Abstract
Lipocalins are a large protein family with only little sequence homology but highly conserved structural similarity. Many lipocalins play crucial roles in the generation of epithelial cancer, influencing pathways which regulate cell motility, cell differentiation and neovascularisation. Thereby they can be used as biomarkers of cancer, in most cases for a rather good prognosis. Glycodelin is a lipocalin existing in three isoforms which differ only by glycosylation, but which have different functions. In breast cancer, glycodelin A is known to contribute to a more differentiated cell morphology and is a biomarker for a favourable prognosis, but also plays a role in angiogenesis. Glycodelin A is a useful prognostic marker as it can be detected in serum samples, but is also a target for therapeutical interventions.
The Lipocalin Family – Structure and Function
Lipocalins are small secreted proteins, belonging to the calycin-protein family and can be found in nearly all organisms from eubacteria to eukaryotic cells (1). Interestingly the members of the lipocalin family exhibit only little sequence identity. They are classified as ‘kernel’ lipocalins, sharing three conserved regions, or as ‘outlier’ lipocalins, sharing less than these three conserved regions (2). Nevertheless, they are characterized by a high structural similarity, consisting of eight anti-parallel β-sheets connected by seven short loops, thereby forming a cup-like structure with loop 1 as a lid (3). Lipocalins bind lipophilic molecules within this cup-like structure; binding specifity is thereby influenced by cup size and conformation, as well as amino acid composition. Apart from molecular binding and transport, a variety of functions of the different members of the lipocalin family are known: binding of cell surface receptors, complex formation, invertebrate coloration, olfaction and pheromone transport, prostaglandin synthesis, regulation of cell homeostasis and modulation of immune response (3).
Glycodelin
General facts. The human glycodelin gene Human Genome Organisation (HUGO) gene symbol: PAEP, progesterone-associated endometrial protein (4); is located on chromosome 9q34 (5), in a region which comprises of many other genes of the lipocalin family (6-8). The gene consists of seven exons and six introns (9). From the primary gene sequence, a 180-amino-acid protein is made, which includes an 18-amino-acid signal sequence and three N-glycosylation sites at Asn 28, 63 and 85 (10). As these glycosylation sites can be differentially glycosylated, influencing functions of the molecule, it was called glycodelin (10-12), although many different names exist for the protein, depending on the tissue in which it was described (10, 13-17). Glycodelin, which is the major lipocalin of the reproductive axis, involved in cell recognition and differentiation (18), shows high structural similarity to β-lactoglobin, the major constituent of whey (19, 20), and has a retinol-binding motif (21), although glycodelin was never found to bind retinol (22).
Glycodelin isoforms and non-oncogenetic functions. Three isoforms of glycodelin are described in the literature, differing only in glycosylation patterns and in their function (23). An important point is that different glycosylation does not influence protein folding (22).
Glycodelins are synthesized in glandular epithelial cells of the secretory endometrium (24), in the Fallopian tubes (10, 25), in the seminal gland and ampulla of the ductus deferens (23, 26), in ovarian tumors and healthy ovaries (27, 28), in bone marrow (16), and in healthy breast tissue and breast tumors (29). It can also be detected in the epithelial cells of the umbilical cord vein (30).
First of all Glycodelin A is found in the amniotic fluid, the endometrium, the decidua and the serum of pregnant women (22). Glycodelin A has a molecular weight of 18.78 kDa determined from the cDNA sequence (20), but the molecular mass determined by gel electrophoresis is significantly higher (28 kDa) due to glycosylation (31), as carbohydrates account for 17.5% of the protein (14). Its expression is stimulated by progesterone (18). Glycodelin A exhibits anti-fertilisation activity by binding spermatozoa and thereby inhibiting the binding of the sperm to the zona pellucida of the oocyte (32). Binding of the sperm to the oocyte is only possible during the so called ‘fertilisation window’, in which the absence of glycodelin A is regulated by oestrogen. While the anti-fertilisation activity is glycosylation-dependent (33), its immunosuppressive activity is not dependent on the state of glycosylation of the molecule. The role of glycodelin A in the regulation of the immune system is inhibition of lymphocyte proliferation (34), natural killer cell cytotoxicity (35), T-cell proliferation and Th1-cytokine response (36), induction of T-cell apoptosis (37) and even regulation of B-cell response (38) by binding the human B-cell receptor CD22 (39). Furthermore, glycodelin A blocks E-selectin-mediated cell adhesion in vitro much more strongly than sialyl-Lewisx (40). An intriguing fact is that the inhibition of natural killer cell activity by glycodelin A permits the implantation of the embryo in the maternal placenta by counteracting the maternal defence against implantation. Glycodelin A is therefore found to be increased during the implantation phase (10, 41). In other words, glycodelin A is expressed during the first two to three days post-menstruation by the glandular cells of the endometrium, then, around the time of ovulation, glycodelin is no longer detectable (42, 43); expression returns around the fifth post-ovulatory day (44-46). In the case of pregnancy, glycodelin A secretion increases profoundly (13, 34, 47); post-menopausally, only very low glycodelin A levels are measured (48, 49).
Glycodelin F, which is mainly expressed in the ovary, and synthesised in the granulosa cells (50), has a function in principle similar to that of glycodelin A (51). It also binds the sperm head, thereby inhibiting acrosome reaction and sperm-egg binding. Upon de-glycosylation, glycodelin F dissociates from the sperm and sperm-egg binding is possible. The de-glycosylation takes place during the passage of the sperm through the corona cell layer. Glycodelin F is thereby important to prevent a premature acrosome reaction.
Glycodelin S is mainly found in the seminal fluid and is differentially glycosylated from glycodelin A, being rich in fucose, with Lewisx/Lewisy antennae (33). It does not inhibit sperm-egg binding, although it also has two binding sites on spermatozoa (52, 53). The binding of glycodelin S is not in competition with glycodelin A/F. Glycodelin S reduces cholesterol efflux from spermatozoa, thereby suppressing capacitation. During the passage of the sperm through the cervix (52), glycodelin S is de-glycosylated and dissociates from the sperm, allowing the sperm to mature.
Glycodelins in cancer. Expression of glycodelins was found in some types of breast cancer (16), ovarian serous carcinoma (54) and bi-phasic synovial carcinoma (55). The presence of glycodelin in cancerous tissue is mostly linked to a better prognosis than is attributed to glycodelin-negative tissue (54), as glycodelin is a protein typical of differentiated tissue. Furthermore, experiments demonstrated that the presence of glycodelin reduces cell proliferation and reverses malignant features (18, 29, 56, 57). In vitro as well as in vivo models of tumorigenesis found that glycodelin up-regulates the expression of tumour-suppressor genes, while it reduces the expression of oncogenes, thus it could be concluded that glycodelin is itself a tumor-suppressor (58). Furthermore expression of glycodelin correlates with tumour grading, meaning it is reduced in G2/FIGO (International Federation of Gynecology and Obstetrics) III-IV tumours (59). On the other hand, there are indications that glycodelin might play a role in neovascularisation, thereby promoting tumour growth (60).
Role of Glycodelin A in Breast Cancer
When it became known that glycodelin A was expressed in tissues outside the reproductive tract, for example in normal and neoplastic glandular epithelia of the breast, its role in these tissues was studied. As glycodelin A was found to be localized in highly differentiated acinar epithelia and a role in organization of epithelial tissue was presumed. To further clarify this finding, cDNA of glycodelin A was transfected into MCF-7 cells (29). After transfection, MCF-7 cells showed an altered growth behaviour, including the formation of acinar configurations, growth instability in semisolid media because of apoptosis, expression of markers of organized epithelia such as cytokeratins (CK) 8 and 18, and E-cadherin, changes in the intracellular distribution of β-catenin and decreased proliferation; after transfection with glycodelin cDNA the cells exhibit a phenotype of organized epithelium (29). The findings of Hautala et al. fit these results: They introduced MCF-7 cells transfected with glycodelin cDNA into mouse mammary fat pads. The result was a formation of smaller tumors with a more differentiated phenotype in comparison to implantation of non-transfected MCF-7 cells (61). It was concluded that glycodelin induces differentiation, reduces the expression of oncogenes, and similarly increases the expression of tumor-suppressor genes, thereby contributing to a more favourable prognosis (58). Nevertheless, these results obtained in model systems have to be treated with caution. It was recently shown by tissue microarray experiments analyzing gene expression that glycodelin expression is associated with a low proliferation rate and well-differentiated cell forms in sporadic cases of breast cancer. In contrast, in familial non-BRCA (breast cancer associated genes)1/2 tumours, glycodelin expression is linked to a worse prognosis, with positive lymph nodal state, expression of human epidermal growth factor receptor 2 (HER2) and increased risk for distant metastasis. There seem to be different gene expression profiles in familial and sporadic breast cancer, resulting in different pathways of disease progression (62), which need to be taken in account when evaluating glycodelin expression and its meaning for the patient.
In regard to sporadic breast cancer, which comprises the largest group within breast cancer, it was found that the expression of glycodelin A is independent of different histological forms (63) and of grading (64), but a slight correlation to steroid receptor expression was found, meaning that glycodelin A could be a marker for differentiation. In 2010, an increase in glycodelin expression was found in estrogen receptor/progesterone receptor (ER/PR)-positive tumor samples which had a lymph node involvement (65), so in addition to being a marker for breast cancer differentiation, there might be some role for glycodelin A in lymph node metastasis. Furthermore, the expression of glycodelin A could be regarded as a prognostic marker, as it decreases with increasing malignancy and exhibits higher expression in tissues from patients with good prognosis (66). However, it should be borne in mind that the presence and expression of the glycodelin protein is the decisive factor, not the expression of its mRNA, which does not decrease with increasing malignancy (67). This fact has to be considered for evaluation of the study of Kostadima et al., who extracted mRNA from paraffin-embedded breast cancer tissues and correlated the mRNA expression to clinicopathological and molecular parameters, and to patient outcome. They found no prognostic utility for glycodelin mRNA expression for overall survival or disease-free survival (68).
The experiments of Song et al. represent a strong contrast to the data above. As it was known that glycodelin A was found in endothelial cells of the umbilical cord and tumor blood vessels, they examined the tube formation and migration of human umbilical vein endothelial cells upon addition of amniotic fluid, which is rich in glycodelin, or of a synthetically-produced protein mimicking glycodelin. An increase in tube formation and cell migration was found, indicating a promotion of angiogenesis in vivo. This effect was blocked by antibodies against the synthetic peptide or vascular epidermal growth factor (VEGF). It could, therefore, be supposed that the effect of glycodelin is mediated by VEGF, as glycodelin increases the release of both VEGF protein and mRNA expression, as well as mRNA expression of VEGF receptor. From all these results, it could be concluded that glycodelin A plays a role in neovascularisation of tumors (60).
A point for therapeutical intervention comes from the group of Ramachandran et al., who found that the expression of glycodelin A is induced by lipophosphatidic acid (LPA) in a dose-dependent manner. LPA is rather similar to phorbol-myristate acid (PMA), which is increased in the serum of patients with cancer. Controlling the amount of LPA/PMA could help prevent glycodelin A expression and thereby reduce neoangiogenesis (69).
Other Lipocalins and their Associations with Breast Cancer
Prolactin. Prolactin is known to induce the proliferation of normal alveoli in pregnancy, especially during lactation, and abnormal alveoli in hormonally-dependent breast cancer (70) and especially increases the risk for male breast cancer by a transforming growth factor α-mediated pathway (71). An intervention into this pathway could be the basis for a therapeutic strategy directed against male breast cancer.
Apolipoprotein D. Apolipoprotein D is a small (24 kDa) glycoprotein, regulated by retinoic acid. The presence of retinoic acid leads to an accumulation of apolipoprotein D, especially in ER-positive breast cancer cell lines, and inhibits proliferation and tumor progression by establishment of a more differentiated phenotype (72). Furthermore it is involved in intracellular ligand binding and inhibits translocation of phosphorylated mitogen-activated protein kinases into the nucleus. Thereby it is associated with a favourable histology and grading in breast cancer. However, apolipoprotein D is associated with an adverse prognosis, if it is found in the tumor stroma, and ER-positive/apolipoprotein D-positive cells have a non-functional ER pathway, hence the cancer cells would not react to a tamoxifen-based therapy (73). Moreover, apolipoprotein D seems to correlate with lymph node metastasis, therefore it also has a prognostic relevance (74).
Lipocalin 2. Also known as neutrophil gelatinase-associated lipocalin (NGAL), lipocalin 2 is a small secreted protein and is known by many different names in the literature. Lipocalin 2 is in fact mostly used for the mouse homolog of NGAL (2).
LCN2 stabilizes matrix metalloproteinase-9 from auto-degradation. If Lipocalin 2 is added to matrix metallo-proteinase-9-positive breast cancer cells, they exhibit a more aggressive phenotype, and when secreted by macrophages, lipocalin 2 induces cellular growth of MCF-7 cells (75). Up-regulation of lipocalin 2 results in increased tumor growth and angiogenesis (76) and is hence involved in metastasis formation (77). But as it can be detected in patients' urine, it seems to be a useful marker for the determination of tumor stage. Furthermore lipocalin 2 is correlated with the disease severity score (78), a negative hormone receptor status, HER2 overexpression, poor grading, lymph node metastasis, and a high proliferation index (Ki-67), and is generally a predictor of poor prognosis (79). However, it has to be distinguished between the low risk groups on one hand, where it is a marker of pathological complete remission after neoadjuvant chemotherapy, and on the other hand primary breast cancers, where it can be regarded as a prognostic factor for reduced disease-free survival (80). One therapeutic strategy directed against the lipocalin 2 pathway uses the transcription factor C/enhancer binding protein ζ, which inhibits the transcription of LCN2 and thereby inhibits migration and invasion in breast cancer (81). Lipocalin 2 is also involved in red blood cell production, leading to anaemia, which is frequently found in patients with cancer. As up-regulation of HER2 results in up-regulation of lipocalin 2 via the nuclear factor κB-pathway, this could be another starting point for therapeutical intervention (45). The most recent research in the field of lipocalin 2 is on micro RNAs, regulating it´s expression. If micro RNA miR-138 is present, then the expression of lipocalin 2 is significantly reduced, as is cell migration. Therefore miR-138 can be regarded as a tumor suppressor, preventing tumor formation and metastasis growth (82).
Similar to the literature on human lipocalin 2 are the findings for lipocalin 2 in mouse. It also seems to be regulated by HER2 via nuclear factor κB-pathways, and is up-regulated in epithelial cancer. Up-regulation of lipocalin 2 is accompanied by an increased expression of mesenchymal markers and a decrease in epithelial markers (83, 84), hence promoting tumor progression by epithelial–mesenchymal transition. Like lipocalin 2 in humans, it is known to stabilize matrix metalloproteinase-9 (85) and to regulate angiogenesis via the VEGF pathway, promoting angiogenesis and tumor progression (84), as well as breast cancer cell invasion in mice (86). Lipocalin 2 double-knockout mice have a delayed onset of mammary tumors, a decreased tumor burden, and a reduced matrix metalloproteinase-9 activity, but no reduction of lung metastasis was found (87). Lipocalin 2 can also be regarded as a biomarker: elevated levels in serum samples indicate a reduced disease-free survival (88). In contrast to that are results from Cramer et al., who found no correlation of Lipocalin 2 with tumor size, tumor appearance, tumor volume and number of metastases, and thus call the role of Lipocalin 2 in tumor development into question (89).
As a conclusion, most kinds of lipocalins play a role in tumor development, progression, angiogenesis and formation of distant metastases, and therefore have a significant impact on cancer diagnosis. Much research will be required to exploit all the therapeutical possibilities which exist on the basis of these molecules.
- Received December 11, 2013.
- Revision received January 7, 2014.
- Accepted January 9, 2014.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.