An invasive ductal carcinoma TMA (cat. no., OD-CT-RpBre01-006; Outdo Biotech Co., Ltd., Shanghai, China) was performed. The tissue microarray contained tissues from 82 cases, as 328 specimen cores (each with a diameter of 1 mm and a height of 4 µm), and the integrity of the microarray was >95%. For all 82 cases, each case included two cores of invasive ductal carcinoma and two cores of paired adjacent non-tumorous breast tissues. All clinical and pathological data of the specimens were provided by Outdo Biotech Co., Ltd., including age, tumor size, whether any axillary lymph node metastasis was present, pathological grade according to the World Health Organization 2012 classification of breast tumors (17) and the staining results of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor-2 (HER2) and Ki-67. A total of 48 cases were classified as grade 2 and 34 cases were grade 3. Overall, 37 cases demonstrated lymph node metastasis and 45 cases demonstrated no lymph node metastasis. The age range of patients was 33–88 years (median, 54.5 years) and all patients were female.

The microarray was washed in 100% xylene (Beijing Chemical Works, Beijing, China) to remove the paraffin and then rehydrated through serial dilutions of alcohol (100, 90, 80 and 70%; Beijing Chemical Works) followed by rinsing in water. This was followed by the quenching of endogenous peroxidase activity using a 0.3% solution of hydrogen peroxide (Beijing Chemical Works) in methanol for 30 min. For antigen retrieval, the section was boiled in 0.01 mol/l sodium citrate buffer (pH 6.0; Beijing Chemical Works) in a microwave oven at 560 W for 15 min. The section was blocked with 1% normal goat serum (cat no. ZLI-9021; ZSGB-BIO, Beijing, China) in PBS for 1 h at room temperature then incubated with anti-β klotho antibody (rabbit polyclonal anti-human; cat no. 109454; 1:500 dilution; LifeSpan BioSciences, Inc., Seattle, WA, USA) overnight at 4°C. The section was laid at room temperature for 30 min prior to additional analysis the following day. Next, the section was incubated with peroxidase-conjugated goat anti-rabbit IgG secondary antibody (cat no. ZB-2301; 1:250 dilution; ZSGB-BIO) for 30 min. The reaction was visualized using 3,3′-diaminobenzidine (Beijing Chemical Works) under light microscopy (magnification, ×40) to control optimal dyeing. The section was then counterstained with hematoxylin (Beijing Chemical Works), dehydrated in graded ethanol and xylene, and embedded using permount TM mounting medium (Beijing Chemical Works). The stained TMA was scanned into digital format with the Leica sSCN400 program (Leica Microsystems GmbH, Wetzlar, Germany).

The staining was scored according to the staining intensity and the percentage of the positive tumor cells. The staining intensity was divided into 0, 1, 2 and 3 points based on color (no color, faint yellow, brownish yellow and brown, respectively). The percentage of positive tumor cells was divided into four levels (<5, 5–25, 26–50, 51–75 and 76–100%), corresponding to the assignment of 0, 1, 2, 3 and 4 points, respectively. The product of the scores for staining intensity and the percentage of the positive tumor cells was then used to divide the results into four groups as follows: Negative (−), score 0; mild (+), score 1–4; moderate (++), score 5–8; and marked (+++), score 9–12.

The use of all specimens was approved by the Ethics Committee of Capital Medical University (Beijing, China). Invasive ductal carcinoma tissues, paired adjacent non-tumorous breast tissues and lymph nodes were obtained from 42 patients who were diagnosed with primary breast invasive ductal carcinoma. The specimens were collected from between January 2007 to December 2012 at Da Xing Hospital of Capital Medical University, and all patients provided written informed consent. No chemotherapy, radiotherapy or hormone therapy was administered to the patients prior to therapeutic resection. All the tissues were fixed in 10% formalin (Beijing Chemical Works) at room temperature for 24 h, embedded in paraffin (Beijing Chemical Works) and cut into slices; the paraffin-embedded tissues were cut into 6 sections, each with a diameter of 4 µm. Next, all sections were stained using hematoxylin and eosin. All cases were reviewed by two pathologists who confirmed the diagnosis of invasive ductal carcinoma and the grading of the tumors according to the 2012 World Health Organization criteria (17). A total of 2 cases were classified as grade 1, 16 cases were classified as grade 2 and 24 cases were classified as grade 3. Overall, 28 cases demonstrated lymph node metastasis and 14 cases demonstrated no lymph node metastasis. Isolating normal epithelial cells and tumor cells with a needle under an inverted microscope, normal epithelial cells and tumor cells were collected and placed into Eppendorf tubes. The same tissue from the same 6 sections were put into the same tube. The genomic DNA was then extracted using QIAamp^®DNA mini kit (cat no. 51304; Qiagen GmbH, Hilden, Germany) for loss of heterozygosity (LOH) examination.

A total of 2 microsatellite markers, D4S251 and D4S3040, were selected due to their close proximity to the KLB locus. The 5′ polymerase chain reaction (PCR) primers were labeled with 6-carboxyfluorescein. The following PCR primers were used: D4S251 forward, 5′-TATGTATATATGTGTGCGTGCG-3′ and reverse, 5′-TATGTATATATGTGTGCGTGCG-3′; and D4S3040 forward, 5′-AGCCTAAGCCTATCACAATCCAG-3′; and reverse, 5′-CTGATTGGAACCAAGATGTATATATG-3′ (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). PCR was performed as follows: 10 min at 94°C, followed by 30 sec at 95°C, then 40 cycles of 30 sec each at 55°C and 30 sec at 72°C, followed by an extension of 10 min at 72°C, in a 20-µl reaction mixture containing 0.2 µM of each primer, 2.5 ng/µl DNA and 10 µl 2X EasyTaq^® PCR SuperMix for PAGE (Beijing TransGen Biotech Co., Ltd., Beijing, China). The PCR products were electrophoresed in an ABI Prism 3730 system (Applied Biosystems; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol and the data were analyzed using the GeneMapper 3.2 software (Applied Biosystems; Thermo Fisher Scientific, Inc.).

The statistical analyses were performed with SPSS 19.0 (IBM Corp., Armonk, NY, USA). The correlation between KLB expression and the clinicopathological parameters was evaluated using the χ² test and Spearman's correlation test. Two-tailed P<0.05 was considered to indicate a statistically significant difference.

In the invasive ductal carcinoma tissues from the TMA, KLB was primarily expressed in the cytoplasm, and in a small number of specimens, KLB was expressed simultaneously in the cell membrane. In general, the expression was identified to be weak, and the rates of negative, mild, moderate and marked expression were 43.90 (36/82), 42.69 (35/82), 10.98 (9/82) and 2.44 (2/82), respectively. In the paired adjacent non-tumorous breast tissues, KLB was primarily expressed in the cell cytoplasm of myoepithelial cells, the expression was marked (+++), and the positive expression rate was 100% (82/82) in the samples. However, mammary glandular epithelial cytoplasm exhibited weak expression (Fig. 1). Overall, KLB expression was significantly decreased in invasive ductal carcinoma tissue compared with the paracancerous tissue (P<0.01).

The correlation between KLB expression and a variety of clinical features was analyzed. As summarized in Table I, decreased KLB expression was frequently associated with lymph node involvement (P=0.008) and pathological grade (P=0.022). Using correlation analysis, decreased KLB expression was correlated with increased lymph node involvement (r=−0.234) and higher pathological grade (r=−0.254). However, the expression of KLB was not associated with the age of the patient (P=0.551) or with the size of the tumor (P=0.352).

However, KLB expression was not associated with the expression of ER (P=0.894), PR (P=0.450), HER2 (P=0.558) or Ki-67 (P=0.162). The data suggest that KLB expression was also not associated with the secretion of estrogen or progestin, tyrosine kinase activities, or the proliferative index.

To examine the mechanisms that regulate the differential expression of KLB in tumors compared with normal tissue, the occurrence of LOH at this gene locus was examined.

As demonstrated in Fig. 2A, the markers D4S251 and D4S3040 were used to detect LOH in the patient samples from Da Xing Hospital of Capital Medical University. As summarized in Table II, compared with normal tissue, evidence of LOH was identified in 24/42 tumors (57.14%); additionally, these two markers were lost in 4/42 tumors (9.52%; #9, #12 and #15 in Fig. 2B). A total of 28/42 demonstrated lymph node metastasis; 12/28 of these cases (42.86%) exhibited accompanying LOH (Table II), and the two aforementioned markers were lost in 2/28 tumors (7.14%; #12 and #15 in Fig. 2B). No association was observed between LOH and lymph node metastasis or pathological grade (P>0.05).