STK15/Aurora-A Expression in Primary Breast Tumors
Is Correlated with Nuclear Grade but Not with
Prognosis
Melanie E. Royce,M.D., Ph.D.1 Weiya Xia,M.D.2 Aysegul A. Sahin,M.D.3 Hiroshi Katayama,Ph.D.4 Dennis A. Johnston,Ph.D.5 Gabriel Hortobagyi,M.D.1 Subrata Sen,Ph.D.4 Mien-Chie Hung,M.D.21Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas.
2Department of Molecular and Cellular Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas.
3Department of Pathology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. 4Department of Molecular Pathology, The Univer-sity of Texas M. D. Anderson Cancer Center, Hous-ton, Texas.
5Department of Biomathematics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas.
Supported by Grants CA 58880 and CA 78633, by a SPORE grant in ovarian cancer (CA 83639, to M.-C.H.), by the Nellie Connally Breast Cancer Research Fund at the M. D. Anderson Cancer Center (M.-C.H.), and by Grants CA 61979 and CA 89716 (S.S.).
Melanie E. Royce and Weiya Xia contributed equally to the current study
Address for reprints: Melanie E. Royce, M.D., Ph.D., Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Box 424, 1515 Holcombe Boulevard, Houston, TX 77030; Fax: (713) 794-4385; E-mail: mroyce@mdanderson.org
Received June 5, 2003; revision received Septem-ber 24, 2003; accepted SeptemSeptem-ber 24, 2003.
BACKGROUND.DNA amplification on chromosome 20q13 is commonly detected in breast carcinoma and is correlated with poor prognosis. STK15 maps to this amplicon. The objective of the current study was to use immunohistochemistry to determine STK15 expression in primary breast tumors. The authors also explored whether STK15 was a prognostic factor for breast carcinoma by comparing the level of STK15 gene expression with clinical parameters that are known prognostic factors for the disease.
METHODS.Archival mastectomy and lumpectomy specimens, randomly selected, were immunohistochemically stained to determine the STK15 gene expression level. The clinical parameters of these same patients were reviewed retrospectively and analyzed for correlations with STK15 expression level, based on a positive-versus-negative scoring system.
RESULTS.Of the 112 human breast tumor specimens analyzed, 26% stained pos-itively for STK15 by immunohistochemistry. Of the tumors, that stained pospos-itively 62.1% had a well-to-moderately differentiated nuclear grade. The correlation be-tween STK15 staining and nuclear grade was nearly statistically significant (P ⫽ 0.05). No association was found between STK15 staining and tumor size, lymph node status, or hormone receptor status. Analysis of recurrence-free survival and overall survival rates also failed to reveal a statistically significant difference be-tween the two groups.
CONCLUSIONS.STK15 expression by immunohistochemistry was noted in
approx-imately one-fourth of primary breast tumors. STK15 expression was associated with nuclear grade, but no correlation was found between the other clinical parameters evaluated. Furthermore, no differences were found in survival rates when they were analyzed by level of STK15 staining. Cancer 2004;100:12–9.
© 2003 American Cancer Society.
KEYWORDS: BTAK, Aurora-A, AIK1, breast carcinoma.
D
NA amplification on chromosome 20q13 is commonly detected in breast carcinoma and is correlated with poor prognosis.1A novelgene, STK15 (also known as BTAK/Aurora-A/AIK1), maps close to the critical region of amplification defined on this amplicon.2– 6STK15 is
a member of the protein serine/threonine kinase family. Its deduced amino acid sequence exhibits conservation of all subdomains pre-dicted in this protein kinase super family.2 The translated STK15
peptide consists of 403 amino acids and has a molecular weight of 46 kilodaltons.2STK15 exhibits significant homology with the previously
cloned prototypic yeast Ip11 and Drosophila aurora protein serine/ threonine kinase– encoding genes. It has 40% identity with the
Sac-charomyces cerevisiae Ip11 gene2and 48% identity with the
Drosoph-© 2003 American Cancer Society DOI 10.1002/cncr.11879
ila melanogaster aurora gene,2 both shown to be
functionally involved in centrosome function and nor-mal chromosome segregation.7,8 STK15 encodes a
centrosome-associated kinase and is involved in the induction of centrosome duplication-distribution ab-normalities and aneuploidy in mammalian cells.8,9
Previous studies have demonstrated that ectopic ex-pression of STK15 in mouse NIH/3T3 cells and Rat 1 fibroblasts causes centrosome amplification and transformation in vitro as well as tumorigenesis in vivo.10,11 Furthermore, overexpression of STK15 in
near-diploid human breast epithelial cells led to ab-normal centrosome numbers and induction of aneu-ploidy.11These findings suggest that STK15 is a critical
kinase-encoding gene whose overexpression leads to centrosome amplification, chromosomal instability, and transformation in mammalian cells.
Fluorescence in situ hybridization (FISH) studies have revealed that the STK15 gene is amplified in approximately 12% of primary breast tumors.11
South-ern blot hybridization results have demonstrated a 2.5– 8-fold amplification in various human tumor cell lines.10,11Increased STK15 gene expression also occurs
in the absence of gene amplification.11,12In one study,
which involved only 33 cases, STK15 was reported to be overexpressed in 94% of invasive ductal carcinomas of the breast by immunohistochemical (IHC) stain-ing.13Another study reported a somewhat lower (62%)
but still a substantial level of STK15 overexpression in human breast carcinomas by real-time polymerase chain reaction (PCR).14In the current study, we
deter-mined STK15 expression by IHC in 112 primary hu-man breast tumor specimens and explored whether
STK15 expression was a prognostic factor for breast
carcinoma by comparing STK15 expression with clin-ical parameters that are known prognostic factors for the disease.
MATERIALS AND METHODS
Patients and SpecimensArchival paraffin-embedded tissue samples were se-lected for IHC staining with the STK15 antibody. These samples were obtained from patients who had ade-quate fresh tumor samples that were available for research. All samples were obtained from female pa-tients who were diagnosed with breast carcinoma and underwent lumpectomy or mastectomy and axillary lymphadenectomy between 1986 and 1993. Clinical data from the patients were reviewed retrospectively using medical records, including written medical charts and electronic records and reports. Of the 125 patient specimens initially stained, accurate and de-tailed information was available in 112 cases. These data then were used to analyze the relation between
STK15 expression level and clinical parameters such as
clinical stage, tumor size, lymph node status, nuclear grade, recurrence-free survival rate, and overall sur-vival rate.
Immunohistochemical Staining
A rabbit polyclonal anti-STK15 antibody developed in our laboratory and raised against a carboxy-terminal (C-terminal) peptide was used for IHC staining of all primary breast tumor specimens.2,11,12The STK15
an-tibody was assayed for specificity by Western blot analyses and by IHC staining of the near-diploid non-tumorigenic breast epithelial cell line, MCF10, and the breast carcinoma cell line BT474 grown in vitro. MCF10 cells display a near-normal diploid copy num-ber and a low expression level of the STK15 gene, whereas BT474 cells reveal amplification and overex-pression of the STK15 gene.11 For the Western blot
assay, subconfluent cell monolayers were lysed in ex-traction buffer as described elsewhere.11Lysates with
approximately 100g of total cell protein were elec-trophoresed on 12% sodium dodecyl sulfate–polyac-rylamide gels in 10% Tris-glycine buffer and trans-ferred by electroblotting (Bio-Rad, Hercules, CA) to nitrocellulose membranes. The membranes were first blocked in 5% bovine calf serum (BCS) in 10 mM Tris, 150 mM NaCl, and 0.05% Tween 20 (TBST). Mem-branes then were incubated either with purified rabbit polyclonal anti-STK15 antibody (1:400 dilution in 0.5% BCS in TBST) or with the antibody preincubated with the cognate peptide against which it was raised (0.1 mg of antibody to 1 mg of peptide). The STK15 protein band was detected with a peroxidase-conjugated an-tirabbit secondary antibody (Amersham, Arlington Heights, IL) and visualized using the enhanced chemi-luminescence-Plus (Amersham) detection system. Fi-nally, Western blots were stripped and reincubated with a primary mouse monoclonal anti– actin anti-body (Amersham) diluted 1:300 in 5% BCS in TBST and with a secondary peroxidase-conjugated anti-mouse antibody (Amersham) to normalize the amount of protein loaded in each lane.
For the IHC assay of STK15 expression in the cell lines, paraffin-embedded sections of cells were stained according to the protocol described below.
The immunoperoxidase staining method used was a modification of the avidin-biotin complex tech-nique.15 In brief, paraffin-embedded tissue sections
were dewaxed and rehydrated in a graded series of alcohol solutions. These sections then were digested in 0.05% trypsin for 15 minutes, blocked in 0.3% H2O2
in methanol for 15 minutes, covered with 10 mM sodium citrate buffer (pH 6.0), heated at 95 °C for 5 minutes, and treated with 1% (v/v) normal goat serum
for 30 minutes. The slides were incubated for 3 hours at room temperature with STK15 antibody diluted 1:40. After extensive washing with phosphate-buffered saline (PBS), the slides were incubated for 60 minutes with biotinylated goat antirabbit IgG antibody diluted 1:200 in PBS. Subsequently, the slides were incubated for 60 minutes at room temperature with the avidin-biotin-peroxidase complex diluted 1:100 in PBS. The peroxidase-catalyzed product was visualized with 0.125% aminoethylcarbazole chromogen stock solu-tion (Sigma, St. Louis, MO). After counterstaining with Mayer modified hematoxylin (Sigma), the slides were dehydrated and mounted. Negative controls, in which PBS was used instead of the primary antibody, were run with each batch of staining samples. A previously identified strongly staining tumor section was used as a positive control. By including these controls with each batch of slides that were stained, interassay and intraassay consistency was maintained.
Slides were evaluated by light microscopy. Stain-ing was assessed blindly and independently by two pathologists. STK15 immunoreactivity initially was scored into five groups according to the percentage and intensity of cytoplasmic staining of the positively stained tumor cells. Specimens with ⬎ 50% of cells stained were scored as strongly positive (3⫹), those with⬎ 20–50% of cells stained were scored as mod-erately positive (2⫹), those with 20% of cells stained were scored as positive (1⫹), and those with ⬍ 20% cells stained were scored as weakly positive (⫹/⫺). Specimens with no staining were scored as negative.
Statistical Analyses
STK15 staining for each specimen was classified as either positive or negative. The tumors were classified as positive for STK15 overexpression when⬎ 20% of the cells exhibited cytoplasmic staining (i.e., speci-mens initially scored as 3⫹ or 2⫹), as conventionally has been done in other studies.12The remaining
spec-imens were classified as negative. The chi-square and Fisher exact tests were used to analyze the relation between the level of STK15 staining and the stage of disease, lymph node status, tumor size, nuclear grade, and hormone receptor status. Analyses were per-formed for overall survival (i.e., the time from initial diagnosis to death) and for recurrence-free survival (i.e., the time from initial diagnosis to recurrence of breast carcinoma). Kaplan–Meier curves for both overall survival and recurrence-free survival were plot-ted and compared. Log-rank tests were used to eval-uate the differences between the two levels of STK15 expression with regard to overall survival and recur-rence-free survival.
RESULTS
Patient Characteristics
Of the 112 patients included in the analysis, the ma-jority (79.5%) had clinical 1998 AJCC Stage II breast carcinoma at diagnosis. This group included no pa-tients with Stage IV disease at diagnosis. Receptor status was available for 65.1% of patients. After mas-tectomy, 92% of patients received chemotherapy, and 33.3% received adjuvant radiation treatment. Very few patients (7.1%) received adjuvant hormonal treat-ment. Patient characteristics are summarized in Table 1.
Immunohistochemical Staining for STK15 expression
The specificity of the STK15 antibody in the immuno-detection of cellular STK15 protein became evident from the parallel Western blot and IHC assays per-formed with cell lysates of the breast epithelial cell
TABLE 1
Patient Characteristics (nⴝ 112)
Characteristic No. of patients (%)
Age (ys) Median 46 Range 26–72 Clinical stage I 8 (7.1) II 89 (79.5) III 15 (13.4) Tumor size T1 27 (24.1) T2 69 (61.6) T3 4 (3.6) T4 3 (2.7) Unknown 9 (8.0)
No. of positive lymph nodes
Median no. of lymph nodes removed (range) 16 (5–39)
0 35 (31.2)
1–3 56 (50)
4–9 15 (13.4)
⬎ 10 5 (4.5)
Unknown 1 (0.9)
Black nuclear grade
I (poorly differentiated) 61 (54.5) II (moderately differentiated) 50 (44.6) III (well differentiated) 1 (0.9) Receptor status ER and/or PR positive 37 (33.0) ER and PR negative 36 (32.1) Unknown 39 (34.8) Previous treatmenta Chemotherapy 103 (92) Radiotherapy 37 (33.3) Hormone 8 (7.1)
ER: estrogen receptor; PR: progesterone receptor.
line, MCF10, and the breast tumor cell line BT474, which expressed low and high levels of STK15 protein, respectively (Fig. 1). Lack of any signal on Western blots with the peptide antigen–preincubated antibody unambiguously proved that the antibody recognizes the specific epitope on the STK15 protein.
Representative staining patterns by STK15 expres-sion level are shown in Figure 2. The results of STK15 staining are graphically represented in Figure 3. Al-though STK15 staining initially was scored into five different groups as previously described, staining re-sults were regrouped for the statistical analyses as either positive or negative. The positive specimens included only those that strongly (3⫹) or moderately strongly (2⫹) expressed STK15. The remaining speci-mens were classified as negative.12We believe that this
paired grouping is more appropriate for the current exploratory study than were the initial five groups, especially given the limited number of patients. The distribution of STK15 staining by clinical stage, tumor size, lymph node status, nuclear grade, and hormone receptor status is summarized in Table 2.
Using IHC to assess STK15 expression, the num-ber of tumors with positive STK15 staining was found to be lower in the current study than in a previous study13that reported 94% positivity in a study
popu-lation comprising 33 tumor samples. There are several possible explanations for this discrepancy. First, a dif-ferent antibody was used for IHC staining, and there may have been differences in the affinity and speci-ficity of the antibodies. Although we have verified the specificity of the antibody used in the current study in detecting the C-terminal peptide epitope against which the antibody was raised, the same is not known for the antibody in the previous study13 that was
raised against the glutathione-S-transferase-conju-gated full-length protein produced in bacteria.
Sec-ond, in both studies, the intensity and degree of stain-ing were measured semiquantitatively by independent pathologists. Therefore, interpretation of the level of
STK15 expression may have differed. Third, the
num-ber of tissue specimens evaluated was different. The initial investigators evaluated only 33 tumor speci-mens, whereas 112 tumor specimens were evaluated in the current study. Although less likely, another ex-planation may be found in the ethnic differences of the patients examined. All patients in the previous study were Japanese, whereas, the current study pri-marily examined Caucasian patients.
Statistical Results
When STK15 staining was compared with clinical stage, tumor size, lymph node status, and hormone receptor status, no statistically significant differences were found (Table 2). However, the results of our analysis of the relation between STK15 staining and nuclear grade were nearly statistically significant (P ⫽ 0.05). The distribution of STK15-positive tumors by Black nuclear grades contained greater proportion (62.1%) of nuclear Grades II and III. In STK15-negative tumors, the pattern was the opposite, with a larger proportion of Black nuclear Grade I tumors (60.2%).
With a median follow-up time of 79 months, only 28 (25.5%) patients experienced disease recurrence, and 27 (24.3%) patients died. The relation between STK15 staining and survival rates was analyzed, and the Kaplan–Meier curves for recurrence-free survival and overall survival by STK15 staining results are shown in Figures 4 and 5, respectively. No statistically significant differences were found with respect to re-currence-free survival rates (P⫽ 0.34) or overall sur-vival rates (P⫽ 0.42). The lack of a difference may be due to the fact that few events (i.e., either recurrence
FIGURE 1.Assessment of specificity of STK15 antibody. (A) Western blot analyses of STK15 expression in the near-diploid nontumorigenic breast epithelial cell line MCF10 and in the breast carcinoma cell line BT474 using the rabbit polyclonal anti-STK15 antibody. The specificity of the antibody is reflected in the detection of a single protein band in the top panel and by the absence of any detectable band in the middle panel, in which the antibody was blocked by the peptide antigen against which the antibody was made. The presence of-actin bands of similar intensity in the lanes in the lower panel indicate comparable amounts of protein loading from the cell lysates. (B) Immunohistochemical assay of STK15 expression in the two cell lines, using paraffin-embedded sections of cells that were stained as previously described.
or death) occurred and may reflect the early stage at diagnosis for the current group of patients.
DISCUSSION
Chromosomal rearrangements and aneuploidy are among the most notable abnormalities found in tu-mor cells.16 –19 The precise mechanisms by which
chromosomes are segregated in mitosis continue to be elucidated, but the general consensus is that bipolar
spindle formation plays an important role(s) in this process.10,20,21Because of the preponderance of
chro-mosomal abnormalities in human malignancies, the contribution to this phenotype made by genes in-volved in chromosome segregation, such as STK15, can be postulated. Evidence suggests that elevated expression of STK15 induces aneuploidy as well as oncogenic transformation in mammalian cells. STK15 transforms both rat fibroblast and mouse NIH3T3 cells
FIGURE 2. Immunohistochemical
staining of primary breast tumors for STK15. (A) Strongly positive staining (3⫹). (B) Moderately positive staining (2⫹). (C) Positive staining (1⫹). (D) Neg-ative staining.
in vitro,10,11and STK15-transformed cells grow as
tu-mors in nude mice.10In human cells, STK15 has been
shown to play a role in chromosome segregation and is associated with tumorigenesis.1–3,10,11,22–26 Given
these data, it is appealing to propose that expression of STK15 in tumors plays a role in the pathogenesis and/or prognosis of human malignancies such as breast carcinoma.
As an initial step to determining whether STK15
may be involved in breast carcinoma, we assessed the level of STK15 expression in primary breast tumors. Although expression of STK15 has been evaluated in human breast carcinoma cell lines,2,11 only two other
studies13,14 have evaluated STK15 expression in
hu-man breast carcinoma tissue specimens. Tanaka et al.13evaluated STK15 expression in human breast
car-cinoma tissue by IHC, and Miyoshi et al.4 assessed
mRNA expression using real-time PCR. Both studies noted STK15 expression in a large proportion of breast tumors, especially with IHC staining. However, the number of patients examined was quite limited (33 and 47, respectively). Tanaka et al.13 did not report
whether STK15 expression was correlated with any known prognostic factors for breast carcinoma. To the best of our knowledge, the current study is the first to evaluate STK15 expression in primary breast carci-noma in an adequate number of patients. We also explored the relation between STK15 expression and clinical parameters that are prognostic factors for this disease.
Given the function of STK15, it would be expected
FIGURE 3.Distribution of STK15 staining (n⫽ 112). TABLE 2
Distribution of STK15 Staining by Clinical Parameters
Parameter
No. of patients by STK15 staining (%) Positive Negative Total
Clinical stage I 3 (10.3) 5 (6.0) 8 (7.1) II 22 (75.9) 67 (80.7) 89 (79.5) III 4 (13.8) 11 (13.3) 15 (13.4) Tumor sizea T1 7 (25.9) 20 (26.3) 27 (26.2) T2 18 (66.7) 51 (67.1) 69 (67.0) T3 2 (7.4) 2 (2.6) 4 (3.9) T4 0 (0) 3 (3.9) 3 (2.9)
Lymph node statusb
Positive 20 (69.0) 56 (67.5) 76 (67.9) Negative 9 (31.0) 27 (32.5) 36 (32.1) Black nuclear grade
I (poorly differentiated) 11 (37.9) 50 (60.2) 61 (54.5) II (moderately/well differentiated) 18 (62.1) 33 (39.8) 51 (45.5) Hormone receptor status
ER and/or PR positive 14 (63.6) 23 (45.1) 37 (50.7) ER and PR negative 8 (36.4) 28 (54.9) 36 (49.3)
ER: estrogen receptor; PR: progesterone receptor.
aNine patients had unknown tumor size and were not included in the analysis.
bFor the statistical analyses, lymph node status was reclassified as either positive or negative. The
distribution of patients by number of positive lymph nodes is shown in Table 1.
FIGURE 4.Recurrence-free survival by STK15 expression (P⫽ 0.34).
that higher levels of STK15 expression would be cor-related with aggressive clinical behavior,12,14,27such as
poorly differentiated tumors. Miyoshi et al.14 found
that histologic Grade III and hormone receptor–nega-tive breast tumors had higher STK15 mRNA expression levels. However, no correlation was found with lymph node status, one of the strongest prognostic factors in breast carcinoma. It is noteworthy that in the current study, a trend toward a more differentiated nuclear grade with positive STK15 staining was observed. It is not clear why this correlation is the opposite of what was published earlier.14It is possible that this is simply
a function of the tumors evaluated, which were ran-domly selected and retrospectively analyzed. There is a preponderance of poorly differentiated tumors in this sample of patients, which is not reflective of the usual observations in early-stage breast carcinoma. Alternatively, it is entirely plausible that the function of STK15 is regulated tightly by its expression level, such that both underexpression and overexpression lead to its aberrant function and result in chromo-somal aneuploidy associated with malignant transfor-mation and progression processes. Both overexpres-sion and inactivation of STK15 kinase have been shown to cause similar mitotic defects leading to chro-mosomal ploidy alterations in cells in vitro.28The lack
of STK15 expression leads to large, aberrant-looking nuclei, suggesting that nuclear atypia may result from the lack of STK15 expression (unpublished data).
That a correlation could not be established be-tween STK15 expression and other clinical parameters evaluated is disappointing, but several possible expla-nations exist. First, analysis of various STK15 staining intensities may be more useful than a simple positive or negative grouping. Although we initially attempted to perform, this analysis, the resulting subgroups were so small that no differences were detected and no correlations were found. However, a study involving a larger population of patients that has the power to detect even small differences may produce interesting results and would be worthwhile to undertake. Sec-ond, IHC analysis of cytoplasmic STK15 may not be the best way to evaluate STK15 expression levels.29
Instead, gene amplification and mRNA expression lev-els may be better correlated with prognosis. For in-stance, HER-2/neu, which is overexpressed in approx-imately 20 –30% of breast carcinomas, may be positive by IHC studies but show no gene amplification.30 –33
We plan to evaluate STK15 gene amplification by FISH to determine whether there is a discrepancy between FISH and IHC staining. Third, it could be postulated that STK15-positive patients do have a poor prognosis but that adjuvant treatment alters the natural history of the disease, compensating for poor prognosis. In
that case, fewer disease recurrences or deaths would occur and no differences between the two groups would be detected in either recurrence-free or overall survival rate.
It is more likely, however, that the patients exam-ined in the current study, the majority of whom were diagnosed at an early stage, already have a good prog-nosis and that a relatively small number of disease recurrences or deaths are expected. A small number of events is unlikely to allow the detection of differences between the two groups, except for the most apparent ones. A correlation may still exist, but the current study is underpowered to detect it. It would be inter-esting to determine whether any correlation exists between STK15 expression and survival in patients with more advanced stages of disease, as many more events are expected in that group of patients. Alterna-tively, the expression of STK15 could be evaluated in preinvasive disease, such as ductal carcinoma in situ (DCIS), and the risk of development of invasive ductal carcinoma could be stratified by STK15 staining re-sults. Preliminary data from our laboratory indicate that high levels of STK15 expression are detected in a significant number of DCIS specimens. We are explor-ing the biologic relevance of this observation. It is relevant in this context to mention that in a rat mam-mary tumor model system, overexpression of the STK15 kinase recently has been demonstrated to be an early event in the carcinogenic transformation pro-cess.34 Thus, it is possible that STK15 kinase
overex-pression may be an early marker of initiation rather than progression of mammary tumorigenesis.
We recognize that the current study is an explor-atory one and that definitive conclusions will have to be drawn from prospectively designed studies with more patients. The current study is nonetheless worthwhile, as it could provide clues for generating hypotheses and designing future studies to elucidate the physiologic targets of STK15 and its potential role in the pathogenesis of breast carcinoma. Furthermore, we have demonstrated that it is possible to measure
STK15 expression in breast tumors and that this
vari-able should be included in future correlative studies, such as those involving microarray analysis.
REFERENCES
1. Isola JJ, Kallioniemi OP, Chu LW, et al. Genetic aberrations detected by comparative genomic hybridization predict out-come in node-negative breast cancer. Am J Pathol. 1995;147: 905–911.
2. Sen S, Zhou H, White RA. A putative serine/threonine kinase encoding gene BTAK on chromosome 20q13 is amplified and overexpressed in human breast cancer cell lines.
3. Tanner MM, Tirkkonen M, Kallioniemi A, et al. Increased copy number at 20q13 in breast cancer: defining the critical region and exclusion of candidate genes. Cancer Res. 1994; 54:4257– 4260.
4. Tanner MM, Tirkkonen M, Kallioniemi A, et al. Amplifica-tion of chromosomal region 20q13 in invasive breast cancer: prognostic implications. Clin Cancer Res. 1995;1:1455–1461. 5. Tanner MM, Tirkkonen M, Kallioniemi A, et al. Independent amplification and frequent co-amplification of three non-syntenic regions on the long arm of chromosome 20 in human breast cancer. Cancer Res. 1996;56:3441–3445. 6. Nigg EA. Mitotic kinases as regulators of cell division and its
checkpoints. Nat Rev Mol Cell Biol. 2001;2:21–32.
7. Francisco L, Wang W, Chan CS. Type 1 protein phosphatase acts in opposition to IpL1 protein kinase in regulating yeast chromosome segregation. Mol Cell Biol. 1994;14:4731– 4740. 8. Glover DM, Leibowitz MH, McLean DA, Parry H. Mutations in aurora prevent centrosome separation, leading to the formation of monopolar spindles. Cell. 1995;81:95–105. 9. Chan CS, Botstein D. Isolation and characterization of
chro-mosome-gain and increase— in-ploidy mutants in yeast.
Genetics. 1993;135:677– 691.
10. Bischoff JR, Anderson L, Zhu Y, et al. A homologue of Dro-sophila aurora kinase is oncogenic and amplified in human colorectal cancers. EMBO J. 1998;17:3052–3065.
11. Zhou H, Kuang J, Zhong L, et al. Tumor amplified kinase STK15/BTAK induces centrosome amplification, aneuploidy and transformation. Nat Genet. 1998;20:189 –193.
12. Sen S, Shou H, Zhang RD, et al. Amplification/overexpres-sion of a mitotic kinase gene in human bladder cancer.
J Natl Cancer Inst. 2002;94:1320 –1329.
13. Tanaka T, Kimura M, Matsunaga K, Fukada D, Mori H, Okano Y. Centrosomal kinase AIK1 is overexpressed in in-vasive ductal carcinoma of the breast. Cancer Res. 1999;59: 2041–2044.
14. Miyoshi Y, Iwao K, Egawa C, Noguchi S. Association of centrosomal kinase STK15/BTAK mRNA expression with chromosomal instability in human breast cancers. Int J
Can-cer. 2001;92:370 –373.
15. Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase technique: a compar-ison between ABC and unlabeled antibody (PAP) proce-dures. J Histochem Cytochem. 1981;29:577–580.
16. Shackney SE, Shankey TV. Common patterns of genetic evolution in human solid tumors. Cytometry. 1997;29:1–27. 17. Loeb LA. Mutator phenotype may be required for multistage
carcinogenesis. Cancer Res. 1991;51:3075–3079.
18. Lengauer C, Kinzler KW, Vogelstein B. Genetic instability in colorectal cancers. Nature. 1997;386:623– 627.
19. Pihan GA, Purohit A, Wallace J, et al. Centrosome defects and genetic instability in malignant tumors. Cancer Res. 1998;58:3974 –3985.
20. Kimura M, Kotani S, Hattori T, et al. Cell cycle-dependent expression and spindle pole localization of a novel human protein kinase, AIK, related to aurora of Drosophila and yeast Ip11. J Biol Chem. 1997;272:13766 –13771.
21. Roghi C, Giet R, Uzbekov R, et al. The Xenopus protein kinase pEg2 associates with the centrosome in a cell cycle-dependent manner, binds to the spindle microtubules and is involved in bipolar mitotic spindle assembly. J Cell Sci. 1998;111:557–572.
22. Muleris M, Salmon RJ, Dutrillauz AM, et al. Characteristic chromosomal imbalances in 18 near-diploid colorectal tu-mors. Cancer Genet Cytogenet. 1987;29:289 –301.
23. Kallioneimi A, Kallioniemi OP, Piper J, et al. Detection and mapping of amplified DNA sequences in breast cancer by comparative genomic hybridization. Proc Natl Acad Sci
U S A. 1994;91:2156 –2160.
24. James LA, Mitchell EL, Menasce L, Varley JM. Comparative genomic hybridization of ductal carcinoma in situ of the breast: identification of regions of DNA amplification and deletion in common with invasive breast carcinoma.
Onco-gene. 1997;14:1059 –1065.
25. Reznikoff CA, Belair CD, Yeager TR, et al. A molecular ge-netic model of human bladder cancer pathogenesis. Semin
Oncol. 1996;23:571–584.
26. Courjal F, Cuny M, Rodriguez C, et al. DNA amplifications at 20q13 and MDM2 define distinct subsets of evolved breast and ovarian tumours. Br J Cancer. 1996;74:1984 –1989. 27. Tanner MM, Grenman S, Koul A, et al. Frequent
amplifica-tion of chromosomal region 20q12-q13 in ovarian cancer.
Clin Cancer Res. 2000;6:1833–1839.
28. Meraldi P, Honda R, Nigg A. Aurora-A overexpression re-veals tetraploidization as a major route to centrosome am-plification in p53⫺/⫺cells. EMBO J. 2002;21:483– 492. 29. Shin SO, Lee KH, Kim JH, et al. Alternative splicing in 5
⬘-untranslated region of STK15 gene, encoding centrosome associated kinase, in breast cancer cell lines. Exp Mol Med. 2000;32:193–196.
30. Pauletti G, Dandekar S, Rong H, et al. Assessment of meth-ods for tissue-based detection of the HER-2/neu alteration in human breast cancer: a direct comparison of fluores-cence in situ hybridization and immunohistochemistry.
J Clin Oncol. 2000;18:3651–3664.
31. Kakar S, Puangsuvan N, Stevens JM, et al. HER-2/neu as-sessment in breast cancer by immunohistochemistry and fluorescence in situ hybridization: comparison of results and correlation with survival. Mol Diagn. 2000;5:199 –207. 32. Ridolfi RL, Jamehdor MR, Arber JM. HER-2/neu testing in
breast carcinoma: a combined immunohistochemical and fluorescence in situ hybridization approach. Mod Pathol. 2000;13:866 – 873.
33. Wang S, Saboorian MH, Frenkel E, Hynan L, Gokaslan ST, Ashfaq R. Laboratory assessment of the status of Her-2/neu protein and oncogene in breast cancer specimens: compar-ison of immunohistochemistry assay with fluorescence in situ hybridization assays. J Clin Pathol. 2000;53:374 –381. 34. Goepfert TM, Adigun YE, Zhong L, Gay J, Medina D, Brinkley
WR. Centrosome amplification and overexpression of Aurora-A are early events in rat mammary carcinogenesis.
