Niels Hilmer Nielsen

Cyclin E overexpression, a negative prognostic factor in breast cancer with strong correlation to oestrogen receptor status

Cyclin E is a G1 cyclin which has been proposed to be one of the key regulators of the important G1/S transition, and could consequently be a potential deregulated molecule in tumours. Recently, it has been observed that cyclin E is overexpressed in a variety of malignancies including breast cancer and that several isoforms of the protein exists. In this study we have characterised the cyclin E expression in 114 tumour specimens from patients with primary breast cancer using Western blotting. Various expression of cyclin E was observed among tumours and a group of 27 patients out of 100 patients with stage I-III disease, identified as having tumours with high cyclin E levels, had a significantly increased risk of death and relapse from breast cancer (P = 0.0002 and P = 0.015 respectively). Even in the subgroup of axillary node-negative patients the cyclin E level was of prognostic importance. There was also a strong association between cyclin E expression and oestrogen receptor status (P < 0.00001), and tumours with high cyclin E expression were in general oestrogen receptor negative, suggesting a potential role for cyclin E in mechanisms responsible for oestrogen-independent tumour growth.

Deregulation of cyclin E and D1 in breast cancer is associated with inactivation of the retinoblastoma protein

Inactivation of the retinoblastoma protein (pRB) by mutations or abnormal phosphorylation is a mechanism by which tumour cells can subdue normal growth control. Among molecules involved in control of pRB phosphorylation, cyclin D1 and E have been found to be deregulated and overexpressed in various types of cancers. In order to study the cell cycle regulatory mechanisms in breast cancer, we have analysed the protein expression of cyclin D1 and E in 114 tumour specimens from patients with primary breast cancer using Western blotting. Twenty-five out of 34 tumours with overexpression of cyclin E showed uniform low cyclin D1 expression, and by immunohistochemical analysis of pRB we present evidence for the existence of pRB defects in approximately 40% of these tumours in contrast to no pRB defects in the other group of tumours. This result was supported by a high protein expression of the cyclin-dependent kinase inhibitor p16 in 44% of the tumours with high cyclin E and low D1 expression, and all immunohistochemical pRB defect tumours showed a high p16 protein level. Additionally, an abnormal low pRB phosphorylation in relation to a high proliferative activity and loss of heterozygosity of the retinoblastoma susceptibility gene locus were found in all but one tumour with immunohistochemical defect pRB. Interestingly, tumours with high cyclin E and low D1 expression were generally oestrogen receptor negative suggesting a role for cell cycle regulators in the mechanisms leading to oestrogen independent tumour growth. Furthermore, the prognosis differed markedly for the patients in the various groups of tumours, indicating that the heterogeneous nature of breast cancer pathogenesis and the clinical course in part could be explained by different and distinctive sets of cell cycle defects.

The cyclin D1 high and cyclin E high subgroups of breast cancer: separate pathways in tumorogenesis based on pattern of genetic aberrations and inactivation of the pRb node

In an attempt to identify subtypes of breast cancer and pinpoint patterns of cell cycle regulatory defects associated with clinical behaviour, proliferation and other transformation associated events, a multitude of cell cycle regulatory proteins were analysed in a material of 113 primary breast cancers. Increased proliferation was observed in two different scenarios; (1) with high cyclin D1 and elevated retinoblastoma protein (pRb) phosphorylation, (cyclin D1high tumours) or (2) with high cyclin E protein but low cyclin D1 and lack of corresponding pRb phosphorylation (cyclin Ehigh tumours) indicative of an interrupted pRb pathway. Characteristic for cyclin Ehigh tumours were further defects in p53, p27 and bcl-2, while c-erbB2 overexpression and c-myc amplification was found in both cyclin D1high and Ehigh tumours. Using transfected cell lines overexpressing cyclin E, cyclin Ehigh and D1high tumours were mimicked and the cyclin D1high cell line normalized the cyclin E kinase activity by an induction and redirection of p21 and p27 to the cyclin E complex whereas cyclin Ehigh cell lines obtained increased kinase activity without redirection of inhibitors. Based on differences in genetic aberrations as well as function of the pRb node we therefore propose a model in which cyclin D1high and cyclin Ehigh tumours represent two alternative mechanisms to inactivate the pRb pathway and thereby achieve unrestrained growth in the tumorogenesis of breast cancer.

Tumor specific VEGF-A and VEGFR2/KDR protein are co-expressed in breast cancer.

Angiogenesis is a prognostic indicator in primary breast cancer regulated by specific angiogenic factors and their receptors. Vascular endothelial growth factor-A (VEGF-A), so far considered the most important, acts through dimerization of the receptor VEGFR2/KDR within the receptor tyrosine kinase family of VEGF receptors. In order to study the interplay between VEGF-A and VEGFR2/KDR in breast cancer we evaluated their expression by immunohistochemistry in 102 breast cancers organized in a tumor tissue array system allowing semi-quantitative evaluation of cytoplasmatic staining intensity. In addition, VEGF-A165 was analyzed by an enzyme immuno assay (ELISA) in protein extracts prepared from frozen tissue from 98 of 102 tumors included in the array. Cytoplasmatic staining of VEGF of varying intensity was observed in all samples and correlated with the ELISA results of VEGF content (p = 0.007). Interestingly, VEGFR2/KDR expression correlated with VEGF expression using immunohistochemistry, indicating that VEGF and VEGFR2/KDR may be co-expressed in breast cancer. Furthermore, high levels of VEGF-A165 in the protein extracts was associated with impaired short time survival but not long term survival whereas immunohistochemically assessed VEGF and VEGFR2/KDR were not significantly associated with survival. In summary, immunohistochemically based analysis of VEGF using a tumor tissue array system seems to be a useful method for VEGF quantification in breast cancer here validated using an ELISA based method. The tumor tissue array system enables opportunities of simultaneous analysis of markers engaged in angiogenesis justifying further studies using larger series of tumors.

Downregulation of the potential suppressor gene IGFBP-rP1 in human breast cancer is associated with inactivation of the retinoblastoma protein, cyclin E overexpression and increased proliferation in estrogen receptor negative tumors.

The complex insulin-like growth factor network of ligands, receptors and binding proteins has been shown to be disturbed in breast cancer. In addition to defects in proteins controling cell cycle checkpoints, this type of aberrations could affect tumor growth and survival thereby influencing both tumor aggressiveness and potential response to treatments. We have previously identified the T1A12/mac25 protein, which is identical to the IGFBP-rP1, as a differentially expressed gene product in breast cancer cells compared with normal cells. Here we compare the expression of IGFBP-rP1 in 106 tumor samples with known status of cell cycle aberrations and other clinicopathological data. This was done using a tumor tissue section array system that allows for simultaneous immunohistochemical staining of all samples in parallel. Cytoplasmic staining of variable intensity was observed in most tumors, 15% lacked IGFBP-rP1 staining completely, 20% had weak staining, 32% intermediate and 33% showed strong staining. Low IGFBP-rP1 was associated with high cyclin E protein content, retinoblastoma protein (pRb) inactivation, low bcl-2 protein, poorly differentiated tumors and higher stage. There was a significantly impaired prognosis for patients with low IGFBP-rP1 protein tumors. Interestingly, IGFBP-rP1 showed an inverse association with proliferation (Ki-67%) in estrogen receptor negative tumors as well as in cyclin E high tumors suggesting a separate cell cycle regulatory function for IGFBP-rP1 independent of interaction with the estrogen receptor or the pRb pathway.

G1‐S transition defects occur in most breast cancers and predict outcome

Cell cycle deregulation is frequently observed in tumors and has moreover been proposed to be a requirement for tumor development. By analyzing the expression of p27 by immunohistochemistry in 100 primary breast tumors and combining the analyses with our earlier characterization of cyclin E, D1, p16, and the retinoblastoma protein (pRB), we have been able to cover the majority of potential G1–S transition defects and observed that 90% of the tumors had alterations in one or several cell cycle regulatory proteins. Considerable variations in protein levels were found among tumors, with low p16 expression as the most common alteration followed by cyclin E or cyclin D1 overexpression, low p27 expression or pRB inactivation in decreasing prevalence. Tumors were grouped according to observed combinations of defects and the proliferative capacity was determined for each group by analyzing Ki–67 labeling index. Low proliferation was observed in tumors with: low p16; high cyclin Dl with normal or high p16 expression; and in tumors without cell cycle defects. Tumors with high cyclin E/low p27 or pRB defects showed higher proliferation. The survival differed noticeably for patients with various combinations of cell cycle defects, and four distinctive clusters were identified showing significantly different breast cancer specific survival (p < 0.0001) for both node-positive (p=0.0006) and node-negative patients (p < 0.0001). In summary, we have shown that G1-S transition defects are nearly obligatory in breast tumors and that the specific type of cell cycle defect influences the clinical behavior of the tumor.

Telomerase activity in relation to p53 status and clinico‐pathological parameters in breast cancer

Cell cycle deregulation can occur at different levels in cancer. In human breast cancer it includes overexpression of cyclins D1 and E, down‐regulation of cyclin‐dependent kinase inhibitors and inactivation of the retinoblastoma and p53 tumor suppressor proteins. Telomerase activity is strongly associated with an immortal phenotype and expression of telomerase is linked to the cell cycle. We have recently demonstrated a connection between specific cell cycle defects within the pRB pathway and levels of telomerase activity in breast cancer. In the present study, 106 tumors were investigated for p53 gene and protein status. By single strand conformation polymorphism (SSCP) analysis, 15% showed mutations within exons 5–8 and by immunohistochemistry (IHC), 29% were p53 positive. Tumors with a telomerase activity above median (i.e., telomerasehigh) were significantly associated with p53 protein accumulation (p = 0.004), but not with p53 gene mutations. The strongest telomerase expression was found in tumors with p53 protein accumulation. Morphologic grade, estrogen and progesterone receptor expression differed significantly between the telomerasehigh and telomeraselow groups (p < 0.0001, p = 0.016 and p = 0.046, respectively), but no difference was observed for stage or nodal status. Telomerasehigh tumors were significantly associated with a poor prognosis for node‐negative (N0) patients (p = 0.008), but not for node‐positive (N+) patients, whereas the opposite was demonstrated for tumors with p53 accumulation. The survival data indicated that telomerase expression has biological importance particularly for N0 tumors, suggesting that telomeraselow tumors constitute a group of “pre‐immortalized” tumors with a good prognosis. Int. J. Cancer (Pred. Oncol.) 79:343–348, 1998.

Cyclin E dependent kinase activity in human breast cancer in relation to cyclin E, p27 and p21 expression and retinoblastoma protein phosphorylation

The cell cycle machinery is regulated by cyclin dependent kinases and sets of activating and inhibitory proteins. The G1-S control mechanism is often deregulated in tumours supposedly leading to increased kinase activity, phosphorylation of substrates and subsequent S phase entrance. Increased kinase activity has been proposed to be essential in cell cycle aberrations, but few studies have actually shown enhanced kinase activity related to specific cell cycle defects in primary tumours. In the present study we have determined the cyclin E dependent kinase activity (cyclin Ekinase) in 59 primary breast cancers, using an H1-kinase assay, and related the activity to the expression of cyclin E, p27 and p21. In a subgroup of 48 tumours, we further characterized the association between cyclin Ekinase, in vivo phosphorylation of the retinoblastoma protein (pRb) and proliferation. The cyclin Ekinase correlated significantly with cyclin E content and inversely with p27 and p21 expression. P27, but not p21, was associated with low cyclin Ekinase in specimens with normal/low levels of cyclin E. At elevated cyclin E levels, suppression of cyclin Ekinase seemed to require high levels of both p21 and p27. The cyclin Ekinase correlated with the phosphorylation status of pRb as well as with proliferation. Surprisingly, pRb phosphorylation did not correlate with proliferation. Our results support that pRb is a substrate for cyclin Ekinase in primary breast cancer and that deregulation of cyclin E and p27 act through increased CDK-kinase activity, but cyclin E associated events beside pRb phosphorylation might be rate-limiting for entrance into S phase.

Cyclin E expression and proliferation in breast cancer

Cyclin E is a part of the cell cycle machinery and aberrantly expressed in several malignancies including breast cancer. Since cyclin E is cell cycle specifically expressed, we wanted to examine the relation between proliferation and expression of cyclin E with special attention to tumours with overexpression of the protein. Seventy-four breast tumours were analysed for the expression of cyclin E by immunohistochemistry and Western blotting and related to the growth fraction determined by Ki-67. Significant correlations were obtained between the growth fraction, the percentage of cyclin E positive cells, the intensity of cyclin E and total amount of cyclin E determined by Western blotting. The majority of the tumours had less cyclin E than Ki-67 positive cells indicating a conserved cell cycle specific expression of the protein which further was supported by flow cytometric analysis of breast cancer cell lines. The cell cycle specificity of cyclin E was found even in tumours with inactivated retinoblastoma protein (pRB) demonstrating the existence of a pRB independent regulation of cyclin E. A fraction of the tumours had considerably elevated cyclin E levels that were not in relation to the proliferative activity as observed for the other tumours. These tumours were in general highly proliferative and considered to overexpress cyclin E. Patients with tumours of high proliferative activity, high total cyclin E levels or disproportionally elevated cyclin E expressions in relation to proliferation had significantly increased risk of death in breast cancer, whereas the intensity of the immunohistochemical cyclin E staining did not affect the survival.