Gene expression profiling of pulmonary neuroendocrine neoplasms: A comprehensive overview

Author’s Accepted Manuscript Gene expression profiling of pulmonary neuroendocrine neoplasms: A comprehensive overview Dorian R.A. Swarts, Frans C.S. Ramaekers, ErnstJan M. Speel www.elsevier.com/locate/ctrc

PII: DOI: Reference:

S2213-0896(15)30015-3 http://dx.doi.org/10.1016/j.ctrc.2015.09.002 CTRC59

To appear in: Cancer Treatment Communications Received date: 13 March 2015 Revised date: 23 April 2015 Accepted date: 8 September 2015 Cite this article as: Dorian R.A. Swarts, Frans C.S. Ramaekers and Ernst-Jan M. Speel, Gene expression profiling of pulmonary neuroendocrine neoplasms: A comprehensive overview, Cancer Treatment Communications, http://dx.doi.org/10.1016/j.ctrc.2015.09.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Gene expression profiling of pulmonary neuroendocrine neoplasms: A comprehensive overview Dorian R.A. Swartsa1; Frans C.S. Ramaekersa; Ernst-Jan M.Speelb* Departments of aMolecular Cell Biology ([email protected]) and bPathology ([email protected]), GROW – School for Oncology and Developmental Biology, Maastricht University Medical Center, Maastricht, The Netherlands *Address correspondence to: Prof. Ernst-Jan M. Speel, PhD Laboratory Molecular Oncopathology & Diagnostics Department of Pathology Maastricht University Medical Center P. Debeyelaan 25 P.O. Box 5800 6202 AZ Maastricht The Netherlands Phone: +31-43-3874614 Fax: +31-43-3876613 Email: [email protected]

Short title: Gene expression of neuroendocrine lung tumors

1

Present address: Department of Biology, The University of York, York, United Kingdom; [email protected].

1

Figure 2 – Proteins involved in the spindle assembly checkpoint for which transcriptional activity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

was found to be upregulated in neuroendocrine lung neoplasms. When CDC20 is bound by BUB1R, BUB3 and MAD2 to form the mitotic checkpoint complex (MCC), its stimulating activity towards the anaphase-promoting complex/cyclosome (APC/C) is blocked. Inhibition of the APC/C will prevent the cell to enter the anaphase until all chromosomes are correctly attached to the mitotic spindle. The APC/C can polyubiquitylate and thus degrade several downstream targets, of which Cyclin B1 and CENP-F are depicted. Both cyclin B1 and CENP-F can be activated by the Forkhead transcription factor FoxM1 [115]. Also stathmin has been described to be a direct transcriptional target of FoxM1 [116]. In green the proteins or protein complexes are indicated for which the transcriptional levels were described to be upregulated in at least two studies (Table 2), while other proteins or protein complexes are shown in blue.

40

Figure 1

Isl1

Jak1 Stat3

DCX

INSM1

Cdk5

NeuroD

INS

secretion

CHGA

TCF3

ASH1

CHGB

secretion

ISL1

NCAM1

miR-375

HES6

HES1

Figure 2

MAD2

Stathmin

CDC20

BUBR1

BUB3

CENP-F

FoxM1

Cyclin B1

APC/C

MCC

Word-count: 4665

Abstract Neuroendocrine neoplasms (NENs) of the lung comprise a heterogeneous group, including small cell lung cancer (SCLC), large cell neuroendocrine carcinoma and pulmonary carcinoids. To unravel their molecular biology, microarray studies have been conducted that provided lists of differentially expressed genes between lung NENs on the one hand and normal tissue and/or non-SCLCs on the other. However, the majority of studies paid little attention to the functions of candidates and their potency as diagnostic markers and/or therapeutic targets. Furthermore, at a first glance, only limited overlap was seen amongst these individual studies concerning differentially expressed transcripts. By combining all originally published gene expression profiling studies on lung NENs, and by reevaluating differentially expressed genes, we were able to identify major factors involved in lung NEN carcinogenesis. Thirty-three genes were found to be frequently deregulated in multiple studies. Amongst these are neuroendocrine-specific factors, including ASH1, INSM1, and ISL1 and genes involved in neuronal differentiation and neurite outgrowth such as DCX and NCAM1. Also, multiple factors were involved in cell cycle progression, including members of the mitotic spindle checkpoint complex, and the regulated secretory pathway, e.g. CHGA and CHGB and CPE. For several of these candidates we propose possible functions in lung NEN carcinogenesis as well as potential roles in diagnosis and as targets for novel therapies. This review elucidates potential genes of interest in pulmonary NENs on basis of the present expression profiling literature. We advocate that a selection of the identified candidates should be examined in depth for their clinical application.

Abstract word count: 250

2

Keywords: cDNA microarrays, oligonucleotide microarrays, gene expression profiling, pulmonary neuroendocrine tumors, candidate genes, therapeutic targets

Contents 1. Introduction 2. Approach 3. Most frequently reported genes in gene expression profiling studies in lung NENs 3.1. Genes involved in neuroendocrine differentiation 3.2. Genes related to cell cycle progression and proliferation 3.3. Genes related to neuronal development and neurite outgrowth 3.4. The chromogranins and the regulated secretory pathway 3.5. Other frequently deregulated genes 4. Concluding remarks

1. Introduction2

Lung cancer is a common type of malignancy, with one of the highest mortality rates in the USA [1]. Approximately 20-25% of lung tumors are neuroendocrine in nature, with small cell lung cancer

2

Abbreviations used: APC/C, anaphase-promoting complex/cyclosome; CgA, chromogranin A; CgB, chromogranin B; CGH, comparative genomic hybridization; HCC, hepatocellular carcinoma; HGNEC, high-grade neuroendocrine carcinoma; LCNEC, large cell neuroendocrine carcinoma; MCC, mitotic spindle checkpoint;NEN, neuroendocrine neoplasm; NGS, next-generation sequencing; NSCLC, non-small cell lung cancer; PNEC, pulmonary neuroendocrine cell; SAC, spindle assembly checkpoint; SCLC, small cell lung cancer

3

(SCLC; 15-20%) being the most common and malignant subtype, while large cell neuroendocrine carcinoma (LCNEC; 30%) and pulmonary carcinoids (1-2%) comprise rarer entities [2]. Pulmonary carcinoids are further subdivided into the low grade typical carcinoid tumors, and intermediate grade atypical carcinoids. While many well-characterized oncogenes and tumor suppressor genes have been extensively studied in lung neuroendocrine neoplasms (NENs), the mechanisms of carcinogenesis of these neoplasias, and in particular of the smoking-unrelated pulmonary carcinoids, remain poorly understood. In the past ten years, a shift in molecular approaches to elucidate mechanisms of oncogenesis has become evident. Candidate gene approaches have been increasingly complemented by genomewide techniques. Next to array comparative genomic hybridization (array CGH) and gene expression profiling studies genome-wide methylome profiling and several sequencing techniques have become available. Up till now, only a few studies using these novel techniques have been performed on pulmonary NENs. Kalari et al. carried out methylation profiling on 18 primary SCLCs and cell lines using the methylated-CpG island recovery assay and identified tumor-specific methylation of several transcription factors involved in the determination of neural cell-fate [3]. Integrative analyses of multiple genome-wide approaches were performed by Peifer et al., who demonstrated that Rb- and P53-alterations are almost universal in SCLCs and identified novel disease-driving factors, including CREBBP and MLL [4]. Using a similar approach Rudin et al. observed frequent amplification of the SOX2 gene in SCLC [5]. Recently, Fernandez-Cuesta et al. showed using genome and exome sequencing that pulmonary carcinoids frequently exhibit mutations of chromatin-remodeling genes [6]. Although these recent studies have yielded important novel insights into the molecular biology of lung NENs, the large majority of genome-wide studies published so far consists of array CGH and gene expression profiling data. In a previous review [7], we have performed a meta-analysis of array CGH studies on pulmonary NENs and briefly discussed the main findings of a number of gene expression profiling studies. This included the important finding that LCNECs and SCLCs cluster either 4

separately or together, but in general more closely towards normal bronchial epithelium, whereas carcinoids grouped always together in a different cluster, which is more closely related to brain tumors than to high-grade neuroendocrine carcinomas (HGNECs) [7]. These studies provided also lists of differentially expressed genes between lung NENs on the one hand and normal tissue and/or NSCLCs on the other, which will be the focus of the current review. Since pulmonary carcinoids and LCNECs comprise relatively rare entities, and SCLCs are only seldom resected [7], relatively few gene expression profiling studies have been undertaken to these entities, rendering them understudied as compared to NSCLCs. This makes a review of available gene expression profiling studies valuable. We noticed that for individual lung NEN expression profiling studies few top-candidates overlap, which is a common problem when comparing microarray studies [8]. This is not surprising, since these studies had different aims, not all of them focusing on lung NENs, and different platforms and control tissues were used (Table 1). Moreover, even when reported by multiple studies, often little attention has been given to putative role(s) of candidate genes in lung NEN oncogenesis. Here, we review candidates reported in at least two different lung NEN gene expression profiling studies. We will discuss the biological functions of these candidates in general and their putative role within lung NEN development and progression, and their potential value for diagnosis and therapy.

2. Approach

We have selected all original research papers (1999-2014) available through PubMed that included gene expression profiling data on pulmonary NENs, using the following keywords: ("expression profiles" OR "expression profiling" OR "cDNA microarray") AND ("pulmonary neuroendocrine tumors" OR "small cell” OR "lung carcinomas" OR "lung carcinoids" OR "pulmonary carcinoids" OR “SCLC” OR “LCNEC” OR “Large Cell Neuroendocrine Carcinoma”). From these, only the studies describing gene expression profiling on primary lung NENs were included (Table 1) [9-21]. Studies not reporting on differentially expressed genes in lung NENs were excluded [18], as well as studies only 5

reporting genes differentially expressed between lung NEN subgroups [21]. We have combined gene lists (limited to the top-100 genes) as well as genes discussed elsewhere in the text body [13] of these individual studies. Subsequently, we have selected candidate genes described in at least two different studies to be differentially expressed in carcinoids, LCNECs and/or SCLCs as compared amongst each other [12] or to normal (lung) tissue and/or NSCLCs [9-11,13-17,22] (Table 2). With respect to our own gene expression profiling study [19], we have extracted the genes that are differentially expressed between carcinoid tumors and the normal reference (see Supplementary Table 1).

3. Most frequently reported genes in gene expression profiling studies in lung NENs

Table 2 lists the 33 genes that are reported in at least two gene expression profiling studies to be upregulated (n=32) or downregulated (n=1) in carcinoids, LCNECs and/or SCLCs. A subset of these genes was furthermore reported to be overexpressed in an SCLC cell line gene expression study [23]. Except for ASH1, CCNB1, CHGA, CHGB, CPE, INSM1, ISL1, NCAM1, STMN1 and TOP2A these factors have not yet been implicated in the biology of lung NENs in other studies (see Table 3), and are thereby novel candidates putatively involved in the carcinogenesis of lung NENs. Several of these 33 genes are involved in neuroendocrine differentiation, cell cycle progression and control, neuronal differentiation and neurite outgrowth, the regulated secretory system, or in multiple of these categories (Table 3A-D). A few genes could not be related to one of these main groups (Table 3E).

3.1 Genes involved in neuroendocrine differentiation

6

Several genes found to be upregulated in pulmonary NENs have established roles in neuroendocrine differentiation (Table 3A). Among these, the achaete-scute homolog (ASCL1, more commonly known as ASH1) gene is most extensively studied. It is an interesting candidate oncogene in lung NEN carcinogenesis, because an ASH1-positive precursor cell can give rise to both pulmonary neuroendocrine cells (PNECs) and other pulmonary cell types [7]. In adult lung its expression seems to be limited to PNECs and NENs. Within pulmonary tumors, the specific expression of ASH1 in SCLCs and its absence in most NSCLCs has long been acknowledged [24]. ASH1 is also expressed in other lung NENs, albeit less frequently (Table 3A), suggesting that ASH1 expression may be related to poorer differentiation grade [25]. The ASH1 protein is an important transcription factor involved in determination of the neuroendocrine lineage (Figure 1) [26]. It induces expression of typical endocrine-related genes such as CHGB (Section 3.4), HES6, INSM1 and SCG2 (secretogranin 2), and represses CDH1 (E-cadherin) as well as DKK1 and DKK3, negative regulators of Wnt-signaling [27-29]. Its antagonist HES1, which is mainly expressed in non-neuroendocrine cells, inhibits both the expression and DNA binding of ASH1 [30,31]. It has been shown that HES6, also upregulated in lung NENs, can relieve this repression [32]. Insulinoma-associated 1 (INSM1, or IA-1), a direct ASH1-target [29], is a zinc-finger transcription factor that is widely expressed in neuronal progenitor cells of the developing nervous system, in neuroendocrine organs and in a wide variety of NENs (Table 3A) [33]. INSM1 is a negative regulator of NEUROD and insulin expression, as well as its own expression (Figure 1). Furthermore, CHGA expression (Section 3.4) is strongly downregulated in INSM1 mutant mice, suggesting that INSM stimulates CHGA transcription [34]. The LIM homeodomain protein Islet-1 (ISL1) may be closely linked to INSM1 (Figure 1). ISL1 is important in the development of pancreatic islets and is expressed in the endocrine (but not exocrine) pancreas, the central and peripheral nervous system and in cardiac progenitor cells [35-37].

7

It is frequently expressed in primary and metastatic pancreatic endocrine tumors [38], but also in lung NENs (Table 3A).

3.2 Genes related to cell cycle progression and proliferation A second main group of upregulated genes in lung NENs consists of factors involved in mitotic control mechanisms and proliferation (Table 3B). To allow proper chromosome separation during mitosis, a spindle-assembly checkpoint (SAC) exists which controls the attachment of the kinetochores of the individual chromosomes to the mitotic spindle [39]. The mitotic checkpoint complex (MCC) involved in this process consists of BUB3, BUBR1, CDC20 and MAD2 [40]. The genes encoding three members of this complex, i.e. BUB1B (BUB1R), CDC20 and MAD2L1 (MAD2), were found to be upregulated in at least two cDNA microarray studies (Table 2). The complex is formed following a direct interaction between BUBR1 and MAD2 [41]. Subsequently this complex inhibits the anaphase-promoting complex/cyclosome (APC/C), preventing the start of the anaphase until all chromosomes are properly segregated [42,43]. The CDC20 member of this complex normally activates the APC/C, but looses this ability when bound to the other MCC members (Figure 2) [43]. The APC/C normally polyubiquitylates amongst others securin and cyclin B, thereby promoting their destruction and exit from mitosis [43]. Gurden et al. reported that CENP-F (centromere protein F), located at the kinetochores, is another target of CDC20 and the APC/C [44,45]. CENP-F is essential for the continuous activation of the SAC and more specifically for the sustained recruitment of BUBR1 and MAD1 at the kinetochore. These data underline the close relationship of multiple genes reported to be upregulated in lung NEN (here SCLC) microarray studies in cell cycle control mechanisms (Figure 2).

8

Deregulation of the SAC can lead to aneuploidy and chromosomal instability and has therefore been implicated in cancer [46,47]. Alterations of MAD2, CENP-F and BUB1R have indeed been implicated in the induction of these processes (Table 3B). A number of other factors not directly related to the SAC are also involved in cell cycle control and proliferation. FoxG1 is a negative regulator of TGF-β signaling by forming a complex with Smad3, Smad4 and FoxO3 [48]. This complex normally induces the expression of the cyclin-dependent kinase inhibitor p21 at the transcriptional level but looses this capability when FoxG1 is bound. Indeed, FoxG1 has been shown to repress p21 expression in multiple neoplasms [48,49]. This is particularly relevant since p21 is a tumor suppressor factor which is functionally involved in cell-cycle inhibition and commonly deregulated in human cancers [50]. In contrast to FoxG1, the Menin tumor suppressor, which is frequently mutated in lung carcinoids [6,51], is a positive regulator of CDKN1A gene expression in the case of DNA damage [52]. It has been reported that lung carcinoids are negative for p21 immunostaining [53], while also 26 out of 32 SCLCs tested in another study were negative [54]. Minichromosome maintenance protein 4 (MCM4) is one of the subunits of the heterohexamer consisting of MCM2-7 proteins [55]. This complex is an essential component of the prereplication complex, which exists from late M to early G1 phase. Deregulation of the MCM4 protein has been implicated in genomic instability and proliferation (Table 3B). Topoisomerase IIA (Top2α) is an enzyme that is able to establish and repair double stranded DNAbreaks [56]. These transient breaks are required to unwind catenated DNA and Top2α has been shown to be essential in the segregation of replicated chromosomes. The enzyme is located specifically at the centrosomes during mitosis, where it needs to separate catenated DNA before cell division is initiated. Top2α is essential for the organization of the centromere and the localization of the chromosome passenger complex, including Aurora B kinase [57]. Depletion of Top2α causes severe defects in chromosome separation during anaphase, linking its function to the SAC. It is a 9

therapeutic target that is expressed by both high- and low-grade NENs, and many other cancers (Table 3B). Two microtubule-binding proteins, protein regulator of cytokinesis 1 (PRC1) and stathmin (STMN1), play important roles in cell cycle regulation. Both factors are Cdk1-phosphorylated, but require dephosphorylation to function [58-60]. Stathmin is a 17kDa cytosolic protein that is inactivated through Cdk1-phosphorylation at the entry of mitosis to enable the formation of the mitotic spindle [59]. It functions as a promoter of microtubule depolymerization by sequestering α-β tubulin heteromers, preventing their incorporation into growing microtubules and stimulating microtubule catastrophe [58,59]. Dephosphorylation of stathmin is subsequently required to exit mitosis. Next to its function in the cell cycle, stathmin has potential roles in regulating hormone secretion, as shown in rodent cell lines of the pituitary and insulinomas [58]. Its expression is decreased during differentiation, and stathmin was found to be overexpressed in poorly differentiated carcinomas (Table 3B) [58]. PRC1 is important in regulating midzone formation and cytokinesis is in late mitosis [60,61]. As a homodimer it probably bridges crosslinked microtubules and recruits other proteins to the microtubule caps [62]. PRC1 has also been shown to be upregulated in several cancers and its knockdown resulted in cell death in breast cancer cells (Table 3B) [63].

3.3 Genes related to neuronal development and neurite outgrowth Several of the genes reported to be upregulated in lung NENs are involved in neur(on)al development and/or neurite outgrowth. Their overexpression may be related to the specific characteristics of the likely lung NEN precursor cells, the PNECs. CELSR3 plays important roles in brain development. It is expressed in postmigratory neurons and is downregulated during neuronal maturation. Probably it mediates interactions between axons and guidepost cells to determine axonal tract formation, possibly by cell adhesion [64]. It has been

10

reported that Celsr3 suppresses neurite outgrowth in the rat [65]. The role of Celsr3 in lung NENs remains to be established, although a function similar to NCAM (described below) could be envisaged because both proteins play roles in controlling neurite outgrowth and are similar in being adhesion molecules combined with a signal transducer function. Doublecortin (DCX) is involved in brain development and neuronal migration. Its overexpression leads to microtubule stabilization and inhibition of neurite outgrowth [66]. For its localization to perinuclear microtubules, doublecortin depends on Cdk5 phosphorylation, and the effect of doublecortin on neuronal migration is at least partly dependent on Cdk5 activity [67]. Interestingly, ASH1 (Section 3.1) has been reported to stimulate lung cancer migration by inducing Cdk5 activity (Figure 1) [68]. Doublecortin may be expressed by many brain tumors [69], but the effect of its expression on tumor growth and progression remains unclear (Table 3C). The neural cell adhesion molecule 1 (NCAM1; CD56), is an important marker which is used as a standard in the classification of lung NENs next to chromogranin A (CgA; Section 3.4) and synaptophysin [70]. NCAM has been reported to be ubiquitously expressed in the spectrum of lung NENs (Table 3C) [71]. It functions as an adhesion molecule and in cell-cell interactions and is a signal transducer [72]. Pathways regulated by NCAM have been implicated in neuronal survival and synaptic plasticity [73]. Furthermore, the protein is involved in the initiation and maintenance of epithelial to mesenchymal transition (EMT), while its abrogation can reverse EMT [74]. Similar to Celsr3, NCAM interacts with the cytoskeleton and has been implicated in the regulation and promotion of neurite outgrowth [73]. The NCAM1 gene is subject to alternative splicing, leading to three main protein isoforms: NCAM-120, NCAM-140 and NCAM-180, which may or may not incorporate 10 amino acids encoded by the small VASE or π-exon [75]. The incorporation of π into the NCAM protein has been reported to inhibit its neurite outgrowth promoting activity [76]. Lung NENs mainly express the NCAM-140 variant, but may also express NCAM-180 (Table 3C). NCAM can furthermore being posttranslationally modified in several ways, the most important modification

11

being the addition of α 2,8-linked polysialic acid (PSA), which has a negative effect on its mediation of cell-cell interactions. Similar to NCAM, NrCAM is a cell adhesion molecule that shows extensive alternative splicing and supports neurite outgrowth [77]. It is mainly expressed in the nervous system, but also in the adrenal gland and the pancreas, and both loss and gain of NrCAM function may lead to cancer progression in several organs (Table 3C). Less is known on four additional upregulated neuronal genes. The ectodermal-neural cortex 1 protein (ENC1), also known as nuclear restricted protein in brain (NRP/B), is a nuclear matrix factor playing roles in differentiation of neurons and the development of the central nervous system [78]. ENC1 has been implicated in the formation of several cancers (Table 3C). Alpha-internexin (INA) is a neuronal intermediate filament protein also reported to be involved in neurite outgrowth [79] and expressed by several neural and neuroendocrine cancers (Table 3C). The LIM homeobox protein LHX2 has a strong impact on transcriptional regulation during neural differentiation, mostly through binding to enhancers, rather than promoter regions [80]. LHX2 is overexpressed in a number of human cancer types (Table 3C) and stimulates PDGFβ signaling and, as a consequence, tumor growth and metastasis in murine breast cancer cells [81]. Neuronal regeneration related protein homolog (NREP, more commonly known as P311) is a neuronal gene which is expressed in the developing distal lung and specifically at the start and during alveolar development [82]. Again, overexpression of the P311 protein promotes neurite outgrowth [83]. Also stathmin (Section 3.2) is abundantly expressed in postmitotic neurons and its levels are strongly increased during neurite outgrowth and synapse formation [59].

3.4 The chromogranins and the regulated secretory pathway

12

Granins are a family of precursor proteins with multiple cleavage sites that can be proteolytically processed by prohormone- and proprotein convertases [84]. The granin family consists of chromogranins and secretogranins. CgA (CHGA) and chromogranin B (CgB; CHGB) are evolutionary conserved and have a relatively high sequence similarity. Some members are involved in regulation of the secretory pathways. CgA is a neuroendocrine marker which is widely used in the differential diagnosis of lung NENs.[70] It is generally secreted by cells from the diffuse neuroendocrine system. The prohormone is intracellularly processed by proteases present in chromaffin granules, resulting in different peptides including betagranin, catestatin, chromacin, (pro)chromacin, EL35, pancreastatin, parastatin, vasostatin I and II and WE-14 [84-86]. Many different functions have been assigned to these peptides, including roles in cardiovascular processes, carbohydrate metabolism and defense against microbial invasions [85]. CgA levels are furthermore increased in patients with systemic diseases and inflammation, including heart failure and rheumatoid arthritis. Although CgB is upregulated in NENs to a similar level as CgA, this protein is less well studied. Similar to CgA, CgB can be proteolytically processed into different peptides, including BAM-1745, CCB and secretolytin [84]. Importantly, next to being secreted, both CgA [reviewed elsewhere [87]] and CgB [88] are also functionally involved in the regulated secretory pathway. The Carboxypeptidase E (CPE) protein has a dual role. Its soluble form resides in secretory granules and functions as an exopeptidase cleaving prohormones and propeptides into their mature derivatives [89]. The membrane-bound fraction of CPE functions as a receptor to sort prohormones at the lipid rafts in the trans-Golgi network [87]. Its cytoplasmic tail can bind to the microtubulebased motor complex to transport the granules towards the secretion site at the cell membrane. A splice variant of CPE lacking its N-terminal part (CPE-ΔN) has been shown to be translocated to the nucleus where it modulates transcription of the metastasis promoting gene NEDD9 by interacting with histone deacetylase 1/2 [90]. Furthermore, while the full-length CPE inhibits Wnt-signaling, CPE-

13

ΔN induces Wnt-signaling, which is often linked to cancer [91]. Both the full-length and the CPE-ΔN variant have been related to neuroendocrine and other cancers (Table 3D).

Less extensively studied within this context are three additional upregulated genes. KIF1A is a neuron-specific motor protein that transports organelles containing synaptic vesicle proteins including synaptotagmin and the neuroendocrine marker synaptophysin [92]. KIF5C is involved in the transport of mRNA-containing granules towards dendrites [93]. Angiotensinogen (AGT) is another peptide precursor, which is processed by renin to form angiotensin I and, following further enzymatic cleavage, angiotensin II. Similar to CgA [86], this peptide has functions in the cardiovascular system and is implicated in the pathogenesis of hypertension [94]. Although upregulated in lung NENs, angiotensinogen delays angiogenesis and growth of different cancer types (Table 3D).

3.5 Other frequently deregulated genes A few differentially expressed candidate genes could not be related to any of the above discussed categories (described below and/or listed in Table 3E). C5 is a member of the complement family of proteins involved in inflammation and the immune response to infection. Circulating C5 is cleaved into two different peptides, C5a, which functions as a potent chemoattractant and inflammatory mediator, and C5b, which is deposited on cell surfaces [95]. Important insights into the role of C5 in cancer were provided by Markiewski et al., who demonstrated that C5a signaling plays an important role in the inhibition of the immune response to tumor cells [95]. More recently, Corrales and coworkers showed that C5-deposition was higher on lung cancer cells than on normal bronchial epithelial cells, and that C5 is also produced in lung cancer cell lines [96]. In addition, they demonstrated that NSCLC patients had elevated C5a plasma levels, and that C5a is able to induce an immunosuppressive microenvironment. 14

Because most gene expression profiling studies focused on mRNA overexpression, only few underrepresentated genes were reported. Caspase-4 was the only one that was reported in multiple studies. Caspase-4 belongs to the caspase-1/interleukin-1β converting enzyme (ICE) gene cluster located at chromosome 11q22.3, which deletion has been reported to be associated with poor survival in lung carcinoids [97]. Caspase-4 is probably not directly involved in the intrinsic or extrinsic apoptosis pathways, but plays an important albeit not essential role in ER-stress initiated apoptosis [98]. The colony-stimulating factor 1 receptor gene (CSF1R) is a proto-oncogene located in close proximity to the PDGFRB gene. Both genes encode tyrosine kinase receptors and are thought to be the result of a duplication event [99]. CSF1R normally functions in the differentiation of macrophages [100], and is implicated in tumor associated macrophages [101]. Although PDGFRβ is expressed in the majority of lung NENs (most frequently in LCNECs) [7], the presence of CSF1R protein remains to be studied in lung cancer. Nonetheless, CSF1R as well as PDGFRβ have been reported to be sensitive to the tyrosine kinase inhibitor imatinib, and specific CSF1R-inhibitors have been developed (Table 3E), which warrants a more extensive analysis of CSF1R expression in lung NENs.

4. Concluding remarks

For this review, a systematic analysis of gene expression profiling studies of lung NENs was performed, which resulted in a selection of consistently differentially expressed candidate genes. Although the number of genes that was found to be differentially expressed in more than one study was relatively low, the 33 genes that were found in multiple studies should be clearly associated with lung NENs. This low number may be explained by the use of different reference tissues (Table 1). 15

While PNECs in the bronchial epithelium are assumed to be the cells of origin for (the majority of) lung NENs, their scarcity makes them unsuitable for use as a normal control [7]. Therefore, most studies used normal lung tissue as a reference. As a result, some of the genes differentially expressed in NENs as compared to normal lung tissue, may also be identified in PNECs and are thereby not tumor-specific but neuroendocrine-specific. The specificity and sensitivity of each individual gene as marker for pulmonary NENs should therefore being separately validated in independent tumor series. Although our approach yielded a number of interesting candidate genes, it has also some limitations. Since their number is limited we have included as many lung NEN gene expression studies as possible, although different microarray platforms [8] and reference tissues were used (Table 1), and different lung NEN subtypes and numbers of cases were included. For these reasons, we choose to perform a qualitative systematic review of these studies, rather than a meta-analysis. A metaanalysis of expression profiling studies in lung tumors has been previously provided by Amelung et al., but this study focused mainly on NSCLC [102]. Importantly, due to the heterogeneity of study design and analysis of differentially expressed genes, our gene selection procedure may have biased the results. On the other hand, by considering only genes reported by multiple studies, the likeliness that they are indeed important increases, which is supported by the fact that many genes have already been described in the context of (neuroendocrine) cancer (Table 3).

The majority of genes identified in this review could be assigned to a few main categories. First, a number of genes encode (neuro)endocrine-specific proteins and/or factors involved in neuronal development and neurite outgrowth. The transcription factor ASH1 (Figure 1) seems to be central in a network of neuroendocrine factors involved in the specific nature of lung NENs. Other factors are involved in cell cycle progression, including multiple members of the mitotic SAC (Figure 2). Another important category consists of proteins involved in the regulated secretory pathway, of which the chromogranins are most extensively studied, and used as major tools in the detection of lung NENs. 16

Among these categories a considerable number of candidates have been described in the literature to possess proto-oncogenic properties. To select the most potent candidates for further study we examined which factors can be used for therapeutic approaches (Table 3), and compared our findings to some of the recent next-generation sequencing (NGS) integrated genome-wide approaches [4-6]. A number of genes described in this review were found to be mutated in these studies. Most notably, 6 mutations of CSF1R were identified in the study by Peifer and colleagues, as well as 2 mutations of DCX, FOXG1 and NRCAM. DCX was moreover found twice to be mutated in the study by Rudin, as was CENPF (also identified by Peifer). Since a number of CSF1R inhibitors are currently being developed (Table 3E), this candidate may be the first choice for further study. Amongst other therapeutic targets are CDC20, also once found to be mutated in NGS [4], as well as TOP2A (Table 3B). NCAM is another putative therapeutic target and a number of immunotherapy strategies have been developed to target its expression (reviewed elsewhere [103]). Recently, an NCAM antibody-drug conjugate has been successfully used to target SCLC xenografts [104]. Also ASH1 has been suggested to be a therapeutic target for lung NENs (Table 3A). Since it is a master regulator in the lung neuroendocrine lineage [7] ASH1 and related factors are important to pursue in further studies. Rudin and colleagues showed that SOX2 is frequently amplified in SCLC [5]. Together with FoxG1, Sox2 is capable of inducing neural precursor cells to become astrocytic and neuronal cells [105]. Also Lhx2 may function in combined use with Sox2 in the formation of induced neural precursor cells [105]. Of the genes described in this review, only STMN1 was found to be mutated in lung carcinoid NGS [6].

Although most relevant for diagnostic and therapeutic approaches, the fact that the majority of expression profiling studies yielded upregulated genes poses a limitation for the understanding of lung NEN biology. Future studies will probably reveal more downregulated genes in lung NENs. For 17

example, we have recently identified CD44 and OTP, which are downregulated in aggressive pulmonary carcinoids, as strong prognostic markers [22]. For several of the identified factors, functions in lung NENs still need to be established. Their further assessment within this context will provide important insights into the carcinogenesis and therapeutic options of this understudied subgroup of lung tumors. We hope that the factors discussed in this review will stimulate researchers to test their biological and clinical significance for lung NENs in upcoming studies.

18

Acknowledgements This work was partially supported by a grant from the Jan Dekkerstichting & Ludgardine Bouwmanstichting (nr. 2010_010). This funding source had no involvement in the study design, collection, analysis and interpretation of data, or in the writing of the text.

Conflict of Interest statement

D.R.A. Swarts declared no conflicts of interests E.-J. M. Speel has honoraria from Speakers Bureau of Pfizer and Lilly and is a consultant/advisory board member of Pfizer. F.C.S. Ramaekers declared no conflicts of interests.

19

Reference List

[1] Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin 2012;62:10-29.

[2] Travis WD. Lung tumours with neuroendocrine differentiation. Eur J Cancer 2009;45 Suppl 1:251-66.

[3] Kalari S, Jung M, Kernstine K H, Takahashi T, Pfeifer G P. The DNA methylation landscape of small cell lung cancer suggests a differentiation defect of neuroendocrine cells. Oncogene 2013;32:3559-68.

[4] Peifer M, Fernandez-Cuesta L, Sos M L, George J, Seidel D, Kasper L H et al. Integrative genome analyses identify key somatic driver mutations of small-cell lung cancer. Nat Genet 2012;44:1104-10.

[5] Rudin CM, Durinck S, Stawiski E W, Poirier J T, Modrusan Z, Shames D S et al. Comprehensive genomic analysis identifies SOX2 as a frequently amplified gene in small-cell lung cancer. Nat Genet 2012;44:1111-6.

[6] Fernandez-Cuesta L, Peifer M, Lu X, Sun R, Ozretic L, Seidel D et al. Frequent mutations in chromatin-remodelling genes in pulmonary carcinoids. Nat Commun 2014;5:3518.

[7] Swarts DR, Ramaekers F C, Speel E J. Molecular and cellular biology of neuroendocrine lung tumors: Evidence for separate biological entities. Biochim Biophys Acta 2012;1826:255-71.

20

[8] Mah N, Thelin A, Lu T, Nikolaus S, Kuhbacher T, Gurbuz Y et al. A comparison of oligonucleotide and cDNA-based microarray systems. Physiol Genomics 2004;16:361-70.

[9] Anbazhagan R, Tihan T, Bornman D M, Johnston J C, Saltz J H, Weigering A et al. Classification of small cell lung cancer and pulmonary carcinoid by gene expression profiles. Cancer Res 1999;59:5119-22.

[10] Bhattacharjee A, Richards W G, Staunton J, Li C, Monti S, Vasa P et al. Classification of human lung carcinomas by mRNA expression profiling reveals distinct adenocarcinoma subclasses. Proc Natl Acad Sci U S A 2001;98:13790-5.

[11] Garber ME, Troyanskaya O G, Schluens K, Petersen S, Thaesler Z, Pacyna-Gengelbach M et al. Diversity of gene expression in adenocarcinoma of the lung. Proc Natl Acad Sci U S A 2001;98:13784-9.

[12] He P, Varticovski L, Bowman E D, Fukuoka J, Welsh J A, Miura K et al. Identification of carboxypeptidase E and gamma-glutamyl hydrolase as biomarkers for pulmonary neuroendocrine tumors by cDNA microarray. Hum Pathol 2004;35:1196-209.

[13] Jones MH, Virtanen C, Honjoh D, Miyoshi T, Satoh Y, Okumura S et al. Two prognostically significant subtypes of high-grade lung neuroendocrine tumours independent of small-cell and large-cell neuroendocrine carcinomas identified by gene expression profiles. Lancet 2004;363:775-81.

21

[14] Rohrbeck A, Neukirchen J, Rosskopf M, Pardillos G G, Geddert H, Schwalen A et al. Gene expression profiling for molecular distinction and characterization of laser captured primary lung cancers. J Transl Med 2008;6:69.

[15] Taniwaki M, Daigo Y, Ishikawa N, Takano A, Tsunoda T, Yasui W et al. Gene expression profiles of small-cell lung cancers: molecular signatures of lung cancer. Int J Oncol 2006;29:567-75.

[16] Virtanen C, Ishikawa Y, Honjoh D, Kimura M, Shimane M, Miyoshi T et al. Integrated classification of lung tumors and cell lines by expression profiling. Proc Natl Acad Sci U S A 2002;99:12357-62.

[17] Wikman H, Seppanen J K, Sarhadi V K, Kettunen E, Salmenkivi K, Kuosma E et al. Caveolins as tumour markers in lung cancer detected by combined use of cDNA and tissue microarrays. J Pathol 2004;203:584-93.

[18] Takeuchi T, Tomida S, Yatabe Y, Kosaka T, Osada H, Yanagisawa K et al. Expression profile-defined classification of lung adenocarcinoma shows close relationship with underlying major genetic changes and clinicopathologic behaviors. J Clin Oncol 2006;24:1679-88.

[19] Swarts DR, van N L, Henfling M E, Eijkenboom I, Eijk P P, Van Velthuysen M L et al. An exploration of pathways involved in lung carcinoid progression using gene expression profiling. Carcinogenesis 2013;34:2726-37.

22

[20] Bari MF, Brown H, Nicholson A G, Kerr K M, Gosney J R, Wallace W A et al. BAI3, CDX2 and VIL1: a panel of three antibodies to distinguish small cell from large cell neuroendocrine lung carcinomas. Histopathology 2013

[21] Toffalorio F, Belloni E, Barberis M, Bucci G, Tizzoni L, Pruneri G et al. Gene expression profiling reveals GC and CEACAM1 as new tools in the diagnosis of lung carcinoids. Br J Cancer 2014;110:1244-9.

[22] Swarts DR, Henfling M E, Van Neste L, van Suylen R J, Dingemans A M, Dinjens W N et al. CD44 and OTP are strong prognostic markers for pulmonary carcinoids. Clin Cancer Res 2013;19:2197-207.

[23] Pedersen N, Mortensen S, Sorensen S B, Pedersen M W, Rieneck K, Bovin L F et al. Transcriptional gene expression profiling of small cell lung cancer cells. Cancer Res 2003;63:1943-53.

[24] Ball DW, Azzoli C G, Baylin S B, Chi D, Dou S, Donis-Keller H et al. Identification of a human achaete-scute homolog highly expressed in neuroendocrine tumors. Proc Natl Acad Sci U S A 1993;90:5648-52.

[25] La Rosa S, Marando A, Gatti G, Rapa I, Volante M, Papotti M et al. Achaete-scute homolog 1 as a marker of poorly differentiated neuroendocrine carcinomas of different sites: a validation study using immunohistochemistry and quantitative real-time polymerase chain reaction on 335 cases. Hum Pathol 2013;44:1391-9.

23

[26] Borges M, Linnoila R I, van d, V, Chen H, Nelkin B D, Mabry M et al. An achaete-scute homologue essential for neuroendocrine differentiation in the lung. Nature 1997;386:852-5.

[27] Osada H, Tomida S, Yatabe Y, Tatematsu Y, Takeuchi T, Murakami H et al. Roles of achaete-scute homologue 1 in DKK1 and E-cadherin repression and neuroendocrine differentiation in lung cancer. Cancer Res 2008;68:1647-55.

[28] Hu Y, Wang T, Stormo G D, Gordon J I. RNA interference of achaete-scute homolog 1 in mouse prostate neuroendocrine cells reveals its gene targets and DNA binding sites. Proc Natl Acad Sci U S A 2004;101:5559-64.

[29] Jacob J, Storm R, Castro D S, Milton C, Pla P, Guillemot F et al. Insm1 (IA-1) is an essential component of the regulatory network that specifies monoaminergic neuronal phenotypes in the vertebrate hindbrain. Development 2009;136:2477-85.

[30] Ball DW. Achaete-scute homolog-1 and Notch in lung neuroendocrine development and cancer. Cancer Lett 2004;204:159-69.

[31] Ito T, Udaka N, Yazawa T, Okudela K, Hayashi H, Sudo T et al. Basic helix-loop-helix transcription factors regulate the neuroendocrine differentiation of fetal mouse pulmonary epithelium. Development 2000;127:3913-21.

[32] Bae S, Bessho Y, Hojo M, Kageyama R. The bHLH gene Hes6, an inhibitor of Hes1, promotes neuronal differentiation. Development 2000;127:2933-43.

24

[33] Lan MS, Breslin M B. Structure, expression, and biological function of INSM1 transcription factor in neuroendocrine differentiation. FASEB J 2009;23:2024-33.

[34] Gierl MS, Karoulias N, Wende H, Strehle M, Birchmeier C. The zinc-finger factor Insm1 (IA-1) is essential for the development of pancreatic beta cells and intestinal endocrine cells. Genes Dev 2006;20:2465-78.

[35] Ahlgren U, Pfaff S L, Jessell T M, Edlund T, Edlund H. Independent requirement for ISL1 in formation of pancreatic mesenchyme and islet cells. Nature 1997;385:257-60.

[36] Zhang H, Wang W P, Guo T, Yang J C, Chen P, Ma K T et al. The LIM-homeodomain protein ISL1 activates insulin gene promoter directly through synergy with BETA2. J Mol Biol 2009;392:566-77.

[37] Cai CL, Liang X, Shi Y, Chu P H, Pfaff S L, Chen J et al. Isl1 identifies a cardiac progenitor population that proliferates prior to differentiation and contributes a majority of cells to the heart. Dev Cell 2003;5:877-89.

[38] Schmitt AM, Riniker F, Anlauf M, Schmid S, Soltermann A, Moch H et al. Islet 1 (Isl1) expression is a reliable marker for pancreatic endocrine tumors and their metastases. Am J Surg Pathol 2008;32:420-5.

[39] Fang G, Yu H, Kirschner M W. The checkpoint protein MAD2 and the mitotic regulator CDC20 form a ternary complex with the anaphase-promoting complex to control anaphase initiation. Genes Dev 1998;12:1871-83.

25

[40] Sudakin V, Chan G K, Yen T J. Checkpoint inhibition of the APC/C in HeLa cells is mediated by a complex of BUBR1, BUB3, CDC20, and MAD2. J Cell Biol 2001;154:925-36.

[41] Tipton AR, Wang K, Link L, Bellizzi J J, Huang H, Yen T et al. BUBR1 and closed MAD2 (CMAD2) interact directly to assemble a functional mitotic checkpoint complex. J Biol Chem 2011;286:21173-9.

[42] Abal M, Obrador-Hevia A, Janssen K P, Casadome L, Menendez M, Carpentier S et al. APC inactivation associates with abnormal mitosis completion and concomitant BUB1B/MAD2L1 up-regulation. Gastroenterology 2007;132:2448-58.

[43] Musacchio A, Salmon E D. The spindle-assembly checkpoint in space and time. Nat Rev Mol Cell Biol 2007;8:379-93.

[44] Gurden MD, Holland A J, van Z W, Tighe A, Vergnolle M A, Andres D A et al. Cdc20 is required for the post-anaphase, KEN-dependent degradation of centromere protein F. J Cell Sci 2010;123:321-30.

[45] Johnson VL, Scott M I, Holt S V, Hussein D, Taylor S S. Bub1 is required for kinetochore localization of BubR1, Cenp-E, Cenp-F and Mad2, and chromosome congression. J Cell Sci 2004;117:1577-89.

[46] Kops GJ, Weaver B A, Cleveland D W. On the road to cancer: aneuploidy and the mitotic checkpoint. Nat Rev Cancer 2005;5:773-85.

26

[47] Hernando E, Nahle Z, Juan G, az-Rodriguez E, Alaminos M, Hemann M et al. Rb inactivation promotes genomic instability by uncoupling cell cycle progression from mitotic control. Nature 2004;430:797-802.

[48] Seoane J, Le H V, Shen L, Anderson S A, Massague J. Integration of Smad and forkhead pathways in the control of neuroepithelial and glioblastoma cell proliferation. Cell 2004;117:211-23.

[49] Chan DW, Liu V W, To R M, Chiu P M, Lee W Y, Yao K M et al. Overexpression of FOXG1 contributes to TGF-beta resistance through inhibition of p21WAF1/CIP1 expression in ovarian cancer. Br J Cancer 2009;101:1433-43.

[50] Abbas T, Dutta A. p21 in cancer: intricate networks and multiple activities. Nat Rev Cancer 2009;9:400-14.

[51] Swarts DR, Scarpa A, Corbo V, van C W, van E M, Gatti G et al. MEN1 Gene Mutation and Reduced Expression Are Associated with Poor Prognosis in Pulmonary Carcinoids. J Clin Endocrinol Metab 2013

[52] Kottemann MC, Bale A E. Characterization of DNA damage-dependent cell cycle checkpoints in a menin-deficient model. DNA Repair (Amst ) 2009;8:944-52.

[53] Zirbes TK, Lorenzen J, Baldus S E, Moenig S P, Wolters U, Ottlik A et al. Apoptosis and expression of bcl-2 protein are inverse factors influencing tumour cell turnover in primary carcinoid tumours of the lung. Histopathology 1998;33:123-8.

27

[54] Stefanaki K, Rontogiannis D, Vamvouka C, Bolioti S, Chaniotis V, Sotsiou F et al. Immunohistochemical detection of bcl2, p53, mdm2 and p21/waf1 proteins in small-cell lung carcinomas. Anticancer Res 1998;18:1167-73.

[55] Shima N, Alcaraz A, Liachko I, Buske T R, Andrews C A, Munroe R J et al. A viable allele of Mcm4 causes chromosome instability and mammary adenocarcinomas in mice. Nat Genet 2007;39:93-8.

[56] Nitiss JL. DNA topoisomerase II and its growing repertoire of biological functions. Nat Rev Cancer 2009;9:327-37.

[57] Coelho PA, Queiroz-Machado J, Carmo A M, Moutinho-Pereira S, Maiato H, Sunkel C E. Dual role of topoisomerase II in centromere resolution and aurora B activity. PLoS Biol 2008;6:e207.

[58] Belletti B, Baldassarre G. Stathmin: a protein with many tasks. New biomarker and potential target in cancer. Expert Opin Ther Targets 2011;15:1249-66.

[59] Rubin CI, Atweh G F. The role of stathmin in the regulation of the cell cycle. J Cell Biochem 2004;93:242-50.

[60] Zhu C, Jiang W. Cell cycle-dependent translocation of PRC1 on the spindle by Kif4 is essential for midzone formation and cytokinesis. Proc Natl Acad Sci U S A 2005;102:343-8.

28

[61] Kurasawa Y, Earnshaw W C, Mochizuki Y, Dohmae N, Todokoro K. Essential roles of KIF4 and its binding partner PRC1 in organized central spindle midzone formation. EMBO J 2004;23:3237-48.

[62] Subramanian R, Ti S C, Tan L, Darst S A, Kapoor T M. Marking and measuring single microtubules by PRC1 and kinesin-4. Cell 2013;154:377-90.

[63] Shimo A, Nishidate T, Ohta T, Fukuda M, Nakamura Y, Katagiri T. Elevated expression of protein regulator of cytokinesis 1, involved in the growth of breast cancer cells. Cancer Sci 2007;98:174-81.

[64] Zhou L, Bar I, Achouri Y, Campbell K, De B O, Hebert J M et al. Early forebrain wiring: genetic dissection using conditional Celsr3 mutant mice. Science 2008;320:946-9.

[65] Shima Y, Kawaguchi S Y, Kosaka K, Nakayama M, Hoshino M, Nabeshima Y et al. Opposing roles in neurite growth control by two seven-pass transmembrane cadherins. Nat Neurosci 2007;10:963-9.

[66] Shmueli O, Gdalyahu A, Sorokina K, Nevo E, Avivi A, Reiner O. DCX in PC12 cells: CREBmediated transcription and neurite outgrowth. Hum Mol Genet 2001;10:1061-70.

[67] Tanaka T, Serneo F F, Tseng H C, Kulkarni A B, Tsai L H, Gleeson J G. Cdk5 phosphorylation of doublecortin ser297 regulates its effect on neuronal migration. Neuron 2004;41:215-27.

29

[68] Demelash A, Rudrabhatla P, Pant H C, Wang X, Amin N D, McWhite C D et al. Achaetescute homologue-1 (ASH1) stimulates migration of lung cancer cells through Cdk5/p35 pathway. Mol Biol Cell 2012;23:2856-66.

[69] Daou MC, Smith T W, Litofsky N S, Hsieh C C, Ross A H. Doublecortin is preferentially expressed in invasive human brain tumors. Acta Neuropathol 2005;110:472-80.

[70] W.D.Travis, E.Brambilla, H.K.Müller-Hermelink et al., Tumours of the Lung, Pleura, Thymus and Heart, edited by W.D.Travis, E.Brambilla, H.K.Müller-Hermelink, and C.C.Harris, IARC Press, Lyon, 2004.

[71] Lantuejoul S, Moro D, Michalides R J, Brambilla C, Brambilla E. Neural cell adhesion molecules (NCAM) and NCAM-PSA expression in neuroendocrine lung tumors. Am J Surg Pathol 1998;22:1267-76.

[72] Gattenlohner S, Stuhmer T, Leich E, Reinhard M, Etschmann B, Volker H U et al. Specific detection of CD56 (NCAM) isoforms for the identification of aggressive malignant neoplasms with progressive development. Am J Pathol 2009;174:1160-71.

[73] Bodrikov V, Sytnyk V, Leshchyns'ka I, den H J, Schachner M. NCAM induces CaMKIIalphamediated RPTPalpha phosphorylation to enhance its catalytic activity and neurite outgrowth. J Cell Biol 2008;182:1185-200.

[74] Lehembre F, Yilmaz M, Wicki A, Schomber T, Strittmatter K, Ziegler D et al. NCAMinduced focal adhesion assembly: a functional switch upon loss of E-cadherin. EMBO J 2008;27:2603-15.

30

[75] Moolenaar CE, Pieneman C, Walsh F S, Mooi W J, Michalides R J. Alternative splicing of neural-cell-adhesion molecule mRNA in human small-cell lung-cancer cell line H69. Int J Cancer 1992;51:238-43.

[76] Doherty P, Moolenaar C E, Ashton S V, Michalides R J, Walsh F S. The VASE exon downregulates the neurite growth-promoting activity of NCAM 140. Nature 1992;356:791-3.

[77] Sakurai T. The role of NrCAM in neural development and disorders--beyond a simple glue in the brain. Mol Cell Neurosci 2012;49:351-63.

[78] Kim TA, Lim J, Ota S, Raja S, Rogers R, Rivnay B et al. NRP/B, a novel nuclear matrix protein, associates with p110(RB) and is involved in neuronal differentiation. J Cell Biol 1998;141:553-66.

[79] Yuan A, Rao M V, Sasaki T, Chen Y, Kumar A, Veeranna et al. Alpha-internexin is structurally and functionally associated with the neurofilament triplet proteins in the mature CNS. J Neurosci 2006;26:10006-19.

[80] Hou PS, Chuang C Y, Kao C F, Chou S J, Stone L, Ho H N et al. LHX2 regulates the neural differentiation of human embryonic stem cells via transcriptional modulation of PAX6 and CER1. Nucleic Acids Res 2013;41:7753-70.

[81] Kuzmanov A, Hopfer U, Marti P, Meyer-Schaller N, Yilmaz M, Christofori G. LIMhomeobox gene 2 promotes tumor growth and metastasis by inducing autocrine and paracrine PDGF-B signaling. Mol Oncol 2014;8:401-16.

31

[82] Zhao L, Leung J K, Yamamoto H, Goswami S, Kheradmand F, Vu T H. Identification of P311 as a potential gene regulating alveolar generation. Am J Respir Cell Mol Biol 2006;35:48-54.

[83] Fujitani M, Yamagishi S, Che Y H, Hata K, Kubo T, Ino H et al. P311 accelerates nerve regeneration of the axotomized facial nerve. J Neurochem 2004;91:737-44.

[84] Zhao E, Zhang D, Basak A, Trudeau V L. New insights into granin-derived peptides: evolution and endocrine roles. Gen Comp Endocrinol 2009;164:161-74.

[85] Helle KB, Corti A, Metz-Boutigue M H, Tota B. The endocrine role for chromogranin A: a prohormone for peptides with regulatory properties. Cell Mol Life Sci 2007;64:2863-86.

[86] Modlin IM, Gustafsson B I, Moss S F, Pavel M, Tsolakis A V, Kidd M. Chromogranin A-biological function and clinical utility in neuro endocrine tumor disease. Ann Surg Oncol 2010;17:2427-43.

[87] Park JJ, Koshimizu H, Loh Y P. Biogenesis and transport of secretory granules to release site in neuroendocrine cells. J Mol Neurosci 2009;37:151-9.

[88] Natori S, Huttner W B. Chromogranin B (secretogranin I) promotes sorting to the regulated secretory pathway of processing intermediates derived from a peptide hormone precursor. Proc Natl Acad Sci U S A 1996;93:4431-6.

32

[89] Murthy SR, Pacak K, Loh Y P. Carboxypeptidase E: elevated expression correlated with tumor growth and metastasis in pheochromocytomas and other cancers. Cell Mol Neurobiol 2010;30:1377-81.

[90] Lee TK, Murthy S R, Cawley N X, Dhanvantari S, Hewitt S M, Lou H et al. An N-terminal truncated carboxypeptidase E splice isoform induces tumor growth and is a biomarker for predicting future metastasis in human cancers. J Clin Invest 2011;121:880-92.

[91] Skalka N, Caspi M, Caspi E, Loh Y P, Rosin-Arbesfeld R. Carboxypeptidase E: a negative regulator of the canonical Wnt signaling pathway. Oncogene 2013;32:2836-47.

[92] Okada Y, Yamazaki H, Sekine-Aizawa Y, Hirokawa N. The neuron-specific kinesin superfamily protein KIF1A is a unique monomeric motor for anterograde axonal transport of synaptic vesicle precursors. Cell 1995;81:769-80.

[93] Kanai Y, Dohmae N, Hirokawa N. Kinesin transports RNA: isolation and characterization of an RNA-transporting granule. Neuron 2004;43:513-25.

[94] Sadoshima J, Xu Y, Slayter H S, Izumo S. Autocrine release of angiotensin II mediates stretch-induced hypertrophy of cardiac myocytes in vitro. Cell 1993;75:977-84.

[95] Markiewski MM, DeAngelis R A, Benencia F, Ricklin-Lichtsteiner S K, Koutoulaki A, Gerard C et al. Modulation of the antitumor immune response by complement. Nat Immunol 2008;9:1225-35.

33

[96] Corrales L, Ajona D, Rafail S, Lasarte J J, Riezu-Boj J I, Lambris J D et al. Anaphylatoxin c5a creates a favorable microenvironment for lung cancer progression. J Immunol 2012;189:4674-83.

[97] Swarts DR, Claessen S M, Jonkers Y M, van Suylen R J, Dingemans A M, de Herder W W et al. Deletions of 11q22.3-q25 are associated with atypical lung carcinoids and poor clinical outcome. Am J Pathol 2011;179:1129-37.

[98] Hitomi J, Katayama T, Eguchi Y, Kudo T, Taniguchi M, Koyama Y et al. Involvement of caspase-4 in endoplasmic reticulum stress-induced apoptosis and Abeta-induced cell death. J Cell Biol 2004;165:347-56.

[99] Roberts WM, Look A T, Roussel M F, Sherr C J. Tandem linkage of human CSF-1 receptor (c-fms) and PDGF receptor genes. Cell 1988;55:655-61.

[100] Bonifer C, Hoogenkamp M, Krysinska H, Tagoh H. How transcription factors program chromatin--lessons from studies of the regulation of myeloid-specific genes. Semin Immunol 2008;20:257-63.

[101] Escamilla J, Schokrpur S, Liu C, Priceman S J, Moughon D, Jiang Z et al. CSF1 Receptor Targeting in Prostate Cancer Reverses Macrophage-Mediated Resistance to Androgen Blockade Therapy. Cancer Res 2015;75:950-62.

[102] Amelung JT, Buhrens R, Beshay M, Reymond M A. Key genes in lung cancer translational research: a meta-analysis. Pathobiology 2010;77:53-63.

34

[103] Jensen M, Berthold F. Targeting the neural cell adhesion molecule in cancer. Cancer Lett 2007;258:9-21.

[104] Whiteman KR, Johnson H A, Mayo M F, Audette C A, Carrigan C N, LaBelle A et al. Lorvotuzumab mertansine, a CD56-targeting antibody-drug conjugate with potent antitumor activity against small cell lung cancer in human xenograft models. MAbs 2014;6:556-66.

[105] Lujan E, Chanda S, Ahlenius H, Sudhof T C, Wernig M. Direct conversion of mouse fibroblasts to self-renewing, tripotent neural precursor cells. Proc Natl Acad Sci U S A 2012;109:2527-32.

[106] Hiroshima K, Iyoda A, Shida T, Shibuya K, Iizasa T, Kishi H et al. Distinction of pulmonary large cell neuroendocrine carcinoma from small cell lung carcinoma: a morphological, immunohistochemical, and molecular analysis. Mod Pathol 2006;19:1358-68.

[107] Peng SY, Wang W P, Meng J, Li T, Zhang H, Li Y M et al. ISL1 physically interacts with BETA2 to promote insulin gene transcriptional synergy in non-beta cells. Biochim Biophys Acta 2005;1731:154-9.

[108] Breslin MB, Zhu M, Notkins A L, Lan M S. Neuroendocrine differentiation factor, IA-1, is a transcriptional repressor and contains a specific DNA-binding domain: identification of consensus IA-1 binding sequence. Nucleic Acids Res 2002;30:1038-45.

35

[109] Breslin MB, Zhu M, Lan M S. NeuroD1/E47 regulates the E-box element of a novel zinc finger transcription factor, IA-1, in developing nervous system. J Biol Chem 2003;278:38991-7.

[110] Keller DM, McWeeney S, Arsenlis A, Drouin J, Wright C V, Wang H et al. Characterization of pancreatic transcription factor Pdx-1 binding sites using promoter microarray and serial analysis of chromatin occupancy. J Biol Chem 2007;282:32084-92.

[111] Nishikawa E, Osada H, Okazaki Y, Arima C, Tomida S, Tatematsu Y et al. miR-375 is activated by ASH1 and inhibits YAP1 in a lineage-dependent manner in lung cancer. Cancer Res 2011;71:6165-73.

[112] Poy MN, Eliasson L, Krutzfeldt J, Kuwajima S, Ma X, Macdonald P E et al. A pancreatic islet-specific microRNA regulates insulin secretion. Nature 2004;432:226-30.

[113] Hao A, Novotny-Diermayr V, Bian W, Lin B, Lim C P, Jing N et al. The LIM/homeodomain protein Islet1 recruits Janus tyrosine kinases and signal transducer and activator of transcription 3 and stimulates their activities. Mol Biol Cell 2005;16:1569-83.

[114] Farkas LM, Haffner C, Giger T, Khaitovich P, Nowick K, Birchmeier C et al. Insulinomaassociated 1 has a panneurogenic role and promotes the generation and expansion of basal progenitors in the developing mouse neocortex. Neuron 2008;60:40-55.

[115] Laoukili J, Kooistra M R, Bras A, Kauw J, Kerkhoven R M, Morrison A et al. FoxM1 is required for execution of the mitotic programme and chromosome stability. Nat Cell Biol 2005;7:126-36. 36

[116] Carr JR, Park H J, Wang Z, Kiefer M M, Raychaudhuri P. FoxM1 mediates resistance to herceptin and paclitaxel. Cancer Res 2010;70:5054-63.

37

Figure Legends

Figure 1 – Proteins involved in neuroendocrine differentiation pathways for which transcriptional activity was found to be upregulated in pulmonary neuroendocrine neoplasms. The ASH1 transcription factor seems to be central in determining the neuroendocrine phenotype, and is likely to play a major role in the development of neuroendocrine lung cancers. INSM1, as well as NeuroD, for which lower expression levels have been associated with poor prognosis in high-grade neuroendocrine lung carcinomas [106] (therefore shown in red), are believed to function downstream of ASH1. In its turn, INSM1 inhibits its own transcription as well as the transcription of insulin (INS) and NeuroD, while ISL1 stimulates insulin transcription through a physical interaction with NeuroD [36,107]. Although NEUROD is inhibited by INSM1 [108], a complex of NeuroD and TCF3 induces INSM1 transcription, thereby creating a negative feedback loop [109]. NeuroD also upregulates CHGA and NCAM1-140 expression. In addition, miR-375, a microRNA that inhibits SCLC growth through the repression of YAP1 translation, is a target of both ASH1 and NeuroD [110,111], and is an inhibitor of insulin secretion, possibly through inhibiting myotrophin [112]. The CgA peptides pancreastatin and betagranin inhibit insulin secretion as well [86]. Furthermore, Isl1 resides in a complex with Jak1 and Stat3 and stimulates Stat3 transcriptional activity through the phosphorylation and kinase activity of Jak1 [113]. Since the Stat3 gene is downregulated in Insm1 knockout mice [114], it is likely that Insm1 also stimulates its expression. Additional interactions shown in this Figure are discussed in the text. Factors listed in Table 2 are shown in green, and other factors in blue. Proven interactions are indicated by solid arrows, whereas indirect interactions or not sufficiently validated interactions are depicted by dashed arrows. The black circles indicate complex formation between NeuroD and TCF3 on the one hand and between NeuroD and ISL1 on the other hand.

38

Figure 2 – Proteins involved in the spindle assembly checkpoint for which transcriptional activity was found to be upregulated in neuroendocrine lung neoplasms. When CDC20 is bound by BUB1R, BUB3 and MAD2 to form the mitotic checkpoint complex (MCC), its stimulating activity towards the anaphase-promoting complex/cyclosome (APC/C) is blocked. Inhibition of the APC/C will prevent the cell to enter the anaphase until all chromosomes are correctly attached to the mitotic spindle. The APC/C can polyubiquitylate and thus degrade several downstream targets, of which Cyclin B1 and CENP-F are depicted. Both cyclin B1 and CENP-F can be activated by the Forkhead transcription factor FoxM1 [115]. Also stathmin has been described to be a direct transcriptional target of FoxM1 [116]. In green the proteins or protein complexes are indicated for which the transcriptional levels were described to be upregulated in at least two studies (Table 2), while other proteins or protein complexes are shown in blue.

Highlights

Published gene expression studies on lung neuroendocrine neoplasms re-evaluated Focus on important factors reported by multiple studies Many genes involved in neuroendocrine development and cell cycle progression Potential roles of these factors in diagnosis and as therapeutic targets proposed

· · · · Tables

Table 1 - Gene expression profiling studies performed on neuroendocrine lung tumors

Yea Number of r carcinoids

Number of LCNECs

Number of SCLCs

199 9

2

0

2

200 1

20

0

6

39

Referen ce Platform materia l 18K cDNA Normal microarray lung s (Genome tissue Systems) and the other NE lung cancer subtype s 63K U95A Normal oligonucleo lung

Author/referenc e Anbazhagan et al.[9]

Bhattacharjee et al.[10]

200 1

0

0

5

200 2

3

0

9

200 4

11

2

3

200 a 4

13

8

17

200 4

0

2

7

200 b 6

0

2

0

200 6

0

0

15

200 8

0

0

9

201 3

10

0

0

201 3

0

4

4

tide probe tissue arrays and (Affymetrix) other lung cancers 23K cDNA Normal microarray lung s tissue and other lung cancers 8K Normal microarray lung s tissue and other lung cancers 9K cDNA The microarray other s NE lung cancer subtype s 40K cDNA Normal microarray lung s tissue and other lung cancers 1K cDNA Normal microarray lung s (Human tissue Cancer and Gene Filter other 1.2, lung Clontech) cancers 22K Pool of oligonucleo cell tide lines microarray s (Agilent) 32K cDNA Normal microarray lung s tissue and NSCLC 9K h6Normal Focus lung GeneChips tissue (Affymetrix) and NSCLC 44K v2 Pool of microarray cell s (Agilent) lines 60K single Normal channel lung arrays tissue

40

Garber et al.[11]

Virtanen et al.[16]

He et al.[12]

Jones et al.[13]

Wikman et al.[17]

Takeuchi et al.[18]

Taniwaki et al.[15]

Rohrbeck et al.[14]

Swarts et al.[19]

Bari et al.[20]

(Agilent) c

Toffalorio et 55K HGNone al.[21] U133 Plus 2.0 oligonucleo tide microarray s (Affymetrix) Abbreviations used: LCNEC, large cell neuroendocrine carcinoma; NE, neuroendocrine; NSCLC, nonsmall cell lung cancer; SCLC, small cell lung cancer a Genes taken for this review were described in the text body. b No list of genes for the two LCNECs was reported. Therefore, this publication was not further analyzed for this review. c Only a comparison was provided between typical and atypical carcinoids. Therefore, this publication was not further analyzed for this review. 201 4

13

0

0

41

Table 2 - Genes found to be altered in pulmonary neuroendocrine neoplasms as compared to normal tissue or non-small cell lung cancers by at least two different genome-wide expression profiling studies Gene abbrevi a-tion

Gene name (NCBI)

Nu Chromos mb omal er Location of stu die s

Carc LC S inoi NE C ds C L C

AGT

Angiotensinogen (SERPINA8) Achaete-scute homolog 1 Budding uninhibited by benzimidazole 1 homolog beta Complement component 5 Caspase-4, apoptosisrelated cysteine peptidase Cyclin B1

2

1q42.2

+

4

12q23.2

+

2

15q15.1

2

9q33.2

+

2

11q22.3

-

2

5q13.2

2

1p34.2

+

2

3p21.31

+

2

1q41

2

14q32.12

+

ASCL1 BUB1B a

C5

CASP4

CCNB1 CDC20

Cell division cycle 20 homolog CELSR3 Cadherin, EGF LAG seven-pass G-type receptor 3 CENPF Centromere protein F, 350/400kDa (mitosin) CHGA Chromogranin A b

Chromogranin B

2

20p12.3

+

CPE

Carboxypeptidase E

2

4q32.3

+ +

c

+ +

+

+

[12,15]

-

-

[12,17]

+

+

+

+

+

+

Doublecortin

2

Xq23

+

+

ENC1

Ectodermal-neural cortex 1 Forkhead box G1

2

5q13.3

+

+

4

14q12

+

Guanine nucleotide binding protein (G protein), gamma 4 Hes family bHLH transcription factor 6 Internexin neuronal intermediate filament protein, alpha

2

1q42.3

2

2q37.3

+

2

10q24.33

+

42

+

+

+

[4,5,14,17,23] [9,15]

DCX

INA

+

+ +

[4,12,15] [14,15]

5q32

HES6

[15,17,23] +

Colony stimulating factor 2 1 receptor

+

[10,11,15,20,23] [14,15]

CSF1R

FOXG1 (B) GNG4

Mu References tat ed in NG S stu die s [16,20]

+

+

CHGB

SCLC cell lines

+

[4,5,17,19] and Supplementary Table 1

+

[5,12,17]

+

[4,12,19] and Supplementary Table 1

+

[4,5,15,20,23] [14,15,23]

+ +

[4,10,11,14,20] [10,20,23]

[15,20] +

[15,16,23]

INSM1

Insulinoma-associated 1 5

20p11.23

ISL1

ISL LIM homeobox

2

5q11.1

KIF1A

Kinesin family member 1A

2

2q37.3

KIF5C

Kinesin family member 5C LIM homeobox 2

2

2q23.1

+

2

2q23.1

+

2

4q27

+

2

8q11.21

2

11q23.2

2

7q31.1

2

5q22.1

+

+

2

15q26.1

+

+

3

1p36.11

+

+

[6,14,16,20]

Xq22.1

+

+

[4,10,20]

LHX2

MAD2L1 Mitosis arrest-deficient 2 like 1 MCM4 Minichromosome maintenance complex component 4 NCAM1 Neural cell adhesion molecule 1 NRCAM Neuronal cell adhesion molecule NREP Neuronal regeneration related protein homolog PRC1 Protein regulator of cytokinesis 1 STMN1 Stathmin 1

TMSB15 Thymosin beta 15a 2 A TOP2A Topoisomerase (DNA) II 3 alpha 170kDa

+

+

+

+

+

[10,11,15,16,20, 23]

+

+

[10,11,23]

+

+

+

+

+

[19,20] and Supplementary Table 1 +

[11,20] +

+

+

[14,15,23] [16,17]

+

+

17q21.2

[16,20,23]

[13,14,23] +

+

[4,14,20] [12,13,23] [14,20]

+

[13,17,20,23]

Abbreviations used: LCNEC, large cell neuroendocrine carcinoma; NEN, neuroendocrine neoplasm; NGS, next-generation sequencing; SCLC, small cell lung cancer +, upregulated; -, downregulated. a Reported once in carcinoid tumors and once in SCLC. b Reported once in carcinoid tumors and once in LCNEC and SCLC. c Reported once in carcinoid tumors and once in LCNEC.

43

Table 3 – Functions and involvement in cancer of proteins encoded by genes found to be altered in neuroendocrine neoplasms by at least two different genome-wide expression a profiling studies G Main e function n e

Reported in lung NENs?

b

Reported in other NE and Therapeutic potential? non-NE neoplasms?

A Genes involved in NE differentiation A S C L 1

Transcription factor in the neuroendocrine cell lineage.

Expressed by 7-13% of TCs; 22-85% of ACs; 57% of LCNECs and 72% of SCLCs. In an additional study, 5/5 carcinoids and 5/5 SCLCs positive.

H E S 6

Basic helixloop-helix transcription factor.

No

I N S M 1

Zinc-finger transcription factor, widely expressed in neuronal progenitor cells of the developing nervous system and NE organs. LIM homeodomain protein important in the development of pancreatic islets.

Highly expressed in SCLC cell lines.

I S L 1

Protein expressed in 21/23 (91%) SCLCs and 8/46 (17%) lung carcinoids.

Expressed in some lung adenocarcinomas, almost exclusively from non-smokers, but rarely in other NSCLCs. Promotes NE phenotype in prostate cancer. Expressed in a subset of NE breast cancers. Upregulated in progressive astrocytoma. NSCLC tumors with NE properties emerged from ASH1/SV40 large T antigen doubly transgenic mice. Overexpressed in glioma, and in breast, lung and renal cell carcinomas, and in metastatic colon cells. Promotes breast cancer cell proliferation. Expressed in a wide variety of NENs, including insulinomas, medullary thyroid carcinomas, medulloblastomas, neuroblastomas, pheochromocytomas, pituitary tumors and retinoblastomas.

Inhibition of ASH1 using siRNA suppresses ASH1expression lung cancer cells and decreases tumor volume of ASH1-positive LCNEC xenografts in mice.

Frequently expressed in primary and metastatic pancreatic endocrine tumors, as well as in rhabdomyosarcoma.

NR

NR

Promoter used for targeted gene therapy of SCLC.

B Genes related to cell cycle progression and proliferation B U B 1 B

Component of the MCC, interacts directly with MAD2; inhibits the APC/C.

C Regulatory C protein

No

Overexpressed in 20% of ACs, 84% of LCNECs and

Upregulated after APC inactivation in colorectal cancer. Mutations shown in families with mosaic variegated aneuploidy. Mutations not observed in human cancers. Overexpression associated with tumor cell proliferation in gastric cancer. Upregulation (correlated with poor overall survival) reported in a metaanalysis in nasopharyngeal carcinoma in six different expression studies. Overexpressed in many cancers, often dislocalized to

44

NR

Target for cancer immunotherapy: antibody

N involved in B mitosis. 1

C D C 2 0

Activator of the APC/C, but looses this ability when bound to MCC members.

C E N P F

Essential for the continuous activation of the spindleassembly checkpoint and sustained recruitment of BUBR1 and MAD1 at the kinetochore.

F O X G 1

Negative regulator of the TGF-β signaling pathway; represses p21 ( expression. B ) M Component of A the MCC, D interacts 2 directly with L BUBR1; 1 inhibits the APC/C; target of E2F.

84% of SCLCs.

the cytoplasm. Expression associated with poor prognosis in NSCLC, breast and gastric cancer.

responses against cyclin-B1 observed in many cancer types, including NSCLC and SCLC and in two premalignant lung lesions. No Highly expressed in several Putative therapeutic target. human cancer types, including Several inhibitors can bladder, breast, cervical and suppress its expression in gastric cancer, and cancer cells, including TAME glioblastomas. Associated and NAHA. with poor prognosis in epithelial ovarian cancer. Predictor of poor prognosis in primary NSCLC. No Upregulation (correlated with NR poor overall survival) reported in a meta-analysis in nasopharyngeal carcinoma in six different expression studies. Elevated protein levels associated with chromosomal instability, aneuploidy and poor patient outcome in breast cancer. Overexpression predicts poor prognosis in prostate cancer patients. A synergistic interaction between FoxM1 and CENPF may drive prostate cancer. Overexpression confirmed by Overexpressed in ovarian NR RT-PCR on SCLC cell lines. cancer compared to normal ovarial tissue and promotes proliferation of ovarian cancer cells. Promotes glioblastoma growth.

No

Reported to be upregulated after APC-inactivation in colorectal cancer. Overexpression associated with poor prognosis in neuroblastomas and NSCLC. Upregulation reported in a meta-analysis in nasopharyngeal carcinoma in six different expression studies. Also overexpressed in liver cancer, B-cell lymphoma, breast cancer, and 94/358 NSCLC samples. Mutant allele identified in mice after a mutagenesis screen causes chromosomal instability and formation of mammary tumors. In NSCLC, protein commonly expressed at higher levels than in normal lung and related to smoking history and poor differentiation. Overexpressed in breast cancer, cholioangiocarcinoma, colon cancer, NSCLC and pancreatic cancer.

M C M 4

Essential component of the prereplication complex.

No

P R C 1

Microtubulebinding protein involved in cytokinesis.

No

45

NR

NR

Promoter may be used for gene therapy in breast cancer.

S T M N 1

T O P 2 A

Cell cycle regulator; promoter of microtubule depolymerizati on; increased levels during synapse formation and neurite outgrowth of postmitotic neurons. Repairs doublestranded DNAbreaks; essential in the segregation of replicated chromosomes.

Particular high protein expression level in SCLC samples.

Related to metastatic disease NR and poor prognosis in several cancers, including NSCLC, where it is correlated with poor differentiation grade. Expressed by many endocrine tumors. Overexpressed in diffuse type of gastric cancer.

Expressed in lung carcinoid tumors; higher proportion of positive nuclei in ACs as compared to TCs.

Expressed in many cancer types. Frequently amplified in breast cancer. Both deletion and amplification associated with poor prognosis in breast cancer.

Targeted by many anti-cancer drugs, including doxorubicin, etoposide and mitoxantrone. Amplification and possibly deletion predictors of chemotherapy response in breast cancer.

C Genes related to neuronal development and neurite outgrowth C E L S R 3

Involved in brain development; mediates interactions between axons and guidepost cells to determine axonal tract formation. D MicrotubuleC binding X cytoplasmic protein. Directs neuronal migration.

E N C 1

No

No

NR

No

Expressed in many invasive brain tumors. Seems to function as a tumor suppressor in glioma. Better survival in DCX-expressing glioma patients. Gene mutation of DCX causes the severe neuronal migration disorder lissencephaly. Mutation not observed in the contect of cancer. Upregulated in colorectal carcinomas and hairy cell leukemia; mutated in glioblastoma cell lines. Mutants promote proliferation, repress caspase activity and suppress cellular apoptosis as induced by cisplatin in brain tumor cell lines. ENC1 protein dislocated from the nucleus to the cytoplasm in primary brain tumors as compared to normal brain tissue. Expressed in neuroblastomas and medulloblastomas, as well as in appendiceal carcinoid tumors. Gene overexpression associated with 1p19q codeletion in gliomas. Expression in gliomas related to oligodendroglial phenotype,

NR

Nuclear matrix No protein playing important roles in differentiation of neurons and the development of the central nervous system.

I Neuronal No N intermediate A filament protein involved in neurite outgrowth.

46

NR

Suggested as a predictive marker of chemosensitivity.

L H X 2

N C A M 1

N R C A M

N R E P

associated with favorable prognosis. LIM homeobox No Highly expressed in a variety protein which of human cancer types, serves as a including glioma, cancers of transcriptional the breast, stomach, germ activator in cell, kidney, skin, soft and neural muscle tissue, and pancreas. differentiation. High gene expression levels correlate with poor prognosis in breast cancer patients. Induces cell migration and invasion in murine breast cancer cells, at least partly by inducing PDGF-B/PDGFRβ signaling. Methylated in lung cancer cells. Adhesion Ubiquitously expressed in the Expressed in almost all molecule spectrum of lung NENs. astrocytomas, gliomas, involved in cell- Human tumor cell lines, as neuroblastomas and cell interactions well as 31 primary SCLCs rhabdomyosarcomas. have been reported to and signal Expressed in the majority of express exclusively the transduction. multiple myelomas and half of NCAM-140 variant. Other Involved in acute myeloid leukemias. Involved in the initiation and studies demonstrated initiation and maintenance of EMT. maintenance of presence of NCAM-180 EMT. Interacts variant in SCLC and/or SCLC with the cell lines and even its cytoskeletion specificity for these tumors. and implicated NCAM-PSA variant more in regulation of often present in SCLCs neurite (95%) than carcinoids (53%). outgrowth. Neuronal cell No Overexpressed in high-grade adhesion astrocytoma, glioma and molecule, glioblastoma tumor tissues, as involved in well as in thyroid papillary neural carcinomas. Overexpressed in development colorectal cancer tissues and and neurite independent predictor of poor outgrowth. outcome. Induced by betacatenin signaling in melanoma and colon cancer. Neuronal gene, No In glioblastomas, expressed expressed in notably at the invasive front of the developing primary tumors. Inhibition in distal lung, glioma cell lines decreased involved in migration. alveolar development, expression promotes neurite outgrowth.

NR

Putative therapeutic target. Number of immunotherapy strategies developed to target its expression. Successful treatment of small cell lung cancer xenografts in mice with the NCAM-targeting antibodydrug lorvotuzumab mertansine.

NR

NR

D Chromogranins and regulated secretory pathway genes A Precursor No G protein of T angiotensin I and angiotensin II, the latter involved in maintaining blood pressure. C Prohormone. More often positive by H immunohistochemistry in

Produced and secreted by glioblastoma cells. Delays angiogenesis and tumor growth of hepatocarcinoma in transgenic mice. M235T polymorphism may predispose for breast cancer.

Suppresses angiogenesis, tumor growth and metastasis by adenovirus-mediated gene transfer.

General marker for NENs. Increased levels in

Important diagnostic biomarker for NENs.

47

G A

carcinoids than HGNECs. Chromacin, parastatin and vasostatin more often expressed in pulmonary carcinoids compared to HGNECs, while chromacin more often present in the latter group.

C Prohormone H G B

CCB peptide has been reported to be produced by SCLC cell lines and is also expressed in primary carcinoid tumors. Expression validated by RTPCR in carcinoids; 76% of TCs positive for CPE protein. Majority of LCNECs and SCLCs negative for CPE protein.

C Prohormone P processing E exopeptidase and receptor to sort prohormones for the regulated secretory pathway.

K I F 1 A

K I F 5 C

Neuron-specific No motor protein belonging to the kinesin superfamily of proteins. Transports organelles containing synaptic vesicle proteins. Neuron-specific No motor protein belonging to the kinesin superfamily of proteins. Transports mRNAcontaining granules towards dendrites.

gastroenteropancreatic NETs, medullary thyroid carcinomas, Merkel cell carcinomas, neuroblastomas, pheochromocytomas, breast, ovarian and prostate carcinomas. Transfection of human CgA in mouse cell lines increased tumor necrosis and multi-nodular growth pattern. Also decreased tumor growth rate, but without affecting the proliferative index. General marker for NENs.

Measurement of plasma levels important for diagnosing and monitoring NET disease.

Less sensitive marker of NETs than CgA, although it may be more sensitive for pheochromocytomas and related tumors. NR

Expressed by pituitary adenomas and found in gene expression profiling studies to be overexpressed in insulinomas, as well as ACTH-producing thymic carcinoids. Expression of CPE-ΔN splice variant drives metastasis in hepatocellular carcinoma (HCC) cells and is an adverse prognostic marker for both HCC, pheochromocytoma and paraganglioma patients. High frequency of methylation May mediate docetaxel in the majority of human resistance in breast cancer tumors. Methylated in breast cells. cancer, and in 98% of primary head and neck squamous cell carcinoma tissues.

No

NR

C5a signaling plays an important role in the inhibition of the immune response to tumor cells. Elevated C5a plasma levels in NSCLC.

Possible therapeutic target. Inhibition of tumor growth could be achieved by a C5a antagonist at a similar rate as with paclitaxel.

E Other frequently deregulated genes C Member of the No 5 complement family, involved in inflammation and the immune response to infection;

48

C A S P 4 C S F 1 R

G N G 4

cleaved in C5a (potent chemoattractan t and inflammatory mediator) and C5b (deposited on cell surfaces). Caspase No protein that plays a role in ER-stress initated apoptosis. Differentiation No of macrophages.

Member of the G protein gamma family, regulating the activities of various enzymes and ion channels. Member of the thymosin-β family of actinbinding molecules.

No

Mutated in rare cases of colorectal carcinomas.

NR

Overexpressed in a variety of human malignancies, including acute myeloid leukemia, Non-Hodgkin lymphoma, prostate cancer, renal cell carcinoma and breast, endometrial and ovarian cancer. Associated with a poor disease outcome in breast , endometrial, and ovarian cancer.

Tyrosine kinase receptor, closely related to PDGFRB; sensitive to imatinib and sunitinib. Combined treatment using CSF1R-blocker PLX3397 and chemotherapy decreases primary breast tumor growth and metastasis and improves survival of mice. PLX3397 works in an extrinsic way and targets tumor associated macrophages rather than the primary tumor. It reverses macrophagemediated resistance to androgen blockade therapy. Simultaneously blocking CSF1R in tumor cells and tumor associated macrophages decreases in vivo invasion to background levels. JNJ-28312141 potent inhibitor of CSF1R; efficiently reduces number of tumor associated macrophages. NR

No

T No Upregulated in metastatic Positive predictor of therapy in M prostate cancer, and in highly triple-negative breast cancer. S metastatic human breast and B melanoma cell lines. 1 Expression levels in prostate 5 cancer correlated with tumor A grade. a See Supplementary Table 2 for a version of this Table listing all referred studies. b In other studies than the gene expression profiling studies listed in Table 1. Abbreviations used: AC; atypical carcinoid; APC, adenomatous polyposis coli; APC/C, anaphase-promoting complex/cyclosome; CgA, chromogranin A; CIN, chromosomal instability; EMT, epithelial to mesenchymal transition; LCNEC, large cell neuroendocrine carcinoma; MCC, mitotic checkpoint complex; NE, neuroendocrine; NET, neuroendocrine tumor; NEN, neuroendocrine neoplasm; NR, not reported; NSCLC, non-small cell lung cancer; SCLC, small cell lung cancer; TC, typical carcinoid.

49

Figure 2 – Proteins involved in the spindle assembly checkpoint for which transcriptional activity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

was found to be upregulated in neuroendocrine lung neoplasms. When CDC20 is bound by BUB1R, BUB3 and MAD2 to form the mitotic checkpoint complex (MCC), its stimulating activity towards the anaphase-promoting complex/cyclosome (APC/C) is blocked. Inhibition of the APC/C will prevent the cell to enter the anaphase until all chromosomes are correctly attached to the mitotic spindle. The APC/C can polyubiquitylate and thus degrade several downstream targets, of which Cyclin B1 and CENP-F are depicted. Both cyclin B1 and CENP-F can be activated by the Forkhead transcription factor FoxM1 [115]. Also stathmin has been described to be a direct transcriptional target of FoxM1 [116]. In green the proteins or protein complexes are indicated for which the transcriptional levels were described to be upregulated in at least two studies (Table 2), while other proteins or protein complexes are shown in blue.

40

Figure 1

Isl1

Jak1 Stat3

DCX

INSM1

Cdk5

NeuroD

INS

secretion

CHGA

TCF3

ASH1

CHGB

secretion

ISL1

NCAM1

miR-375

HES6

HES1

Figure 2

MAD2

Stathmin

CDC20

BUBR1

BUB3

CENP-F

FoxM1

Cyclin B1

APC/C

MCC