Reproductive Hormones and Their Receptors May Affect Lung Cancer Open Access

Lung cancer remains the leading cause of cancer-related death globally, and outcomes for patients diagnosed with advanced non-small cell lung cancer (NSCLC) are poor despite recent advances in treatment [1]. In 2016, there were 222,500 new cancer cases and 155,870 cancer deaths related to the lung and bronchus system in the United States. In contrast to the steady increase in survival for most cancers, that for lung cancer has been slow, with a current 5-year relative survival of 18% [2]. There are a number of factors associated with the poor prognosis, such as drug resistance [3] and gender and age differences [4]. For a majority of patients, the disease condition is initially chemosensitive but then relapses with acquired chemoresistance [3]. The low survival rate of patients suffering from lung cancer is caused mainly by delayed detection and diagnosis, resulting in the identification of disease at an advanced stage and leaving the patient with limited treatment options [5].

Although some molecular alterations [6, 7] are shared among various histologic subtypes of cancer, the majority of genomic alterations remain distinct [8]. Accumulating evidence [9-13] indicates that the presence and progression of lung cancer is affected by gender-dependent factors, especially oestrogens. Oestrogens affect cellular processes by binding to oestrogen receptors (ERs) and signalling in two different ways: genomic and non-genomic [14]. In genomic signaling, oestrogens bind to ERs, mainly ERα and ERβ, which causes ER dimerization, translocation to the nucleus and binding to specific DNA regions known as oestrogen response elements (EREs) [15, 16].

Some modern medical studies [14, 17, 18] have shown that sex hormones play an important role in the development of lung tissue, the formation of lung inflammation and perhaps the molecular biology of lung cancer. Many researchers found that the outcomes of patients with lung cancer were related to the use of anti-oestrogens [19]. Fan S et al also reported that ERβ activation in lung cancer cells promotes metastasis by increasing the expression of MMP-2, which is associated with cellular invasiveness [20]. Thus, sex hormones and their corresponding receptors may be potential molecular biomarkers that predict the occurrence, development, treatment, and prognosis of lung cancer.

Herein, we established a sex hormone deficiency model in C57BL/6 female mice by ovariectomy. After subcutaneous inoculation of a tumour cell suspension in all mice, body weight, transplanted tumour volume and weight, serum levels of oestradiol (E₂) and testosterone (T) and ER expression were determined to identify potential differences upon ovariectomy. In addition, the specific mechanisms by which sex hormones and their receptors influence the development of lung cancer were explored.

Twenty-four C57BL/6 female mice aged 56-62 days and weighing 20±2 g were purchased from Beijing Weitong Lihua Experimental Animal Technology Company (certificate number: SCXK 2012-0001). These 24 mice were randomly divided into the ovariectomized group and the control group. Pentobarbital sodium (no. 57330, 5 g/bottle) was from Merck (Germany). Benzylpenicillin sodium (1600 thousand units per bottle) was from Huabei Pharmaceutical Company. Instant PBS powder (LB12231), citrate buffer powder (LB00085), neutral balsam (20160308), Haematoxylin Stain reagent (SLBK4907V) and other reagents were purchased from Zhengzhou Saiboer Biotechnology Company. Enzyme-linked immunosorbent assay (ELISA) test kits for E₂ and T (KGE014 and KGE010) were purchased from R&D Systems (USA). Polyclonal antibodies against AR, ERα, and ERβ (ab133273, ab75635, and ab3577) were from Abcam (USA).

First, all female mice were weighed and then injected with 1% pentobarbital sodium (0.07 ml/10 g) to induce anaesthesia. After disinfecting the mouse back, we made a small approximately 1.5-cm incision on the left or right flank 0.5 cm from the intersection of the lines extending from the hind legs and the midline of the back. The skin and muscular layers were exposed sequentially, and when we pulled out the fat pad on the left or right side with forceps, the ovary and fallopian tube were exposed. The fat pad was returned to the abdominal cavity after ligating fallopian tube and removing the ovary. The two sides were treated in the same way. Lastly, we sutured the muscle layer and the epidermis continuously with absorbable sutures. The control group was treated similarly as the case group without removing the ovary or ligating the fallopian tube, namely, the fat pad was held with blunt forceps, a section of the adipose tissue was cut, and the layers were sutured continuously with absorbable sutures. To prevent infection, all mice were treated with penicillin (60,000 units per mouse) for seven days after surgery.

Tumour-bearing mice with Lewis lung cancer were decapitated and then soaked in 75% medical-grade alcohol. After 5 minutes, we placed the mouse on a clean bench, removed the tumour, eliminated necrotic tissue and normal tissue, and selected tumour sections in good condition. The selected tumour sections were cut into small pieces with scissors and then ground with a glass homogenizer in normal saline at a 1: 3 ratio of tumour (g) to normal saline (ml). Then, the mixture was filtered through a 200 mesh screen to prepare a single cell suspension, and the percentage of viable cells was greater than 95% based on 0.2% trypan blue staining. Then, we adjusted the single cell suspension to a concentration of 1×10⁷ cells per ml. After 3 weeks of recovery from surgery, each mouse in the ovariectomized and control groups was inoculated subcutaneously with 0.3 ml of the single cell suspension under the right axillary fold. On the 5^th day after inoculation, the subcutaneous node was palpable. In addition, beginning on the 6^th day, we measured the longest (a) and shortest diameters (b) of the transplanted tumours and weighed all mice every 3 days. We calculated the tumour volume using the following formula: V (mm³)=1/2•ab². The growth curve was generated based on the arithmetic mean of the weight or volume in each group. After 21 days, all the mice were weighed, blood was drawn, and the mice were killed by decapitation.

On the 21^st day after inoculation, all the mice in the ovariectomized and control groups were weighed. After rapidly removing the eyeball with sterile ophthalmic forceps, blood samples were collected into sterile tubes. Then, all the samples were allowed to clot for 30 minutes at room temperature before centrifugation for 15 minutes at 1000 ×g. The upper yellow-clear serum was removed, labelled and stored at ≤ -20°C. Based on the operational guidelines for ELISA kits (KGE014 and KGE010; R&D Systems, USA), we reconstituted the E₂ and T standards and generated the standard curves. According to the optical density (OD) of each well at 450 nm and the standard curve, the serum levels of E₂ and T could be calculated.

Total RNA was extracted from tissues using TRIzol reagent (Invitrogen, USA), and RNA concentration and purity were detected by a UV spectrophotometer. A cDNA kit (Thermo, USA) was used for reverse transcription; specifically, genomic DNA was removed at 42°C for 2 minutes, and the reverse transcription reaction was performed at 37°C for 15 minutes and 85°C for 5 seconds. Using cDNA as the template, PCR was performed with FastStart Universal SYBR Green Master Mix (Roche, USA). GAPDH expression levels were used as the internal reference. The following primer sequences were used: GAPDH 5'-GTTCCTACCCCCAATGTGTCC-3' (forward) and 5'-TAGCCCAAGATGCCCTTCAGT-3' (reverse); ERα 5'-CAGCAGTAACGAGAAAGGAAACA-3' (forward) and 5'-TCGGCGGTCTTTCCGTATG-3' (reverse); ERβ 5'-TGATGATGTCCCTCACGAAGC-3' (forward) and 5'-AGAACGAGGTCTGGAGCAAAG-3' (reverse); and AR 5'-AAATGGGACCTTGGATGGAGA-3' (forward) and 5'-TTTCCCTTCAGCGGCTCTTT-3' (reverse). The relative mRNA expression levels were quantitated by the 2^-ΔΔCt method. Each experiment was repeated independently three times.

When each mouse was killed by decapitation, the transplanted tumour and lung tissue were dissected and stored in liquid nitrogen. To extract total protein, we first ground the stored tissue with liquid nitrogen and then added a mixture of 10 µl of phenylmethanesulfonyl fluoride (PMSF) and 990 µl of RIPA, with continued grinding for 30 minutes. The obtained mixture was centrifuged for 15 minutes at 12000 ×g at 4℃, and the supernatants containing total protein were harvested. The bicinchoninic acid (BCA) method was used to detect the protein concentration. Laemmli sample buffer was added, and the samples were boiled. The proteins were separated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) with subsequent transfer onto a polyvinylidene difluoride (PVDF) membrane. The membranes were incubated overnight at 4°C with antibodies against AR (1: 1000, ab133273), ERα (1: 1000, ab75635), ERβ (1: 500, ab3577) or β-actin (1: 1000, GB13001-1). After subsequent incubation with secondary antibodies, the membranes were subjected to chemiluminescence detection and exposed on X-ray films. Band density was analysed by ImageJ2x software. Each experiment was repeated independently three times.

All the transplanted tumour and lung tissues from mice in the case and control groups were removed on ice after the mice were killed by decapitation. Then, 4-μm formalin-fixed, paraffin-embedded sections were created from these tissues. In brief, endogenous peroxidase was quenched with 3% hydrogen peroxide, and sections were incubated with normal goat serum for 30 minutes to block nonspecific antibody binding sites. Then, the sections were incubated with antibodies against AR, ERα or ERβ (ab133273, ab75635, and ab3577; Abcam, USA) overnight at 4°C and then with a biotin-free HRP-labelled polymer. DAB solution and haematoxylin were used to visualize the positive reactions. Immunohistochemical scoring was based on the staining intensity and the percentage of positively stained cells. The staining intensity was scored as follows: 0, no staining; 1, weakly stained; 2, moderately stained; and 3, strongly stained. The percentage of positively stained cells was scored as follows: 0, 0%; 1, 1–25%; 2, 26–50%; 3, 51–75%; and 4, > 75%. The final staining score was the product of the staining intensity score and the percent positive score: -, final staining score < 3; +, final staining score = 3; ++, final staining score = 4; and +++, final staining score ≥ 5. We classified positive expression as a final staining score ≥ 4 and negative expression as < 4 in the mouse tumour and lung tissues. Each experiment was repeated independently three times.

The data were analysed by SPSS version 17.0. The rates of positive expression of AR, ERα and ERβ in lung and tumour tissues from the two groups were analysed by the χ² test. The median mouse body weight, transplanted tumour weight, transplanted tumour volume and E₂ and T serum levels were analysed by the t test. The data are presented as the mean ± standard deviation, and P < 0.05 was considered significant.

Compared with the control mice, the ovariectomized mice were timid, easily frightened and less active, and their hair colour was not as bright. On the 5^th day after inoculation of the single cell suspension, a subcutaneous node was palpated. Beginning on the 6^th day, we measured the longest (a) and shortest diameters (b) of the transplanted tumour and the mouse body weight every 3 days. We calculated the tumour volume (V=1/2•ab², in mm³) and generated a growth curve. The transplanted tumours were generally larger in the ovariectomized group than in the control group (Fig. 1). The growth curves showed that the transplanted tumours in the ovariectomized group grew faster than those in the control group (P<0.05). In addition, the tumour weight differed significantly between the two groups (1.83±0.40 and 3.13±0.43, P<0.05) (Fig. 2).

Three weeks after inoculation of the single cell suspension, blood samples were rapidly removed via retroorbital bleeding from all the mice in the two groups. The blood samples were allowed to clot for 30 minutes at room temperature and then centrifuged for 15 minutes at 1000 ×g. Based on the operational guidelines for the ELISA kits (KGE014 and KGE010; R&D Systems, USA), the serum levels of E₂ and T were calculated according to the OD and the E₂ and T standard curves. During the ovariectomy operation, 1 mouse died. During the tumour formation process, tumours in 2 mice in the control group and 1 mouse in the ovariectomized group did not grow. Thus, there were 10 available blood samples from each group. The analysis showed that the serum levels of E₂ and T in the ovariectomized group decreased exponentially, whereas the E₂/T ratio increased at certain times (55.88±11.45 and 78.21±9.37, 0.82±0.14 and 1.46±0.16, 69.62±14.43±29.81 and 52.22±5.42, P<0.05) compared with the control group (Table 1).

Total RNA was extracted from tissues, and the relative AR, ERα and ERβ mRNA expression levels in transplanted tumour and lung tissues are shown in Fig. 3. Expression levels in transplanted tumour tissues were higher in the ovariectomized group than in the control group, especially for ERβ mRNA (P<0.01). Furthermore, expression levels were higher in lung tissues from the ovariectomized group than in those from the control group, with a significant increase in ERα mRNA expression levels (P<0.01). The higher mRNA expression levels of AR, ERα and ERβ in the ovariectomized group may be associated with the low serum levels of E₂ and T, which led to the imbalance in the murine microenvironment.

According to the Western blot analysis, ERβ protein expression levels were highest in transplanted tumour tissues in both the control and ovariectomized groups (P<0.0001; Fig. 4). In lung tissues, ERα protein expression levels were highest in the ovariectomized group (P<0.0001). The expression levels of AR, ERα and ERβ were clearly higher in the ovariectomized group than in the control group. The IHC results showed that AR, ERα and ERβ staining was more evident in the ovariectomized group than in the control group (Fig. 5). Staining score analysis supported these observations and revealed that the ERβ protein expression score was highest in transplanted tumour tissue from the ovariectomized group (P<0.05; Fig. 6). In addition, in lung tissue, the ERα expression score was the highest (P<0.05). The rates of high expression of AR, ERα and ERβ in lung and tumour tissues from the two groups indicated that tumour ERβ expression and lung ERα expression were significantly different in the ovariectomized group compared with the control group (both P<0.05, Table 2). Therefore, Western blot and IHC results revealed high expression of AR, ERα and ERβ in transplanted tumour tissue and lung tissue from ovariectomized mice.

For half a century, the incidence and mortality of lung cancer have been increasing, and lung cancer is the leading cause of cancer-related death in most developed countries [2]. As the number of female smokers continues to increase, the ratio of male to female patients with lung cancer has risen rapidly. Oestrogens can signal through either ERα or ERβ; these receptors are involved in numerous kinds of malignant tumours, including hormone-dependent tumours such as breast cancer [21] and prostate cancer [22]. However, many studies [23-25] have found that oestrogens and ER have significant effects on lung development and physiology, which provided a potential molecular mechanism for the gender differences in lung cancer. Thus, we performed this study to explore the previously unknown regulators of lung cancer.

Bilateral ovariectomy [26] was used to decrease sex hormones in mice in the ovariectomized group, and mice in the control group underwent partial removal of perinephric fat near the bilateral ovaries. Next, all mice were inoculated with a single cell suspension of Lewis lung cancer cells from C57BL/6 mice. Then, we observed significant differences between the control and ovariectomized groups in terms of body weight curves and transplanted tumour volume. In addition, the transplanted tumour weight of mice in the ovariectomized group was heavier than that in the control group (1.83±0.40 and 3.13±0.43, p<0.05). The serum analysis showed that the serum levels of E₂ and T in the ovariectomized group decreased exponentially, while the E₂/T ratio was increased at certain time points compared with the control group (E₂: 55.88±11.45 and 78.21±9.37; T: 0.82±0.14 and 1.46±0.16; ratio: 69.62±14.43±29.81 and 52.22±5.42, p<0.05). Further analysis of AR, ERα and ERβ mRNA and protein expression levels in tumour and lung tissues revealed higher expression in the ovariectomized group, particularly regarding ERβ in tumour tissue and ERα in lung tissue. Therefore, ovariectomy decreased the serum levels of E₂ and T and increased the E₂/T ratio in mice, and the resulting imbalance in the internal environment promoted the growth of transplanted tumours. ERα and ERβ played key roles in this process.

Clinical studies [27, 28] support the conclusion that there is an imbalance in sex hormones and their receptors in patients with lung cancer. E₂ levels were much higher in patients with lung cancer than in a normal control group, while T levels were lower in patients. In addition, several studies [29-31] have now reported that exposure to hormone replacement therapy (HRT) is associated with negative effects on lung cancer survival. Ganti et al [29] performed a retrospective chart review of women diagnosed with lung cancer and collected data including age, past history of cancer, HRT use and overall survival. They found that overall survival was significantly higher in patients without HRT compared with those who received HRT. HRT may affect the endocrine environment in such a way that promotes lung cancer.

The mechanism of lung tumourigenesis is complex, and lung cancer is mainly caused by various carcinogenic factors that cause gene changes and induce normal cells to become cancerous. However, it is not known how gene polymorphisms affect E₂ metabolism [32]. The body of clinical trials suggests that the patients with lung cancer who respond well to EGFR TKIs are mainly female, not male [33], which may indicate potential cross-talk in signalling between EGFR and ER in lung cancer [34, 35]. Other studies found that abnormal ER expression is related to mutations in other genes. Both Mah V [16] and Marquez-Garban DC [36] reported that oestrogen and ER can activate the Src family of nonreceptor tyrosine kinases, MAPKs (mitogen-activated protein kinases) and PI3K (phosphatidylinositol-3 kinase), which affected cancer cell growth, proliferation and apoptosis. Thus, it will be worthwhile to study what affects the overexpression of ER-related receptors and subsequently promotes the formation of malignant tumours.

This research was supported by grants from the National Natural Science Foundation of China (81473497), the Social Research Project of Henan Provincial Science and Technology Department (132102310106), and the Natural Science Foundation of Henan Provincial Department of Education (12A320077).

None.

M. Dou and K. Zhu contributed equally to this article.