxylin-eosin (H&E) for conventional histology. The diagnosis of SCLC was verified by pankeratin, TTF-1, and CD56 immunhistochemical staining.
Fig. 1 Immunhistochemical analysis in SCLC showing dot-like (A) or membranous (A, inset) staining with anti-PODXL-1 and strong nuclear Bmi-1 expression in tumor cells (B). Scattered, sparse PODXL-1 positive cells are found in the tumor adjacent bronchial epithelium (C), whereas Bmi-1 staining is ubiquitously expressed in most epithelial cells (D) (scale bar = 50 μm).
Stem cell markers in SCLC
1599
2.2. Immunohistochemistry
2.4. DNA isolation, bisulphite modification, and methylation-specific polymerase chain reaction
Sections (2-3 μm) were deparaffinized by routine techniques, placed in 200 ml of target retrieval solution (citrate buffer of pH 6.0) and heated for 30 minutes at boiling temperature (microwave/600 W). After cooling for 20 minutes, sections were washed with Tris-buffered saline. Immunohistochemical staining was performed by an automated immunohistochemical staining stainer (“Tech Mate 500” Autostainer, DAKO Cytomation, Glostrup, Denmark) using the streptavidine-biotin-peroxidase complex technique. Endogenous peroxidase activity was blocked by treatment with 3% H2O2 for 5 minutes. Slides were developed with EnVisionTM (DAKO). Primary antibodies used were PODXL-1 (R&D Systems, Wiesbaden-Nordenstadt, Germany; clone: AF 1658, 1:50, overnight), Bmi-1 (Chemicon, Hofheim, Germany; clone: F6, 1:50, overnight), Oct4 (Santa Cruz, Biotechnology, Santa Cruz, CA; clone: C-10, 1:50, overnight), AP2γ (Santa Cruz, clone: H-77, 1:500, overnight), p53 (DAKO, Hamburg, Germany; clone: DO-7, 1:100, 30 minutes), Wilms tumor suppressor 1 (WT1) (DCS, clone: WLM04, 1:400, overnight), CD31 (DAKO, clone: JC/70A, 1:100), and p63 (DAKO, clone: 4A4, 1:100, overnight). Endothelial cells served as internal staining controls for PODXL-1 and normal respiratory epithelium for Bmi-1.
2.3. Scoring Standard sections were evaluated for morphology and immunostaining by two independent investigators. PODXL-1 expression was found in a characteristic dot like or membranous pattern in SCLCs (Fig. 1A). Immunolabeling in less than 5% of tumor cells was scored as negative, and staining in greater than 5% of tumor cells, as positive, as described before [10]. All tumor cells within SCLC biopsies were scored. As in SCLCs, different staining intensities of PODXL-1 were found; strong staining in more than 5% of tumor cells was distinguished from weak staining. Nuclear Bmi-1 staining in SCLCs was scored in a 3-tiered manner as described above for PODXL-1. Strong nuclear p53 immunolabeling in more than 10% of tumor cell nuclei and nuclear p63 staining were interpreted as positive [11,12]. In addition, frequency of p53 immunoreactivity was assessed in SCLCs by subdividing the cases into three groups: nonimmunoreactive and immunoreactive in 50% or less or more than 50% of nuclei [13].
Table 1
Positive Negative
Four 5-μm slices of formalin-fixed, paraffin-embedded tissue biopsies were deparaffinized using xylol and ethanol. The Qiagen Robot M48 (Hilden, Germany) was used to isolate genomic DNA using the MagAttract DNA Mini Kit (Quiagen; final elution volume: 100 μL). As positive and negative controls for polymerase chain reaction (PCR) analysis, testes embryonal carcinoma (Tera-1) cell line– derived DNA and healthy volunteers' white blood cell (WBC) DNA was isolated. Universal methylated DNA was prepared by treating of WBC-DNA with SssI CpGMethylase (New England Biolabs, Frankfurt, Germany). We performed methylation (MSP)/unmethylation-specific (USP) PCR of the promoter region of PODXL-1 (MSP: forward 5′-TTT TTT AAG GCG CGG AGG TC-3′ and reverse 5′-CGC TAA ACC GTA AAC AAT AAA CA-3′; USP: forward 5′-TTA AGG TGT GGA GGT TGT TGG-3′ and reverse 5′- CCA AAA CCA AAA CTA AAC AAA CAC-3′). These primer sets targeted the CpGs (CpG no. −23 to −27 relative to transcription start site) previously shown to be differentially methylated in Tera-1 (unmethylated, PODXL-1 expression) and Meg01 cells (human megakaryoblastic cells; hypermethylated and PODXL-1 not expressed) [14]. All PCR experiments were carried out in a volume of 10 μL with 384-well plates and an ABIPrism 7900HT (Perkin Elmer, Foster City, CA). Samples (1.5-μL bisulphite-treated DNA) were analyzed in triplicate containing 5-μL Power SYBR Green Master Mix (Applied Biosystems) and 200 nmol/L of each forward and reverse primer. PCR amplification was performed as follows: 95°C for 10 minutes, followed by 40 cycles at 95°C for 15 seconds and 59°C for 60 seconds. Following the last cycle, a dissociation curve analysis was performed.
2.5. Statistical analysis Comparative analysis of ordinal data, such as p53 expression in SCLC compared with either PODXL-1 or Bmi-1 (subdivided by intensity in “negative,” “weak,” and “strong”), was performed using the Cochran-Armitage test. For dichotomous data such as sex and age groups (b63 years versus ≥63 years) the Fisher exact test was used. Statistical significance was determined when the 2-sided exact P value of a test was less than .05. All analyses were performed using SPSS version 14.0.2 (SPSS, Chicago, IL).
Immunhistochemical expression of stem cell marker PODXL-1, Bmi-1, and p53 in SCLC (n = 64) Intensity
PODXL-1–reactive
Bmi-1–reactive
p53-reactive
Strong Weak Negative
73.4% (47/64) 14.1% (9/64) 12.5% (8/64)
92.2% (59/64) 6.3% (4/64) 1.5% (1/64)
65.6% (42/64) 34.4% (22/64)
1600
3. Results 3.1. Stem cell marker and p53 expression in SCLC Overall, 87.5% (56/64) of SCLCs showed coexpression of PODXL-1 and Bmi-1 (Fig. 1). Of the PODXL-1 immunoreactive SCLCs, 73.4% (47/64) could be subdivided in a strong 14.1% (9/64) and a weak dot-like (Fig. 1A) or membranous staining pattern (Fig. 1A insert). Of SCLCs, 12.5% (8/64) were PODXL-1 negative; 92.2% (59/64) cases of SCLC strongly expressed Bmi-1 in all examined tumor nuclei, 6.3% (4/64) stained weakly, and 1.5% (1/64) were negative for Bmi-1 (Table 1). In contrast, putative embryonic stem cell markers AP2γ (0/37), Oct4 (0/37), and WT1 (0/37) were not expressed in SCLCs. Likewise, SCLCs were negative for p63 and CD31 (n = 11, data not shown). Nuclear p53 expression was observed in 65.6% (42/64) of SCLCs and no p53 immunostaining was observed in 34.4% (22/64) of cases. No significant correlation was seen when comparing p53 expression in SCLC with either PODXL-1 or Bmi-1 (P = 1.000 and P = .344, respectively). When p53 frequencies
L. -K. Koch et al. were taken into account by subdividing SCLCs into groups with 50% or less or more than 50% of nuclear p53 reactivity, no significant correlation with PODXL-1 (P = .877) nor Bmi-1 immunostaining was observed (P = .721). No correlation was found when comparing age (b63 years: n = 24, ≥63 years: n = 40) and sex (male: n = 43, female: n = 21) of patients suffering from SCLC with PODXL-1 and Bmi-1 expression in tumor sections.
3.2. PODXL-1 promoter methylation of CpG islands in SCLC Next, we assessed the methylation status of CpG-rich islands within the PODXL-1 promoter, which has been shown to correlate with transcriptional activity [14]. In 8 of 9 SCLCs, we detected unmethylated PODXL-1 products with the USP primer set (peak in dissociation curve at ∼72°C). Here, no signal was observed using the MSP primer set (Fig. 2). The Tera-1 cell line DNA was unmethylated at the investigated CpG sites (Fig. 2), as has been described [14]. Universal methylated WBC-DNA demonstrated a peak 75.7°C using the MSP primers.
Fig. 2 Methylation status of CpG island of the PODXL-1 promoter in SCLC and control cell lines and peripheral blood leucocytes (WBC DNA), identified by dissociation curve analysis using unmethylation specific (USP) (left) and methylation specific primer sets (MSP) (right). A peak at 72.8°C using the USP-primer set indicates unmethylated DNA (Tera-1 cell line, SCLC). A peak using MSP primers at 75.7°C is seen with universal methylated control DNA prepared by treating WBC-DNA with SssI CpG-Methylase (Sssl). Eight of nine SCLC samples showed similar peaks at 72.8°C (SCLC) indicating unmethylated CpG islands. During this procedure, the double-stranded PCR product is denatured according to its unique melting Temperature. A peak at 75.7°C indicated a methylated DNA in the MSP, and a product at 72.8°C indicated unmethylated DNA in the USP. Abbreviations: NTC, nontissue control.
Stem cell markers in SCLC
1601
Fig. 3 Immunhistochemical analysis of fetal lung tissue. Pseudoglandular phase: H&E (A), PODXL-1 dot-like expression in very few stromal cells (B), strong ubiquitous Bmi-1 staining (C), and p63 expression limited to the proximal bronchial tree (D). Canalicular phase: H&E (E), clustered epithelial cells show PODXL-1 expression in particular in deep pockets of proximal bronchial and tracheal epithelium (F), ubiquitous nuclear Bmi-1 expression in epithelial and stromal cells (G), and p63 staining in basal cells (H) (scale bar = 25 μm).
1602 Table 2
L. -K. Koch et al. Immunhistochemical expression of PODXL-1, Bmi-1, and p63 in fetal (n = 21) and normal adult bronchial tissue (n = 33)
Phase of lung development
PODXL-1
Bmi-1
p63
Pseudoglandular phase (5-17 wk) Canalicular phase (18-26 wk)
1% of stromal cells, dot like
Ubiquitous in stromal and epithelial cells Ubiquitous in stromal and epithelial cells
Basal cells, proximal bronchial tree Basal cells, proximal bronchial tree
Saccular phase (26-38 wk)
Normal bronchial tissue
Clustered epithelial cells, especially deep pockets, proximal bronchi and trachea, dot like Clustered epithelial cells, especially deep pockets, proximal bronchi and trachea, dot-like No expression in normal bronchial epithelium
3.3. Stem cell marker expression in fetal lung Analysis of Bmi-1 and PODXL-1 expression in fetal lung tissue (n = 21) was subdivided into three periods of lung formation: pseudoglandular phase (5-17 weeks) (n = 7), canalicular phase (18-26 weeks) (n = 10), and saccular phase (26-38 weeks) (n = 3). During the pseudoglandular phase, there were no morphologically detectable clustered epithelial cells. Few stromal cells in fetal lung tissue during this phase showed a weak dot-like PODXL-1 expression (Fig. 3B). However, CD31 staining was observed in single positive cells and stromal capillary vessels (data not shown) indicating that at least some of the PODXL-1 positive cells may correspond to sprouting endothelial cells. In contrast, we observed PODXL-1 staining during the canalicular phase in sparse clustered epithelial cells in major bronchi (Fig. 3F). Here, PODXL-1–positive cells were frequently observed in deep pockets of major bronchi or lobar bronchi in between cartilaginous nests. p63 highlighted basal cells underlining respiratory epithelial cells within major bronchi or lobar bronchi. These p63-positive cells were located lower within the epithelium than PODXL-1–positive cells and were found at high abundance within the trachea and major bronchi (Fig. 3D and H, Table 2). Beginning with the saccular phase, the frequency of PODXL-1 stained cells decreased until birth, where only very few epithelial cells showed PODXL-1 expression (data not shown). Bmi-1 was expressed in stromal as well as epithelial cells from the fifth week until maturity (n = 4, Fig. 3C and G, Table 2).
3.4. Stem cell marker expression in normal bronchial, reactive respiratory epithelium in adult lung, and tumor-adjacent normal bronchial epithelium Bronchial biopsies showing normal respiratory epithelium (n = 27), metaplastic lesions (squamous cell metaplasia
Ubiquitous in stromal and epithelial cells Normal bronchial epithelial cells and lymphocytes
Basal cells, proximal bronchial tree
n = 3), and proximal bronchi from pulmonary lobectomy specimens (n = 6) with normal respiratory epithelium showed no PODXL-1 expression (Fig. 4B and F). In contrast, half of the cases of normal bronchial epithelium located adjacent to tumor (20/40) showed focal PODXL-1 expression (Fig. 1C). Normal respiratory epithelial cells of major bronchi, metaplastic epithelium, and bronchial epithelium adjacent to tumor cells demonstrated a ubiquitous nuclear Bmi-1 staining pattern within most of the nuclei. p63 highlighted basal cells within the respiratory epithelium above the basement membrane in adult bronchi and was also detected in squamous metaplasia (Fig. 4D and H).
4. Discussion Here, we show that PODXL-1 and Bmi-1, but not AP2γ, Oct4, and WT1, are expressed in SCLC (Fig. 1). The expression of stem cell markers such as PODXL-1 and Bmi1 in tumor cells underscores the possible link between cancer stem cells and normal stem cells [15]. This hypothesis is supported by tumor cell transplantation studies using CD133 as a marker of cancer stem cells in lung tumors and observations that CD133 expressing cells are increased in regenerating lung tissue [16]. PODXL-1 is a heavily glycosylated sialomucin expressed in hematopoietic, embryonic, and mesenchymal stem cells [4,5,17]. Like other stem cell markers, it has also been described in human malignancies, in particular, in pancreatic [10] and breast carcinoma [18] as well as in leukaemias [19]. Inappropriate expression or function of PODXL-1 may be caused by hypermethylation of the PODXL-1 promoter [14]; gene amplification, which has been proposed for prostate cancer [20]; or inappropriate expression of PODXL-1 due to changes in the WT1 and/or p53 activity [9]. In comparing the expression of PODXL-1 and p53, no significant correlation was found indicating that protein levels of PODXL-1 in
Fig. 4 Immunhistochemical staining of normal adult bronchial epithelium (A-D), and squamous metaplasia (E-H): H&E (A and E), lack of PODXL-1 expression (B and F), ubiquitous Bmi-1 expression within the majority of epithelial nuclei (C and G). p63 highlights cells of the respiratory epithelium above the basement membrane within the lower level of bronchial epithelium (D and H) (scale bar = 50 μm).
Stem cell markers in SCLC
1603
1604 SCLC may be influenced by other factors. Instead, we observed unmethylated DNA in CpG-rich islands of the promoter region, suggesting that DNA demethylation may be involved in PODXL-1 regulation in SCLC (Fig. 2). Once airway epithelial stem cells have been identified unambiguously, it will be interesting to compare the epigenetic regulation of the PODXL-1 gene in these cells and SCLCs. The cell of origin of SCLC is still enigmatic. SCLCs, because of the expression of neuroendocrine markers such as chromogranin, neuron-specific enolase [21], and calcitonin gene–related peptide have been proposed to be derived from neuroendocrine cells (NECs) [22]. On the other hand, SCLC also shares characteristics of proximal bronchial precursor cells that give rise to respiratory epithelial cells during development and repair and, like these cells, depend on the ligand-dependent Sonic Hedgehog (Shh) activation pathway [23]. Although it is unclear, at the present time, whether these cells truly fulfill stem cell characteristics including long-term cell division, Shh-dependent epithelial airway regeneration occurs from these cells, followed by neuroendocrine differentiation [23]. In order to identify a corresponding counterpart of SCLC, we investigated PODXL-1 expression in developing lung and bronchial tissue, where these cells might be more frequent than in adult bronchiopulmonary biopsies. There are many different cell types described within the developing respiratory epithelium, including ciliated and nonciliated (Clara) cells, basal cells, and NECs. The latter can not only be found as scattered single cells within the proximal bronchial tree but also in groups related with nerve fibers called neuroepithelial bodies [24,25]. Within the unipotent precursor population of NECs within neuroepithelial bodies, variant Clara cells with multipotent stem cell features have been identified in mice [25]. Here, we find that PODXL-1 is expressed in a small subset of cuboidal epithelial cells within proximal airways (Fig. 3F) during the canalicular phase (1826 weeks) of human lung development. Whether these cells are similar to variant Clara cells in adult bronchi cannot be investigated at the present time, as defining human markers are lacking. Colocalization and functional studies are needed to address this possibility. During the pseudoglandular phase of lung morphogenesis (5-17 weeks), PODXL-1 was expressed in developing vessels and within single stromal cells. Here, because of the high abundance of sprouting endothelial cells during the pseudoglandular phase highlighted by CD31 staining, single PODXL-1–positive cells may correspond to sprouting endothelial cells rather than precursor cells. Interestingly, however, the distribution of PODXL-1–positive cells in human fetal lung is similar to Shh-activated cells in developing murine lung: Shh-activated single cells are observed within the stroma during the pseudoglandular phase. During the canalicular phase, clustered epithelial calcitonin gene–related peptide and patched double-positive cells are found within proximal bronchi next to Shh expressing epithelial cells [23].
L. -K. Koch et al. Basal cells have been described in mice as multipotent reserve cells, which can differentiate into each of the mature cell types of the bronchial epithelium [26]. It is unclear whether, in our study, some p63 positive cells in fetal lungs show coexpression of PODXL-1 expression. However, PODXL-1 staining cells were found predominantly in deep folds of bronchial epithelium, whereas p63 stained a cell layer underlying the cuboidal epithelial cells of major bronchi during the canalicular phase (Fig. 3H). This observation is in line with previous evidence, supporting that squamous carcinoma but not SCLC is derived from basal cells [3]. Bmi-1 expression was observed in 98.4% (63/64) cases of SCLC in all examined tumor nuclei. Bmi-1 is a member of the Polycomb group family of proteins, which represent epigenetic chromatin modifiers [27,28]. It is described in numerous other human cancers [29]. Dysregulation of normal self-renewal pathways such as the Shh pathway and the products of Bmi-1 and other Polycomb Group genes may explain tumorigenesis as because of dysregulated stem cell proliferation [6]. Our study shows that Bmi-1 in the lung is not a phenotypic marker for stem cells but, instead, stains self-renewing cells in normal bronchial epithelium, proliferating cells in SCLC, and proliferating cells during fetal lung development. Interestingly, Bmi-1 is a downstream effector of the extracellular signaling molecule Shh. As mentioned before, this pathway plays a pivotal role during normal lung development, airway repair, and in SCLC proliferation [23]. Therefore, it is likely, that Bmi-1 overexpression in SCLC is a result of juxtacrine Shh signaling. In summary, the present study is the first to demonstrate differential expression of PODXL-1 and Bmi-1 in SCLC, human fetal lung tissue, and adult bronchial epithelium. The localization of PODXL-1 positive cells during lung morphogenesis underscores the hypothesis that there is a possible histogenetic link between SCLC and multipotent airway epithelial cells in the proximal bronchial tree that have retained expression of stem cell markers. In the future, it will be interesting to examine whether PODXL-1–expressing cells during the canalicular phase of human lung development bear functional stem cell characteristics and how epigenetic mechanisms govern SCLC tumorigenesis.
Acknowledgments We thank S Reynolds and H Schorle for helpful discussions and C Esch for technical assistance.
References [1] Hirsch FR, Matthews MJ, Aisner S, et al. Histopathologic classification of small cell lung cancer. Changing concepts and terminology. Cancer 1988;62:973-7.
Stem cell markers in SCLC [2] Zochbauer-Muller S, Gazdar AF, Minna JD. Molecular pathogenesis of lung cancer. Annu Rev Physiol 2002;64:681-708. [3] Giangreco A, Groot KR, Janes SM. Lung cancer and lung stem cells: strange bedfellows? Am J Respir Crit Care Med 2007;175: 547-53. [4] Doyonnas R, Nielsen JS, Chelliah S, et al. Podocalyxin is a CD34related marker of murine hematopoietic stem cells and embryonic erythroid cells. Blood 2005;105:4170-8. [5] Brandenberger R, Wei H, Zhang S, et al. Transcriptome characterization elucidates signaling networks that control human ES cell growth and differentiation. Nat Biotechnol 2004;22:707-16. [6] Burkert J, Wright NA, Alison MR. Stem cells and cancer: an intimate relationship. J Pathol 2006;209:287-97. [7] Kim CG, Lee JJ, Jung DY, et al. Profiling of differentially expressed genes in human stem cells by cDNA microarray. Mol Cells 2006;21: 343-55. [8] Nichols J, Zevnik B, Anastassiadis K, et al. Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 1998;95:379-91. [9] Stanhope-Baker P, Kessler PM, Li W, Agarwal ML, Williams BR. The Wilms tumor suppressor-1 target gene podocalyxin is transcriptionally repressed by p53. J Biol Chem 2004;279:33575-85. [10] Ney JT, Zhou H, Sipos B, et al. Podocalyxin-like protein 1 expression is useful to differentiate pancreatic ductal adenocarcinomas from adenocarcinomas of the biliary and gastrointestinal tracts. HUM PATHOL 2007;38:359-64. [11] Han H, Landreneau RJ, Santucci TS, et al. Prognostic value of immunohistochemical expressions of p53, HER-2/neu, and bcl-2 in stage I non–small-cell lung cancer. HUM PATHOL 2002;33: 105-10. [12] Sheikh HA, Fuhrer K, Cieply K, Yousem S. p63 expression in assessment of bronchioloalveolar proliferations of the lung. Mod Pathol 2004;17:1134-40. [13] Lee JS, Yoon A, Kalapurakal SK, Ro JY, et al. Expression of p53 oncoprotein in non-small-cell lung cancer: a favorable prognostic factor. J Clin Oncol 1995;13:1893-903. [14] Butta N, Larrucea S, Alonso S, et al. Role of transcription factor Sp1 and CpG methylation on the regulation of the human podocalyxin gene promoter. BMC Mol Biol 2006;7:17. [15] Jordan CT, Guzman ML, Noble M. Cancer stem cells. N Engl J Med 2006;355:1253-61.
1605 [16] Eramo A, Lotti F, Sette G, et al. Identification and expansion of the tumorigenic lung cancer stem cell population. Cell Death Differ 2007. [17] Schopperle WM, DeWolf WC. The TRA-1-60 and TRA-1-81 human pluripotent stem cell markers are expressed on podocalyxin in embryonal carcinoma. Stem Cells 2007;25:723-30. [18] Somasiri A, Nielsen JS, Makretsov N, et al. Overexpression of the antiadhesin podocalyxin is an independent predictor of breast cancer progression. Cancer Res 2004;64:5068-73. [19] Kelley TW, Huntsman D, McNagny KM, Roskelley CD, Hsi ED. Podocalyxin: a marker of blasts in acute leukemia. Am J Clin Pathol 2005;124:134-42. [20] Casey G, Neville PJ, Liu X, et al. Podocalyxin variants and risk of prostate cancer and tumor aggressiveness. Hum Mol Genet 2006;15: 735-41. [21] Wistuba II. Molecular pathology: applications of molecular biology in pathological anatomy. Rev Med Chil 2001;129:791-804. [22] Mason JF, Murray JF, Courtney Broaddus V, Nadel JA, editors. Textbook of respiratory medicine. Philadelphia: Elsevier Saunders; 1994. [23] Watkins DN, Berman DM, Burkholder SG, Wang B, Beachy PA, Baylin SB. Hedgehog signalling within airway epithelial progenitors and in small-cell lung cancer. Nature 2003;422:313-7. [24] Khoor A, Gray ME, Singh G, Stahlman MT. Ontogeny of Clara cell– specific protein and its mRNA: their association with neuroepithelial bodies in human fetal lung and in bronchopulmonary dysplasia. J Histochem Cytochem 1996;44:1429-38. [25] Reynolds SD, Giangreco A, Power JH, Stripp BR. Neuroepithelial bodies of pulmonary airways serve as a reservoir of progenitor cells capable of epithelial regeneration. Am J Pathol 2000;156:269-78. [26] Hong KU, Reynolds SD, Watkins S, Fuchs E, Stripp BR. Basal cells are a multipotent progenitor capable of renewing the bronchial epithelium. Am J Pathol 2004;164:577-88. [27] Sewalt RG, Lachner M, Vargas M, et al. Selective interactions between vertebrate Polycomb homologs and the SUV39H1 histone lysine methyltransferase suggest that histone H3-K9 methylation contributes to chromosomal targeting of Polycomb group proteins. Mol Cell Biol 2002;22:5539-53. [28] Satijn DP, Otte AP. Polycomb group protein complexes: do different complexes regulate distinct target genes? Biochim Biophys Acta 1999; 1447:1-16. [29] Gil J, Bernard D, Peters G. Role of Polycomb group proteins in stem cell self-renewal and cancer. DNA Cell Biol 2005;24:117-25.