SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Aldehyde dehydrogenase 1 (ALDH1) is expressed in stem/progenitor cells, including cancer-initiating cells (CIC) of various organs. In the present study, ALDH1 expression was immunohistochemically examined in uterine endometrioid adenocarcinoma. The ALDH1 was expressed in a small portion of tumor cells, and these ALDH1-expressing cells were less mature than ALDH1-non-expressing cells. The ALDH1-expressing (ALDH1-hi) cells were more tumorigenic, resistant to anti-cancer agents and more invasive than ALDH1-lo cells. Culture of the sorted ALDH1-hi cells yielded both ALDH1-hi and ALDH1-lo cells, whereas ALDH1-lo cells yielded ALDH-lo cells alone. Clinically, a high-level of ALDH1 expression in tumor cells was correlated with T category, lymphatic invasion, recurrence and prognosis of patients. Patients with high ALDH1 expression showed poorer prognoses than those with low expression (P = 0.015 for disease-free survival [DFS] and P = 0.010 for overall survival [OS]), and high ALDH1 expression was an independent factor for poor prognosis. Aldehyde dehydrogenase 1 is a candidate for CIC marker for uterine endometrioid adenocarcinoma. (Cancer Sci 2011; 102: 903–908)

Tumors consist of heterogeneous cell populations derived from a single clone. Recently, it has been demonstrated that cells with tumorigenic potential are limited to a small population among tumor cells, called cancer-initiating cells (CIC), in cancers of blood (leukemia), breast, brain and colon.(1–11) Cancer-initiating cells efficiently efflux anti-tumor agents and degrade reactive oxygen species that are related to radiation-induced apoptosis. Furthermore, CIC are in a quiescent state for cell division, and thus escape the attack of various anti-cancer drugs targeting the rapidly dividing tumor cells. These characteristics enable CIC to be resistant to anti-tumor drugs and radiation therapy.(12–15)

Endometrioid adenocarcinoma is one of the most common malignancies of the female genital system.(16,17) Despite the advances in methods for detection and treatment, prognosis of patients with endometrioid adenocarcinoma still remains unfavorable. Therapeutic strategies targeting CIC would be necessary to improve cure rates, but studies on CIC of endometrioid adenocarcinoma are limited. Gotte et al.(18) demonstrated that Musashi-1, highly expressed in neural stem cells, was co-expressed with Notch-1 in a subpopulation of endometrial cells and endometrioid adenocarcinoma cells. Kato et al.(19) demonstrated that the side-population of endometrioid adenocarcinoma, which is considered to contain CIC, possessed higher tumorigenic activities than non side-population cells. To our knowledge, the relationship of stem cell marker expression to prognosis has not been reported in endometrioid adenocarcinoma.

Aldehyde dehydrogenase 1 (ALDH1), a predominant isoform of the ALDH family in mammals, oxidizes retinol to retinoic acid in early stages of stem cell differentiation, and hematopoietic and neural stem cells show high ALDH1 activity.(20–22) Cancer-inducing cells of human multiple myeloma, acute myeloid leukemia and cancers of brain, lung and breast also show high ALDH1 activity.(23–27) The activity of ALDH1 might be a common marker for both normal and malignant stem cell populations. In the present study, ALDH1 expression was immunohistochemically examined in endometrioid adenocarcinoma and its clinical implications were evaluated.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Patients.  Ninety-eight patients who underwent surgery for uterine endometrioid adenocarcinoma at Osaka University Hospital from January 1998 to January 2007 were examined. Clinicopathological findings in these 98 patients are summarized in Table 1. The age of patients ranged from 22 to 75 years (median, 55.9 years). Resected specimens were macroscopically examined to determine the location and size of the tumors. Histological specimens were fixed in 10% formalin and routinely processed for paraffin embedding. Paraffin-embedded specimens were stored in the dark room in the Department of Pathology of Osaka University Hospital at room temperature, sectioned at 4-μm thickness at the time of staining, and stained with H&E and immunoperoxidase procedures. The histological stage was determined according to the International Federation of Obstetricians and Gynecologists (FIGO) staging system. All patients were followed up with laboratory examinations including routine peripheral blood cell counts at 1- to 6-month intervals, X-ray, computed tomographic scan and pelvic examination at 6- to 12-month intervals. The follow-up period for survivors ranged from 8 to 122 months (median, 89 months). The study was approved by the ethical review board of the Graduate School of Medicine, Osaka University.

Table 1.   Summary of characteristics in 98 endometrioid adenocarcinoma patients
 Number of patients
Tumor
 T170
 T28
 T320
Lymph node
 N073
 N125
Tumor histological grade
 Grade 138
 Grade 239
 Grade 321
Estrogen receptor status
 Positive40
 Negative58
Progesterone receptor status
 Positive75
 Negative23
Ki67 labeling index
 ≥ 20%82
 <20%16
Response to chemotherapy
 Non-respond17
 Respond32
Recurrence
 Positive20
 Negative78
Prognosis
 Dead15
 Alive 83

Immunohistochemistry for ALDH1, ER, PgR, CD9 and MIB-1.  Expression of ALDH1, estrogen receptor (ER), progesterone receptor (PgR) and Cluster Differentiation (CD) was examined with anti-ALDH1 (BD Biosciences, Franklin Lakes, NJ, USA), anti-ER (Dako, Glostrup, Denmark), anti-PgR (Dako) and anti-CD9 (Abcam Ltd, Cambridge, UK) antibodies, respectively. The proliferative activity of cancer cells was examined with monoclonal antibody MIB-1 (Immunotech, Marseilles, France), recognizing the proliferation-associated antigen Ki67. The antigen retrieval with Pascal pressurized heating chamber (Dako) was done for the staining of ER, PgR, CD9 and MIB-1. The sections were incubated with anti-ALDH1 (×100), -ER (×2), -PgR (×6), -CD9 antibody (×100) or MIB-1 (×50), then treated with a ChemMate EnVision kit (Dako). Diaminobenzidine (DAB) (Dako) was used as a chromogen. As the negative control, staining was carried out in the absence of primary antibody. Stained sections were evaluated independently by two pathologists (JI, EM). As described previously,(24) cases with more and <10% of cells positive for ALDH1 were regarded as ALDH1-hi and ALDH1-lo, respectively. The proportion and intensity of ER and PgR expression were evaluated as described previously.(28) The MIB-1 labeling index was defined as the percentage of stained nuclei per 1000 cells. The patients were divided into MIB-1-high and MIB-1-low groups using the median as cut-off value.

Double staining of ALDH1 and CD9, ER and PgR.  Double staining of ALDH1 and CD9, ER and PgR was done with the EnVision G/2 doublestain system (Dako) according to the manufacturer’s protocol. First, the ALDH1 was stained with DAB, and subsequently the staining of CD9, ER and PgR was done with Permanent Red. Since the red fluorescence is released from Permanent Red, the signal of CD9, ER and PgR was detected with a fluorescence microscope (Biozero, Keyence, Osaka, Japan).

Cell lines and isolation of ALDH1-hi population.  Endometrioid adenocarcinoma cell lines HEC-1, -1A, -108, -116, -6, -88nu and -251; and SNG-M and -II were obtained from the Health Science Research Resources Bank of Osaka, Japan. Cells were cultured in DMEM (Wako, Osaka, Japan) supplemented with 10% FCS (Nippon Bio-Supply Center, Tokyo, Japan). To isolate the population with high ALDH1 enzymatic activity, the Aldefluor kit (Stem Cell Technologies, Vancouver, BC, Canada) was used according to the manufacturer’s instructions. Briefly, cells were suspended in Aldefluor assay buffer containing ALDH1 substrate and BODIPY–aminoacetaldehyde (BAAA). The BAAA was taken up by living cells and converted by intracellular ALDH1 into BODIPY–aminoacetate, which causes the cells to fluoresce brightly. The brightly fluorescent ALDH1-expressing cells were detected with FACS Calibur or FACS Aria (BD Biosciences). As a negative control, cells were stained under identical conditions with the specific ALDH1 inhibitor, diethylaminobenzaldehyde (DEAB; Sigma, St Louis, MO, USA). Data were analyzed by using Cell Quest software (BD Biosciences). In endometrioid adenocarcinoma cell lines, cells with bright fluorescence were judged as ALDH-hi, and those with no or faint fluorescence as ALDH-lo. The criteria for ALDH1-hi and ALDH1-lo in cell lines were different from those for ALDH1 immunohistochemistry in clinical samples.

Effects of the anticancer drug cisplatin.  Cisplatin is commonly used for the treatment of endometrioid adenocarcinoma. The effect of cisplatin on ALDH-hi cells was compared to that on ALDH-lo cells. Cells (1 × 104) were seeded onto cell culture plates with DMEM-10% FBS, cultured for 20 h, and various concentrations of cisplatin (0, 1, 4, 8 μg/mL) were added. After 24 h, the viability of cells was assessed with the Premix WST-1 cell assay system (Takara Bio Inc., Kyoto, Japan). The absorbance of cisplatin-treated cells at 450 nm was subtracted from the background absorbance (600 nm). The resultant value was divided by that of cells not treated with cisplatin, and the results are shown as the viability index.

Matrigel invasion assay.  Invasion of tumor cells into Matrigel was examined with a BD BioCoat Matrigel Invasion Chamber (BD Biosciences). Briefly, cells were seeded in DMEM without FCS in the Matrigel invasion upper chamber and cultured for 72 h. The lower chamber contained DMEM and 10% FBS. Invading cells were stained with a Diff-quick staining kit (Siemens, Munich, Germany). The number of invading cells was counted in four microscopic fields per well at a magnification of ×20 and the extent of invasion was expressed as the average number of cells per square millimeter.

In vitro colony formation assay.  Cells were suspended in 0.1 mL of DMEM and 10% FBS, and 1000 cells were plated in culture dishes with 1 mL of methylcellulose-containing DMEM supplemented with 15% FBS. The number of colonies was counted on day 14.

Statistical analysis.  Statistical analyses were performed using StatView software (SAS Institute Inc., Cary, NC, USA). The Chi-square and Fisher’s exact probability tests were used to analyze the correlation between ALDH1 expression and clinicopathological factors in endometrioid adenocarcinoma. Kaplan–Meier methods were used to calculate overall survival (OS) and disease-free survival (DFS) rates, and differences in survival curves were evaluated with the log-rank test. Cox’s proportional hazards regression model with a stepwise manner was used to analyze the independent prognostic factors. The P values of <0.05 were considered to be statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Immunohistochemical findings.  The expression of ALDH1 was examined in normal endometrium and 98 endometrioid adenocarcinoma tissues. No signals were detected in normal proliferating and secretory phase of endometrium (Fig. 1a,b, respectively). Strong signals were found in the cytoplasm of a small portion of tumor cells in several cases (Fig. 1c), whereas signals were found in most tumor cells in other cases (Fig. 1d). The expression of ALDH1 was not detected in some cases. Based on the criteria of Jiang et al.,(24) cases with more than 10% of cells positive for ALDH1 were regarded as ALDH1-hi: 40 (40.8%) of 98 cases were categorized as ALDH1-hi, and the remaining as ALDH1-lo.

image

Figure 1.  Expression of ALDH1 in normal endometrium and endometrioid adenocarcinoma. Normal proliferative (a) and secretory (b) phases of endometrial epithels did not express ALDH1. (c) Small portions of endometrioid adenocarcinoma were positive for ALDH1 in some cases. (d) Most tumor cells were positive for ALDH1. (e) Double staining of ALDH1 for CD9, ER or PgR was done. Red fluorescence from CD9, ER or PgR staining signals is shown at left. Scale lines, (a–d) 200 μm, (e) 20 μm.

Download figure to PowerPoint

Expression of CD9, ER and PgR is one of the differentiation markers of endometrium. The expression level of CD9, ER and PgR was significantly lower in ALDH1-expressing cells than in non-expressing cells (Fig. 1c).

ALDH1 activity in endometrioid adenocarcinoma cell lines.  ALDH1 activity was examined with Aldefluor assay in nine endometrioid adenocarcinoma cell lines: HEC-1, -1A, -108, -116, -6, -88nu and SNG-M contained ALDH-hi population, whereas HEC-251 and SNG-II did not (Fig. 2). In the subsequent experiments, HEC-1, which proliferates rapidly and is easy to handle, was used as a representative endometrioid adenocarcinoma cell line containing an ALDH-hi population.

image

Figure 2.  Activity of ALDH1 in various endometrioid adenocarcinoma cell lines. The left side of each cell line shows dot–blot of Aldefluor assay with inhibitor (DEAB), and the right side shows dot–blot without inhibitor. The ALDH1-hi population is boxed.

Download figure to PowerPoint

Cancer-inducing cells are known to yield both CIC and non-CIC, whereas non-CIC does not yield any CIC. To examine whether ALDH1 could be used as a CIC marker for endometrioid adenocarcinoma, ALDH1-hi and ALDH1-lo HEC-1 were sorted separately. After culture for 5 days, cells derived from sorted ALDH1-hi HEC-1 yielded both ALDH-hi and ALDH-lo cells, whereas few ALDH1-hi cells were detected in cells derived from ALDH1-lo cells (Fig. 3).

image

Figure 3.  ALDH1 activity of ALDH-hi and ALDH-lo HEC-1 cells after culture for 5 days. Dot–blot of Aldefluor assay without inhibitor is shown in the left side, and that with inhibitor on the right side.

Download figure to PowerPoint

Comparison of ALDH1-hi cells to ALDH1-lo cells in resistance against anti-tumor drug, abilities of invasion and in vitro colony formation.  As cisplatin is commonly used for the treatment of endometrioid adenocarcinoma, the effect of ALDH1 on resistance of HEC-1 to cisplatin was examined. The ALDH-hi and ALDH-lo cells were cultured for 20 h, and cisplatin was added. Then, viability was examined at 44 h after sorting. Since 5-day culture of ALDH-hi cells yielded both ALDH-hi and ALDH-lo as described above, there is a possibility that the ALDH-hi cells yielded significant number of ALDH-lo cells at the time of cisplatin addition and viability check. However, this possibility was unlikely, because the percentage of sorted ALDH-hi cells remained at similar levels at 20 and 44 h to that just after sorting (Fig. 4a). The ALDH1-lo cells were more vulnerable to cisplatin than ALDH-hi cells (Fig. 4b). The invasion ability of ALDH-hi cells was compared to that of ALDH1-lo cells with Matrigel invasion assay: the number of invading cells was lower in ALDH1-lo cells than in ALDH1-hi cells, indicating that ALDH1-hi cells possessed stronger invasive capability than ALDH1-lo cells (Fig. 4c). Next, the ability for in vitro colony formation was evaluated. In comparison to ALDH1-hi cells, ALDH1-lo cells formed fewer colonies in vitro (Fig. 4d).

image

Figure 4.  Effects of ALDH1 on resistance to the anticancer drugs, cell invasion activity and in vitro colony formation activity. (a) Percentage of ALDH1-hi population in the sorted cells at 0, 20 and 44 h after sorting. (b) Viabilities of ALDH1-hi and ALDH1-lo HEC-1 cells compared in the presence of various amounts of cisplatin. (c) Matrigel invasion assay. HEC-1 cells invade through Matrigel (×40), and the number of invading cells per square millimeter are shown. (d) Comparison of colony number derived from the ALDH1-hi and ALDH1-lo HEC-1 cells per square millimeter. The values are the means ± SE of three experiments. *P < 0.05 (Student’s t-test).

Download figure to PowerPoint

Correlation of ALDH1 expression with clinical variables.  Correlation of ALDH1 expression with clinicopathological features was evaluated. Positive correlation was observed between ALDH1 expression and T factor (= 0.047), lymph node metastasis (= 0.002), resistance to chemotherapy = 0.009), relapse rate (= 0.014) and poor prognosis (= 0.027). Other parameters including tumor histological grade, ER, PgR and Ki67 labeling index did not correlate with ALDH1 expression (Table 2). The 5-year DFS and OS were 86.7% and 90.6%, respectively. Tumors recurred in 22 patients. Of these, 14 patients died due to the tumors. There was a statistically significant difference in DFS (= 0.015) and OS rates (= 0.010) between patients with ALDH1-hi and ALDH1-lo tumors (Fig. 5).

Table 2.   Correlation between ALDH1 expression in adenocarcinoma and clinicopathological parameters
 ALDH1 expression in cancerP
LowHigh
Tumor
 T14327 
 T253 
 T310100.047
Lymph node
 N05023 
 N18170.002
Tumor histological grade
 Grade 12810 
 Grade 21920 
 Grade 311100.065
Estrogen receptor status
 Positive1030 
 Negative11470.474
Progesterone receptor status
 Positive4827 
 Negative10130.080
Ki67 labeling index
 ≥20%4735 
 <20%1150.499
Response to chemotherapy
 Non-respond413 
 Respond20120.009
Recurrence
 Positive713 
 Negative51270.014
Prognosis
 Dead510 
 Alive53300.027
image

Figure 5.  Kaplan–Meier plots. Disease-free (a) and overall (b) survival curves are shown. The ALDH1-hi cases showed less favorable disease-free survival and overall survival.

Download figure to PowerPoint

Univariate analysis showed that T factor, lymph node metastasis, tumor histological grade and ALDH1 expression were significant factors for OS. For DFS, T factor, lymph node metastasis, tumor histological grade, PgR expression and ALDH1 expression were significant factors (Table 3). Multivariate analysis revealed that ALDH1 expression, lymph node metastasis and tumor histological grade were independent prognostic factors for OS, and ALDH1 expression was an independent prognostic factor for DFS (Table 3).

Table 3.   Univariate and multivariate analyses of prognostic factors for overall and disease-free survivals
 Overall survivalDisease-free survival
UnivariateMultivariateUnivariateMultivariate
HR (95% CI)P valueHR (95% CI)P valueHR (95% CI)P valueHR (95% CI)P value
  1. CI, confidence interval; HR, hazard ratio.

Tumor3.32 (1.58–7.00)0.0021.59 (0.66–3.80)0.3023.29 (1.58–6.83)0.0021.49 (0.61–3.61)0.383
Lymph node2.64 (1.66–4.21)<0.0012.19 (1.21–3.97)0.0101.59 (1.26–2.00)<0.0011.29 (0.97–1.70)0.077
Tumor histological grade3.32 (1.58–7.00)0.0022.93 (1.11–7.74)0.0293.29 (1.58–6.83)0.0022.43 (0.97–6.12)0.060
Estrogen receptor status0.68 (0.23–2.00)0.482  0.64 (0.22–1.89)0.421  
Progesterone receptor status0.40 (0.14–1.10)0.076  0.36 (0.13–0.99)0.0480.44 (0.15–1.28)0.132
Ki67 labeling index2.56 (0.34–19.5)0.364  2.85 (0.38–21.7)0.311  
ALDH1 expression3.78 (1.28–11.2)0.0164.89 (1.37–17.5)0.0143.51 (1.19–10.4)0.0233.65 (1.03–13.0)0.045

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Normal stem/progenitor cells of various lineages, such as hematopoietic, neural and mesenchymal stem cells, show high ALDH1 activity.(20–22) In addition, CIC have been reported to show high ALDH1 activity: the ALDH1-hi population is tumorigenic and resistant to chemotherapy in cancers of colon, breast, lung, pancreas, bladder, prostate and ovary.(23–27) To our knowledge, the role of ALDH1 in uterine cancer has never been studied. In the present study, we showed that ALDH1-hi endometrioid adenocarcinoma cells to be more tumorigenic, resistant to anti-cancer agents and invasive than ALDH-lo cells. Culture of the sorted ALDH-hi cells yielded both ALDH-hi and ALDH-lo cells, whereas culture of the ALDH-lo cells yielded ALDH-lo cells alone. These findings suggest that the ALDH-hi population possessed the character of CIC in endometrioid adenocarcinoma of uterus, like cancers of other organs.

In clinical specimens, ALDH1-expressing tumor cells were mostly negative for CD9, ER and PgR, indicating that ALDH1 expression was detected in tumor cells of a less mature state. Since CIC are in the immature state, this was consistent with the notion that ALDH1 was expressed in cells with CIC character.

The expression of ALDH1 was limited to a small portion of endometrioid adenocarcinoma cells, which were randomly located in the tumor tissues. Cells with CIC character in squamous cell carcinoma are reported to be located in the outer layer of cancer nests,(29,30) contrasting with the absence of any specific location of CIC in endometrioid adenocarcinoma.

Diffuse expression of ALDH1 was found in some clinical cases of endometrioid adenocarcinoma (Fig. 1d). This appeared to be incompatible with the concept that CIC comprise a small population of cancer cells with multiple differentiations and long-term repopulation capabilities. In several reports, CIC markers were expressed in most tumor cells in clinical samples, such as ALDH1 in breast cancers.(20,31) When most tumor cells possess CIC character, the tumor character might become aggressive. Alternatively, ALDH1 might be a marker of undifferentiated cancer cells but not a CIC marker.

The clinical implication of ALDH1 expression was evaluated in 98 cases of endometrioid adenocarcinoma. The characteristics of patients, such as age and stage, in the current study were similar to those in a previous report,(32) indicating that the results obtained from the current study are commonly applicable to endometrioid adenocarcinoma worldwide. The present study showed that a high level of ALDH1 expression was correlated with T category, lymphatic invasion, resistance to chemotherapy, recurrence, and prognosis of patients. Patients with higher ALDH1 expression showed poorer prognoses than those with lower expression (P = 0.015 for DFS and P = 0.010 for OS), and high ALDH1 expression was an independent poor prognostic factor. These findings were consistent with the previous observation that a high percentage of ALDH1-expressing cells in most types of epithelial tumors, such as breast, lung, pancreatic, bladder, ovarian and prostate, is associated with a poorer outcome of these patients.(23–27) Thus, ALDH1 might be a common marker for CIC among cancers of various organs.

Endometrioid adenocarcinoma is the most common invasive malignancy of the female genital system, and novel therapeutic strategies targeting CIC would be necessary to improve cure rate. Very recently, Yang et al.(33) reported that LIN28 positively and let-7 negatively regulates ALDH1 expression in breast and ovarian cancers, and suggested that targeting ALDH1 expression via a LIN28/let-7 axis by small chemical compounds could be a therapeutic modality. Then, ALDH1 would be an effective target for therapies to CIC not only in breast and ovarian cancers but also in endometrioid adenocarcinoma of the uterus. Further studies on ALDH1 regulation may open a new therapeutic modality for endometrioid adenocarcinoma.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The authors thank Ms. Megumi Sugano, Ms. Etsuko Maeno and Ms. Takako Sawamura for their technical assistance. This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology, Japan (No. 20590364, No. 20014010).

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References