- Email: [email protected]
Serum cytokine profile in patients with breast cancer
Cytokine 89 (2017) 173–178
Contents lists available at ScienceDirect
Cytokine journal homepage: www.journals.elsevier.com/cytokine
Serum cytokine profile in patients with breast cancer Linhai Li a,⇑,1, Lidan Chen a,1, Weiyun Zhang a, Yang Liao a, Jianyun Chen a, Yuling Shi a, Shuhong Luo b,⇑ a
Department of Laboratory Medicine, Guangzhou General Hospital of Guangzhou Military Command of PLA, Guangzhou, Guangdong 510010, PR China State Key Laboratory of Organ Failure, Guangdong Provincial Key Laboratory of Tropical Disease Research, School of Biotechnology, Southern Medical University, 1838 N. Guangzhou Avenue, Guangzhou, Guangdong 510515, PR China b
a r t i c l e
i n f o
Article history: Received 30 June 2015 Received in revised form 22 December 2015 Accepted 23 December 2015 Available online 15 February 2016 Keywords: Serum cytokine Antibody array Breast cancer
a b s t r a c t Breast cancer is the leading cause of cancer-related death among women, with a more 20% 5-year survival rate after metastases. It is therefore critical to improve early diagnosis in order to improve disease prognosis. This study investigates cytokine profiles of breast cancer serum with the aim of identifying biomarkers for early diagnosis. A solid-phase antibody array was used for screening 274 biomarkers in serum from breast cancer patients. ELISA assay was carried out to identify biomarkers with differential expression. The serum levels of IL-8, MIP-1 alpha, MIP-1 beta, MMP-8, Resistin, FLRG, and BCAM were significantly higher in breast cancer patients, but LAP and TSH-b levels were lower. ELISA assay results confirmed those of the antibody array. Our results suggest that these cytokines, screened by antibody array, might serve as novel inflammatory markers in breast cancer patients. Whether these biomarkers are specific for breast cancer and can help to improve diagnoses and prognoses of breast cancer needs further investigation. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction In women, breast cancer is the second most frequently diagnosed cancer, but it is the leading cause of cancer deaths [1]. Worldwide, breast cancer kills over 521,000 women annually, and, in 2012, breast cancer was diagnosed in 1.7 million women [2]. Nearly 90% of early-stage breast cancer patients can survive more than 5 years, but this proportion drops to just 20% upon metastasis [3]. However, it is hard to predict micrometastasis by all the current means of prognosis alone. Although magnetic resonance imaging (MRI) is the main method of early diagnosis, it has a drawback: a high rate of false positives that result in unnecessary follow-up examinations, in turn causing further stress and costs for the patient [4]. Therefore, it is suggested that novel biomarkers should be developed as a promising tool for the early detection and monitoring of breast cancer. So far, few such reliable biomarkers have been identified [5], and their identification remains an important focus of research.
⇑ Corresponding authors at: Department of Laboratory Medicine, Guangzhou General Hospital of Guangzhou Military Command of PLA, No. 111 Liuhua Road, Yuexiu District, Guangzhou, Guangdong 510010, PR China (L. Li) and State Key Laboratory of Organ Failure, Guangdong Provincial Key Laboratory of Tropical Disease Research, School of Biotechnology, Southern Medical University, 1838 N. Guangzhou Avenue, Guangzhou, Guangdong 510515, P.R. China (S. Luo). E-mail addresses: [email protected] (L. Li), [email protected] (S. Luo). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.cyto.2015.12.017 1043-4666/Ó 2015 Elsevier Ltd. All rights reserved.
In this study, antibody array technology was used to identify novel serum cytokine biomarkers for breast cancer. 2. Materials and methods 2.1. Patients Serum samples were obtained from 11 Chinese patients with breast cancer who were hospitalized between 2010 and 2011 in the Pathology Department of the First Affiliated Hospital of Zhongshan University (Guangzhou, China). The subjects ranged in age from 37 to 54, with a median of 47 years. The patients had been diagnosed for the first time, and had not yet received chemotherapy or radiotherapy. Blood samples were collected before surgical resection. 10 healthy cases who were receiving regular health examinations in the same hospital were recruited as control subjects, with ages ranging from 38 to 50 and a median of 44 years (Table 1). All subjects signed written informed consent forms prior to their inclusion in this study. All documentation and procedures were supervised by and approved by the Ethics Committee of the First Affiliated Hospital of Zhongshan University. 2.2. Antibody array processing Human sera of the patients and control subjects were measured, according to the manufacturer’s instructions, with a semi-
174
L. Li et al. / Cytokine 89 (2017) 173–178
Table 1 Clinical data of patients and controls. Patient n Age (mean ± SD), year Sex Disease stage Treatment Survival time (mean ± SD), month Control n Age (mean ± SD), year Sex P-value (age, patient vs control)
11 47.36 ± 6.10 Female = 100% III = 72.7%, IV = 27.3% No 23.9 ± 12.1 10 43.7 ± 4.11 Male = 30%, female = 70% 0.127
quantitative human cytokine antibody array (RayBio Human Cytokine Antibody Array G series 4000, Raybiotech, Norcross GA, USA) that detects 274 cytokines (details in Supplementary Table S1) in one experiment. The array consisted of five glass slides (array 6, 7, 8, 9, and 10) that were imprinted with 274 antibodies (Supplementary Table S1). Briefly, serum was incubated in these assay pools for 2 h. After washing, the array glass slides were incubated with a biotin-conjugated anti-cytokine mix for another 2 h. The slides were washed again and developed for a further 2 h with Cy3-conjugated streptavidin. After the experimental procedure, the slides were scanned with a GenePix 4000B scanner (Axon Instruments, GenePix version 5.0) and the signal values were analyzed with the Raybiotech analysis tool, which is based on Microsoft Excel software and specifically designed to analyze the data of Human Cytokine Antibody Array G series 4000. In this analysis tool, signals are normalized using internal positive and negative controls included on the array. 2.3. Identification of cytokine levels by ELISA After screening by antibody array, serum cytokine levels were measured by ELISA (Raybiotech, Norcross GA, USA), according to the manufacturer’s instructions. Serum dilution factors were specific to individual serum biomarkers. After dilution, samples were coated in the plates for 2.5 h at room temperature. The plates then were incubated with a biotin-conjugated antibody for 2 h. After washing, HRP-conjugated streptavidin was added to combine with any biotin catalyzed by the TMB reagent. Finally, sulfuric acid was used to stop the catalytic reaction and the optical density determined via a microplate reader (Biorek, USA, ELx800NB). 2.4. Statistical analysis Protein microarray data were statistically analyzed with IBM Statistical Package for Social Science Statistics 20 software. Differences between groups were determined by the Mann–Whitney U test and were considered significant if the two-sided P values were < 0.05. In addition, fold change (FC) was calculated, and the values given to indicate the relative expression levels of cytokines. 3. Results 3.1. Profiles of cytokines from breast cancer serum Serum samples from the breast cancer patients and the control subjects were spotted into the glass slides and the levels of 274 biomarkers were assessed. After scanning, the signal values of the fluorescent spots of 274 proteins were read using GenePix 5.0 software and statistically analyzed by the Mann–Whitney U test with SPSS 13.0 software. Based on the significance score (P < 0.05), we found that 19 cytokines had differential values
between breast cancer and control subjects (detailed data in Supplementary Table S2). Then, in order to reduce anomalies arising from individual deviation as well as issues with the reliability of measuring positive expression of cytokines (in view of their signal value), we screened 9 cytokines showing more positive expression: IL-8, MIP-1 alpha, MIP-1 beta, MMP-8, Resistin, FLRG, BCAM, LAP, and TSH-b. Their profiles in arrays of both breast cancer and control subjects are shown in Fig. 1, and these are the most representative among the corresponding groups. Fig. 1 shows that each protein was measured in duplicate (collated with colored boxes in Fig. 1) and the fluorescent signals of these proteins were clearly variable between the two groups, meaning that these cytokines were differentially expressed in serum from breast cancer patients and controls. 3.2. Data analysis of cytokines As shown in Table 2, the levels of the nine biomarkers from breast cancer patients were significantly different from those of control group. Using the signal values of these markers as listed in Table 2, we calculated the fold changes of the breast cancer group compared to the control group and these showed that the expression levels of cytokines IL-8, MIP-1 alpha, MIP-1 beta, MMP-8, Resistin, FLRG, and BCAM in the cancer subjects’ serum were clearly increased, while that of LAP and TSH-b were decreased, when compared to those in the serum of the control subjects (Fig. 2). Afterwards, those factors with a significantly variant expression in the two groups were subjected to an unsupervised-hierarchical cluster using Cluster 3.0 software (Fig. 3), and the results showed that the samples of both the breast cancer and control groups were accurately classified into their respective groups, data which further confirmed that these proteins were significantly differentially expressed in breast cancer and control groups. 3.3. Validation of microarray data by ELISA After the cytokine microarray analysis was performed, ELISA was performed to validate the results of the microarray analysis. Of the nine cytokines, six (MIP-1 alpha, IL-8, MIP-1 beta, FLRG, BCAM, TSH-b) were selected for validation by ELISA based on the results from the microarray experiments, historical research data on serum biomarkers in breast cancer, and the availability of commercial test kits. In this ELISA, the number of samples used was raised to 20 cases of breast cancer and 20 disease-free control samples all of which were chosen similarly to the array assay. The results of these validation experiments are shown in Fig. 4. All of the six cytokines were significantly different between patients and controls, a result identical to that of the microarray. 4. Discussion Although there are extant studies showing that the expression of various cytokines changes in cases of breast cancer, and that this can be detected in body fluid [6], to our knowledge, this is the first study using a high-throughput solid protein array approach with sufficient clinical specificity and sensitivity to identify novel serum biomarkers for breast cancer. In this study, the primary screen deployed a novel cytokine antibody array simultaneously identifying 274 cytokines. The array, from RayBiotech (RayBio Human Cytokine Array), yielded nine cytokines differentially expressed between the breast cancer and healthy control subjects: IL-8, MIP-1 alpha, MIP-1 beta, MMP-8, Resistin, FLRG, BCAM, LAP, and TSH-b, according to a Mann–Whitney U test analysis. Furthermore, to confirm the results
L. Li et al. / Cytokine 89 (2017) 173–178
175
Fig. 1. Identification of serum biomarkers for breast cancer by antibody arrays. The serum cytokines differentially expressed between breast cancer and control groups were identified by arrays 7, 8 and 9. Each protein was measured in duplicate. Colored boxes indicate the location of nine significantly different proteins on the arrays, and different colored boxes represent different cytokines.
Table 2 Protein microarray data of differential cytokines from patients and controls serum. Cytokines
p value
Fold change (Cancer/normal)
Mean (Normal, n = 10)
Mean (Breast cancer, n = 11)
MIP-1 alpha IL-8 MIP-1 beta Resistin FLRG MMP-8 BCAM LAP TSH-b
0.000 0.000 0.000 0.001 0.002 0.001 0.001 0.045 0.000
737.502 35.169 12.574 3.280 3.831 2.662 1.946 0.797 0.618
7.591 172.922 572.697 156.018 275.829 2285.479 149.430 3160.955 493.926
5598.506 6081.530 7200.901 511.794 1056.580 6084.275 290.721 2517.709 305.149
derived from the microarray analysis, we used a second, independent analytic method, ELISA, along with fresh samples. In this validation, six of the 9 cytokines were selected for ELISA for various reasons. For all of the six cytokines, the patients’ data were significantly different to those of the control subjects, using both methods. In addition, the unsupervised-hierarchical cluster analysis accurately grouped the breast cancer and normal control samples using the microarray data. These results suggest that protein array is a potent tool in biomedical discovery in this field, that the 9 cytokines may have roles in specific patho-physiological processes of breast cancer and, thus, are potential serum biomarkers for detection. IL-8 is an important inflammatory factor and many studies show that inflammation within the tumor microenvironment plays an important role in breast cancer progression [7]. Consistent with our study results, previous research has shown that serum levels of IL-8 are higher in patients with breast cancer and appear to be an independent prognostic indicator for breast cancer [8,9]. Although no previous studies had shown that chemokines MIP1 alpha and MIP-1 beta were with significant differences in the sera of breast cancer compared to healthy control subjects, some
researchers found that MIP-1 alpha was elevated in nodenegative patients compared to node-positive patients with breast cancer, and MIP-1 alpha and beta levels showed significant differences between different ER+ groups [10,11]. Fortunately, in this study, we have revealed a higher expression of MIP-1 alpha and MIP-1 beta in breast cancer patients when compared to diseasefree control subjects, suggesting that MIP-1 alpha and MIP-1 beta play a role in the pathogenesis and development of breast cancer. Resistin, a newly discovered adipocytokine, not only has high expression in breast cancer tissue and serum, according to our findings, but is also associated with a more malignant clinic pathological status, poor patient survival, and is a risk factor for postmenopausal breast cancer [12,13]. Therefore, Resistin may have promise as an independent early screening, diagnostic, and prognostic predictor for breast cancer. FLRG (also named FSTL3), an extracellular inhibitor of activin displaying anti-proliferative properties in several cell types including breast cancer cells, can bind to activin to promote cell growth [14,15]. Although there have been no reports about FLRG over expression in the sera of breast cancer patients, previous findings have shown that FLRG was up-regulated in all diseased breast tissue and cells, including florid hyperplasia without atypia fibroadenoma, ductal carcinoma, or infiltrating ductal carcinoma [16]. Furthermore, our study has revealed that the level of FLRG is higher in the blood of patients with breast cancer. When combined, these studies suggest a role for this protein in the progression of breast cancer, and as a promising target for the diagnosis and prognostic prediction of breast cancer. Matrix metalloproteinases (MMPs), as a family of zinc dependent endopeptidases, can degrade all extracellular matrix components, which are key mediators of tumor invasion, metastasis, proliferation, survival, and angiogenesis [17,18]. Although the present study had not found difference in MMP8 between control subjects and breast cancer patients, plasma MMP8 levels were positively associated with lymph node involvement but showed a negative correlation with the risk of distant metastasis, suggesting
176
L. Li et al. / Cytokine 89 (2017) 173–178
Fig. 2. Boxplot display of serum cytokines differentially expressed between breast cancer and control groups. Array signals were scanned with a GenePix 4000B scanner, and signal values were statistically analyzed by the Mann–Whitney U test. Differential serum cytokines are shown with boxplot. Center line indicates the median for each data set.
Fig. 3. Unsupervised-hierarchical cluster analysis of significant cytokines. The breast cancer and control groups were distinguished using array data of 9 cytokines that obtained a significant score (P < 0.05) by cluster 3.0 software. The levels of these proteins were shown in color. Low concentrations are in green, median concentrations in black and high concentrations in red. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
L. Li et al. / Cytokine 89 (2017) 173–178
177
Fig. 4. Validation of six differentially expressed cytokines in patients with breast cancer and normal controls. ELISA assay was carried out for this validation, and data is shown in scatter diagram with median. Groups were compared using the Mann–Whitney U test, and the P-value of each protein was less than 0.05.
it has a protective effect against lymph node metastasis [19]. Luckily, our study found the concentration of MMP8 in serum from breast cancer patients to be higher than that from healthy subjects, indicating that it too plays a role in the occurrence and development process of breast cancer. Lu/B-CAM is an Ig superfamily transmembrane protein, which competes with integrins to bind to laminin a5, a subunit of LM511, and is a major component of basement membranes [20]. Previous studies have shown that Lu/B-CAM promotes tumor cell migration by modulating integrin-mediated cell attachment to Laminin-511 protein, and it has been identified as an upregulated antigen in ovarian carcinoma [21,22]. However, we found that Lu/B-CAM also was over expressed in breast cancer, suggesting its involvement in tumor progression. LAP is found in the cell membrane of many immune cells, including Tregs, and is involved in immune regulation. The present study shows that LAP might confer more potent suppressive activity on the human Treg cell population, suggesting LAP could be considered as a regulatory marker [23]. Jayashri Mahalingam discovered that the population of LAP-positive CD4+Foxp3+Tregs was significantly larger in peripheral blood and cancer tissues of colorectal cancer patients than in healthy subjects, indicating their potential role in controlling immune response to cancer [24]. However, our study shows that LAP was significantly down-regulated in the serum of breast cancer patients, but that whether or not the population of LAP+ Treg in peripheral blood of breast cancer patient also decreased needs to be established. This finding suggests LAP might act mainly as a tumor suppressor and inhibit cellular proliferation in breast cancer, making it a target candidate in the prognosis of breast cancer. Thyrotropin (TSH) is a glycoprotein hormone that is produced only by thyrotrope cells of the anterior pituitary gland, containing an a-subunit and a b-subunit. The present study shows that in mouse thyrotropic tumor, thyroid hormones decreased the
secretion of TSH a-subunit and b-subunit [25]. Moreover, rhTSH has been employed as a radio ablative adjunct in patients with differentiated thyroid carcinoma [26]. However, the relationship of TSH, and especially that of b-subunit, with breast cancer has not been reported. This study reveals that TSH-b exists in serum of patients with breast cancer at a lower level than that in normal controls, suggesting that TSH-b might be an adjunct drug for breast cancer as well. Although small numbers of patients and healthy controls were used to screen serum biomarkers of breast cancer by protein array, further cases were prepared for validation by ELISA. The validation results were identical with that obtained by protein array, revealing that the data from the protein array is reliable. Among the markers, the expression profiles of IL-8 and Resistin in breast cancer serum were found to be identical with that from previous studies, while the other seven biomarkers are reported for the first time as significantly differentially expressed in breast cancer serum compared with normal controls. These results suggest that these cytokines, screened by antibody array, might serve as novel inflammatory markers in breast cancer patients. Whether these biomarkers are specific for breast cancer and can help to improve diagnoses and prognoses of breast cancer needs further investigation.
Acknowledgements This study was funded by grants from Guangdong Province, the China Science and Technology Development Project of Guangdong Province (2010B011300018-7) and the Natural Science Foundation of Guangdong Province (8451051501000491). This work was partially supported by grants from the National Natural Science Foundation of China (NSFC) (No. 31170147 to SL). National High-Tech Research and Development Program (863): The development of Homogeneous Time-resolved fluorescent
178
L. Li et al. / Cytokine 89 (2017) 173–178
instrument and reagents for protein-protein interaction detection (No: 2014AA020904 to SL). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cyto.2015.12.017. References [1] A. Jemal, F. Bray, M.M. Center, J. Ferlay, E. Ward, D. Forman, Global cancer statistics, CA Cancer J. Clin. 61 (2011) 69–90. [2] World Health Organization Media Centre, Fact sheet N0297, Cancer, World Heath Organization, 2014,
[12] Y.C. Lee, Y.J. Chen, C.C. Wu, S. Lo, M.F. Hou, S.S. Yuan, Resistin expression in breast cancer tissue as a marker of prognosis and hormone therapy stratification, Gynecol. Oncol. 125 (3) (2012) 742–750. [13] A.M. Assiri, H.F. Kamel, M.F. Hassanien, Resistin, visfatin, adiponectin, and leptin: risk of breast cancer in pre- and postmenopausal Saudi females and their possible diagnostic and predictive implications as novel biomarkers, Dis. Markers 2015 (2015) 253519. [14] V. Maguer-Satta, R. Rimokh, FLRG, member of the follistatin family, a new player in hematopoiesis, Mol. Cell Endocrinol. 225 (2004) 109–118. [15] D. Razanajaona, S. Joguet, A.S. Ay, I. Treilleux, S. Goddard-Léon, L. Bartholin, R. Rimokh, Silencing of FLRG, an antagonist of activin, inhibits human breast tumor cell growth, Cancer Res. 67 (15) (2007) 7223–7229. [16] E. Bloise, H.L. Couto, L. Massai, P. Ciarmela, M. Mencarelli, L.E. Borges, M. Muscettola, G. Grasso, V.F. Amaral, G.D. Cassali, F. Petraglia, F.M. Reis, Differential expression of follistatin and FLRG in human breast proliferative disorders, BMC Cancer 9 (2009) 320. [17] M. Egeblad, Z. Werb, New functions for the matrix metalloproteinases in cancer progression, Nat. Rev. Cancer 2 (2002) 161–174. [18] E.I. Deryugina, J.P. Quigley, Matrix metalloproteinases and tumor metastasis, Cancer Metastasis Rev. 25 (2006) 9–34. [19] Julie Decock, Wouter Hendrickx, Ulla Vanleeuw, et al., Plasma MMP1 and MMP8 expression in breast cancer: protective role of MMP8 against lymph node metastasis, BMC Cancer 8 (2008) 77. [20] N.M. Burton, R.L. Brady, Molecular structure of the extracellular region of Lutheran blood group glycoprotein and location of the laminin binding site, Blood Cells Mol. Dis. 40 (2008) 446–448. [21] Y. Kikkawa, T. Ogawa, R. Sudo, Y. Yamada, F. Katagiri, K. Hozumi, M. Nomizu, J. H. Miner, The lutheran/basal cell adhesion molecule promotes tumor cell migration by modulating integrin-mediated cell attachment to laminin-511 protein, J. Biol. Chem. 288 (43) (2013) 30990–31001. [22] I.G. Campbell, W.D. Foulkes, G. Senger, J. Trowsdale, P. Garin-Chesa, W.J. Rettig, Molecular cloning of the B-CAM cell surface glycoprotein of epithelial cancers. A novel member of the immunoglobulin superfamily, Cancer Res. 54 (1994) 5761–5765. [23] E.M. Shevach, T.S. Davidson, E.N. Huter, R.A. Dipaolo, J. Andersson, Role of TGFBeta in the induction of Foxp3 expression and T regulatory cell function, J. Clin. Immunol. 28 (2008) 640–646. [24] J. Mahalingam, C.Y. Lin, J.M. Chiang, P.J. Su, Y.Y. Chu, H.Y. Lai, J.H. Fang, C.T. Huang, Y.C. Lin, CD4+ T cells expressing latency-associated peptide and Foxp3 are an activated subgroup of regulatory T cells enriched in patients with colorectal cancer, PLoS One 9 (9) (2014) e108554. [25] J.A. Gurr, I.A. Kourides, Regulation of thyrotropin biosynthesis. Discordant effect of thyroid hormone on alpha and beta subunit mRNA levels, J. Biol. Chem. 258 (17) (1983) 10208–10211. [26] M. Luster, F. Lippi, B. Jarzab, rhTSH-aided radioiodine ablation and treatment of differentiated thyroid carcinoma: a comprehensive review, Endocr. Relat. Cancer 12 (1) (2005) 49–64.
Contents lists available at ScienceDirect
Cytokine journal homepage: www.journals.elsevier.com/cytokine
Serum cytokine profile in patients with breast cancer Linhai Li a,⇑,1, Lidan Chen a,1, Weiyun Zhang a, Yang Liao a, Jianyun Chen a, Yuling Shi a, Shuhong Luo b,⇑ a
Department of Laboratory Medicine, Guangzhou General Hospital of Guangzhou Military Command of PLA, Guangzhou, Guangdong 510010, PR China State Key Laboratory of Organ Failure, Guangdong Provincial Key Laboratory of Tropical Disease Research, School of Biotechnology, Southern Medical University, 1838 N. Guangzhou Avenue, Guangzhou, Guangdong 510515, PR China b
a r t i c l e
i n f o
Article history: Received 30 June 2015 Received in revised form 22 December 2015 Accepted 23 December 2015 Available online 15 February 2016 Keywords: Serum cytokine Antibody array Breast cancer
a b s t r a c t Breast cancer is the leading cause of cancer-related death among women, with a more 20% 5-year survival rate after metastases. It is therefore critical to improve early diagnosis in order to improve disease prognosis. This study investigates cytokine profiles of breast cancer serum with the aim of identifying biomarkers for early diagnosis. A solid-phase antibody array was used for screening 274 biomarkers in serum from breast cancer patients. ELISA assay was carried out to identify biomarkers with differential expression. The serum levels of IL-8, MIP-1 alpha, MIP-1 beta, MMP-8, Resistin, FLRG, and BCAM were significantly higher in breast cancer patients, but LAP and TSH-b levels were lower. ELISA assay results confirmed those of the antibody array. Our results suggest that these cytokines, screened by antibody array, might serve as novel inflammatory markers in breast cancer patients. Whether these biomarkers are specific for breast cancer and can help to improve diagnoses and prognoses of breast cancer needs further investigation. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction In women, breast cancer is the second most frequently diagnosed cancer, but it is the leading cause of cancer deaths [1]. Worldwide, breast cancer kills over 521,000 women annually, and, in 2012, breast cancer was diagnosed in 1.7 million women [2]. Nearly 90% of early-stage breast cancer patients can survive more than 5 years, but this proportion drops to just 20% upon metastasis [3]. However, it is hard to predict micrometastasis by all the current means of prognosis alone. Although magnetic resonance imaging (MRI) is the main method of early diagnosis, it has a drawback: a high rate of false positives that result in unnecessary follow-up examinations, in turn causing further stress and costs for the patient [4]. Therefore, it is suggested that novel biomarkers should be developed as a promising tool for the early detection and monitoring of breast cancer. So far, few such reliable biomarkers have been identified [5], and their identification remains an important focus of research.
⇑ Corresponding authors at: Department of Laboratory Medicine, Guangzhou General Hospital of Guangzhou Military Command of PLA, No. 111 Liuhua Road, Yuexiu District, Guangzhou, Guangdong 510010, PR China (L. Li) and State Key Laboratory of Organ Failure, Guangdong Provincial Key Laboratory of Tropical Disease Research, School of Biotechnology, Southern Medical University, 1838 N. Guangzhou Avenue, Guangzhou, Guangdong 510515, P.R. China (S. Luo). E-mail addresses: [email protected] (L. Li), [email protected] (S. Luo). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.cyto.2015.12.017 1043-4666/Ó 2015 Elsevier Ltd. All rights reserved.
In this study, antibody array technology was used to identify novel serum cytokine biomarkers for breast cancer. 2. Materials and methods 2.1. Patients Serum samples were obtained from 11 Chinese patients with breast cancer who were hospitalized between 2010 and 2011 in the Pathology Department of the First Affiliated Hospital of Zhongshan University (Guangzhou, China). The subjects ranged in age from 37 to 54, with a median of 47 years. The patients had been diagnosed for the first time, and had not yet received chemotherapy or radiotherapy. Blood samples were collected before surgical resection. 10 healthy cases who were receiving regular health examinations in the same hospital were recruited as control subjects, with ages ranging from 38 to 50 and a median of 44 years (Table 1). All subjects signed written informed consent forms prior to their inclusion in this study. All documentation and procedures were supervised by and approved by the Ethics Committee of the First Affiliated Hospital of Zhongshan University. 2.2. Antibody array processing Human sera of the patients and control subjects were measured, according to the manufacturer’s instructions, with a semi-
174
L. Li et al. / Cytokine 89 (2017) 173–178
Table 1 Clinical data of patients and controls. Patient n Age (mean ± SD), year Sex Disease stage Treatment Survival time (mean ± SD), month Control n Age (mean ± SD), year Sex P-value (age, patient vs control)
11 47.36 ± 6.10 Female = 100% III = 72.7%, IV = 27.3% No 23.9 ± 12.1 10 43.7 ± 4.11 Male = 30%, female = 70% 0.127
quantitative human cytokine antibody array (RayBio Human Cytokine Antibody Array G series 4000, Raybiotech, Norcross GA, USA) that detects 274 cytokines (details in Supplementary Table S1) in one experiment. The array consisted of five glass slides (array 6, 7, 8, 9, and 10) that were imprinted with 274 antibodies (Supplementary Table S1). Briefly, serum was incubated in these assay pools for 2 h. After washing, the array glass slides were incubated with a biotin-conjugated anti-cytokine mix for another 2 h. The slides were washed again and developed for a further 2 h with Cy3-conjugated streptavidin. After the experimental procedure, the slides were scanned with a GenePix 4000B scanner (Axon Instruments, GenePix version 5.0) and the signal values were analyzed with the Raybiotech analysis tool, which is based on Microsoft Excel software and specifically designed to analyze the data of Human Cytokine Antibody Array G series 4000. In this analysis tool, signals are normalized using internal positive and negative controls included on the array. 2.3. Identification of cytokine levels by ELISA After screening by antibody array, serum cytokine levels were measured by ELISA (Raybiotech, Norcross GA, USA), according to the manufacturer’s instructions. Serum dilution factors were specific to individual serum biomarkers. After dilution, samples were coated in the plates for 2.5 h at room temperature. The plates then were incubated with a biotin-conjugated antibody for 2 h. After washing, HRP-conjugated streptavidin was added to combine with any biotin catalyzed by the TMB reagent. Finally, sulfuric acid was used to stop the catalytic reaction and the optical density determined via a microplate reader (Biorek, USA, ELx800NB). 2.4. Statistical analysis Protein microarray data were statistically analyzed with IBM Statistical Package for Social Science Statistics 20 software. Differences between groups were determined by the Mann–Whitney U test and were considered significant if the two-sided P values were < 0.05. In addition, fold change (FC) was calculated, and the values given to indicate the relative expression levels of cytokines. 3. Results 3.1. Profiles of cytokines from breast cancer serum Serum samples from the breast cancer patients and the control subjects were spotted into the glass slides and the levels of 274 biomarkers were assessed. After scanning, the signal values of the fluorescent spots of 274 proteins were read using GenePix 5.0 software and statistically analyzed by the Mann–Whitney U test with SPSS 13.0 software. Based on the significance score (P < 0.05), we found that 19 cytokines had differential values
between breast cancer and control subjects (detailed data in Supplementary Table S2). Then, in order to reduce anomalies arising from individual deviation as well as issues with the reliability of measuring positive expression of cytokines (in view of their signal value), we screened 9 cytokines showing more positive expression: IL-8, MIP-1 alpha, MIP-1 beta, MMP-8, Resistin, FLRG, BCAM, LAP, and TSH-b. Their profiles in arrays of both breast cancer and control subjects are shown in Fig. 1, and these are the most representative among the corresponding groups. Fig. 1 shows that each protein was measured in duplicate (collated with colored boxes in Fig. 1) and the fluorescent signals of these proteins were clearly variable between the two groups, meaning that these cytokines were differentially expressed in serum from breast cancer patients and controls. 3.2. Data analysis of cytokines As shown in Table 2, the levels of the nine biomarkers from breast cancer patients were significantly different from those of control group. Using the signal values of these markers as listed in Table 2, we calculated the fold changes of the breast cancer group compared to the control group and these showed that the expression levels of cytokines IL-8, MIP-1 alpha, MIP-1 beta, MMP-8, Resistin, FLRG, and BCAM in the cancer subjects’ serum were clearly increased, while that of LAP and TSH-b were decreased, when compared to those in the serum of the control subjects (Fig. 2). Afterwards, those factors with a significantly variant expression in the two groups were subjected to an unsupervised-hierarchical cluster using Cluster 3.0 software (Fig. 3), and the results showed that the samples of both the breast cancer and control groups were accurately classified into their respective groups, data which further confirmed that these proteins were significantly differentially expressed in breast cancer and control groups. 3.3. Validation of microarray data by ELISA After the cytokine microarray analysis was performed, ELISA was performed to validate the results of the microarray analysis. Of the nine cytokines, six (MIP-1 alpha, IL-8, MIP-1 beta, FLRG, BCAM, TSH-b) were selected for validation by ELISA based on the results from the microarray experiments, historical research data on serum biomarkers in breast cancer, and the availability of commercial test kits. In this ELISA, the number of samples used was raised to 20 cases of breast cancer and 20 disease-free control samples all of which were chosen similarly to the array assay. The results of these validation experiments are shown in Fig. 4. All of the six cytokines were significantly different between patients and controls, a result identical to that of the microarray. 4. Discussion Although there are extant studies showing that the expression of various cytokines changes in cases of breast cancer, and that this can be detected in body fluid [6], to our knowledge, this is the first study using a high-throughput solid protein array approach with sufficient clinical specificity and sensitivity to identify novel serum biomarkers for breast cancer. In this study, the primary screen deployed a novel cytokine antibody array simultaneously identifying 274 cytokines. The array, from RayBiotech (RayBio Human Cytokine Array), yielded nine cytokines differentially expressed between the breast cancer and healthy control subjects: IL-8, MIP-1 alpha, MIP-1 beta, MMP-8, Resistin, FLRG, BCAM, LAP, and TSH-b, according to a Mann–Whitney U test analysis. Furthermore, to confirm the results
L. Li et al. / Cytokine 89 (2017) 173–178
175
Fig. 1. Identification of serum biomarkers for breast cancer by antibody arrays. The serum cytokines differentially expressed between breast cancer and control groups were identified by arrays 7, 8 and 9. Each protein was measured in duplicate. Colored boxes indicate the location of nine significantly different proteins on the arrays, and different colored boxes represent different cytokines.
Table 2 Protein microarray data of differential cytokines from patients and controls serum. Cytokines
p value
Fold change (Cancer/normal)
Mean (Normal, n = 10)
Mean (Breast cancer, n = 11)
MIP-1 alpha IL-8 MIP-1 beta Resistin FLRG MMP-8 BCAM LAP TSH-b
0.000 0.000 0.000 0.001 0.002 0.001 0.001 0.045 0.000
737.502 35.169 12.574 3.280 3.831 2.662 1.946 0.797 0.618
7.591 172.922 572.697 156.018 275.829 2285.479 149.430 3160.955 493.926
5598.506 6081.530 7200.901 511.794 1056.580 6084.275 290.721 2517.709 305.149
derived from the microarray analysis, we used a second, independent analytic method, ELISA, along with fresh samples. In this validation, six of the 9 cytokines were selected for ELISA for various reasons. For all of the six cytokines, the patients’ data were significantly different to those of the control subjects, using both methods. In addition, the unsupervised-hierarchical cluster analysis accurately grouped the breast cancer and normal control samples using the microarray data. These results suggest that protein array is a potent tool in biomedical discovery in this field, that the 9 cytokines may have roles in specific patho-physiological processes of breast cancer and, thus, are potential serum biomarkers for detection. IL-8 is an important inflammatory factor and many studies show that inflammation within the tumor microenvironment plays an important role in breast cancer progression [7]. Consistent with our study results, previous research has shown that serum levels of IL-8 are higher in patients with breast cancer and appear to be an independent prognostic indicator for breast cancer [8,9]. Although no previous studies had shown that chemokines MIP1 alpha and MIP-1 beta were with significant differences in the sera of breast cancer compared to healthy control subjects, some
researchers found that MIP-1 alpha was elevated in nodenegative patients compared to node-positive patients with breast cancer, and MIP-1 alpha and beta levels showed significant differences between different ER+ groups [10,11]. Fortunately, in this study, we have revealed a higher expression of MIP-1 alpha and MIP-1 beta in breast cancer patients when compared to diseasefree control subjects, suggesting that MIP-1 alpha and MIP-1 beta play a role in the pathogenesis and development of breast cancer. Resistin, a newly discovered adipocytokine, not only has high expression in breast cancer tissue and serum, according to our findings, but is also associated with a more malignant clinic pathological status, poor patient survival, and is a risk factor for postmenopausal breast cancer [12,13]. Therefore, Resistin may have promise as an independent early screening, diagnostic, and prognostic predictor for breast cancer. FLRG (also named FSTL3), an extracellular inhibitor of activin displaying anti-proliferative properties in several cell types including breast cancer cells, can bind to activin to promote cell growth [14,15]. Although there have been no reports about FLRG over expression in the sera of breast cancer patients, previous findings have shown that FLRG was up-regulated in all diseased breast tissue and cells, including florid hyperplasia without atypia fibroadenoma, ductal carcinoma, or infiltrating ductal carcinoma [16]. Furthermore, our study has revealed that the level of FLRG is higher in the blood of patients with breast cancer. When combined, these studies suggest a role for this protein in the progression of breast cancer, and as a promising target for the diagnosis and prognostic prediction of breast cancer. Matrix metalloproteinases (MMPs), as a family of zinc dependent endopeptidases, can degrade all extracellular matrix components, which are key mediators of tumor invasion, metastasis, proliferation, survival, and angiogenesis [17,18]. Although the present study had not found difference in MMP8 between control subjects and breast cancer patients, plasma MMP8 levels were positively associated with lymph node involvement but showed a negative correlation with the risk of distant metastasis, suggesting
176
L. Li et al. / Cytokine 89 (2017) 173–178
Fig. 2. Boxplot display of serum cytokines differentially expressed between breast cancer and control groups. Array signals were scanned with a GenePix 4000B scanner, and signal values were statistically analyzed by the Mann–Whitney U test. Differential serum cytokines are shown with boxplot. Center line indicates the median for each data set.
Fig. 3. Unsupervised-hierarchical cluster analysis of significant cytokines. The breast cancer and control groups were distinguished using array data of 9 cytokines that obtained a significant score (P < 0.05) by cluster 3.0 software. The levels of these proteins were shown in color. Low concentrations are in green, median concentrations in black and high concentrations in red. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
L. Li et al. / Cytokine 89 (2017) 173–178
177
Fig. 4. Validation of six differentially expressed cytokines in patients with breast cancer and normal controls. ELISA assay was carried out for this validation, and data is shown in scatter diagram with median. Groups were compared using the Mann–Whitney U test, and the P-value of each protein was less than 0.05.
it has a protective effect against lymph node metastasis [19]. Luckily, our study found the concentration of MMP8 in serum from breast cancer patients to be higher than that from healthy subjects, indicating that it too plays a role in the occurrence and development process of breast cancer. Lu/B-CAM is an Ig superfamily transmembrane protein, which competes with integrins to bind to laminin a5, a subunit of LM511, and is a major component of basement membranes [20]. Previous studies have shown that Lu/B-CAM promotes tumor cell migration by modulating integrin-mediated cell attachment to Laminin-511 protein, and it has been identified as an upregulated antigen in ovarian carcinoma [21,22]. However, we found that Lu/B-CAM also was over expressed in breast cancer, suggesting its involvement in tumor progression. LAP is found in the cell membrane of many immune cells, including Tregs, and is involved in immune regulation. The present study shows that LAP might confer more potent suppressive activity on the human Treg cell population, suggesting LAP could be considered as a regulatory marker [23]. Jayashri Mahalingam discovered that the population of LAP-positive CD4+Foxp3+Tregs was significantly larger in peripheral blood and cancer tissues of colorectal cancer patients than in healthy subjects, indicating their potential role in controlling immune response to cancer [24]. However, our study shows that LAP was significantly down-regulated in the serum of breast cancer patients, but that whether or not the population of LAP+ Treg in peripheral blood of breast cancer patient also decreased needs to be established. This finding suggests LAP might act mainly as a tumor suppressor and inhibit cellular proliferation in breast cancer, making it a target candidate in the prognosis of breast cancer. Thyrotropin (TSH) is a glycoprotein hormone that is produced only by thyrotrope cells of the anterior pituitary gland, containing an a-subunit and a b-subunit. The present study shows that in mouse thyrotropic tumor, thyroid hormones decreased the
secretion of TSH a-subunit and b-subunit [25]. Moreover, rhTSH has been employed as a radio ablative adjunct in patients with differentiated thyroid carcinoma [26]. However, the relationship of TSH, and especially that of b-subunit, with breast cancer has not been reported. This study reveals that TSH-b exists in serum of patients with breast cancer at a lower level than that in normal controls, suggesting that TSH-b might be an adjunct drug for breast cancer as well. Although small numbers of patients and healthy controls were used to screen serum biomarkers of breast cancer by protein array, further cases were prepared for validation by ELISA. The validation results were identical with that obtained by protein array, revealing that the data from the protein array is reliable. Among the markers, the expression profiles of IL-8 and Resistin in breast cancer serum were found to be identical with that from previous studies, while the other seven biomarkers are reported for the first time as significantly differentially expressed in breast cancer serum compared with normal controls. These results suggest that these cytokines, screened by antibody array, might serve as novel inflammatory markers in breast cancer patients. Whether these biomarkers are specific for breast cancer and can help to improve diagnoses and prognoses of breast cancer needs further investigation.
Acknowledgements This study was funded by grants from Guangdong Province, the China Science and Technology Development Project of Guangdong Province (2010B011300018-7) and the Natural Science Foundation of Guangdong Province (8451051501000491). This work was partially supported by grants from the National Natural Science Foundation of China (NSFC) (No. 31170147 to SL). National High-Tech Research and Development Program (863): The development of Homogeneous Time-resolved fluorescent
178
L. Li et al. / Cytokine 89 (2017) 173–178
instrument and reagents for protein-protein interaction detection (No: 2014AA020904 to SL). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cyto.2015.12.017. References [1] A. Jemal, F. Bray, M.M. Center, J. Ferlay, E. Ward, D. Forman, Global cancer statistics, CA Cancer J. Clin. 61 (2011) 69–90. [2] World Health Organization Media Centre, Fact sheet N0297, Cancer, World Heath Organization, 2014,
[12] Y.C. Lee, Y.J. Chen, C.C. Wu, S. Lo, M.F. Hou, S.S. Yuan, Resistin expression in breast cancer tissue as a marker of prognosis and hormone therapy stratification, Gynecol. Oncol. 125 (3) (2012) 742–750. [13] A.M. Assiri, H.F. Kamel, M.F. Hassanien, Resistin, visfatin, adiponectin, and leptin: risk of breast cancer in pre- and postmenopausal Saudi females and their possible diagnostic and predictive implications as novel biomarkers, Dis. Markers 2015 (2015) 253519. [14] V. Maguer-Satta, R. Rimokh, FLRG, member of the follistatin family, a new player in hematopoiesis, Mol. Cell Endocrinol. 225 (2004) 109–118. [15] D. Razanajaona, S. Joguet, A.S. Ay, I. Treilleux, S. Goddard-Léon, L. Bartholin, R. Rimokh, Silencing of FLRG, an antagonist of activin, inhibits human breast tumor cell growth, Cancer Res. 67 (15) (2007) 7223–7229. [16] E. Bloise, H.L. Couto, L. Massai, P. Ciarmela, M. Mencarelli, L.E. Borges, M. Muscettola, G. Grasso, V.F. Amaral, G.D. Cassali, F. Petraglia, F.M. Reis, Differential expression of follistatin and FLRG in human breast proliferative disorders, BMC Cancer 9 (2009) 320. [17] M. Egeblad, Z. Werb, New functions for the matrix metalloproteinases in cancer progression, Nat. Rev. Cancer 2 (2002) 161–174. [18] E.I. Deryugina, J.P. Quigley, Matrix metalloproteinases and tumor metastasis, Cancer Metastasis Rev. 25 (2006) 9–34. [19] Julie Decock, Wouter Hendrickx, Ulla Vanleeuw, et al., Plasma MMP1 and MMP8 expression in breast cancer: protective role of MMP8 against lymph node metastasis, BMC Cancer 8 (2008) 77. [20] N.M. Burton, R.L. Brady, Molecular structure of the extracellular region of Lutheran blood group glycoprotein and location of the laminin binding site, Blood Cells Mol. Dis. 40 (2008) 446–448. [21] Y. Kikkawa, T. Ogawa, R. Sudo, Y. Yamada, F. Katagiri, K. Hozumi, M. Nomizu, J. H. Miner, The lutheran/basal cell adhesion molecule promotes tumor cell migration by modulating integrin-mediated cell attachment to laminin-511 protein, J. Biol. Chem. 288 (43) (2013) 30990–31001. [22] I.G. Campbell, W.D. Foulkes, G. Senger, J. Trowsdale, P. Garin-Chesa, W.J. Rettig, Molecular cloning of the B-CAM cell surface glycoprotein of epithelial cancers. A novel member of the immunoglobulin superfamily, Cancer Res. 54 (1994) 5761–5765. [23] E.M. Shevach, T.S. Davidson, E.N. Huter, R.A. Dipaolo, J. Andersson, Role of TGFBeta in the induction of Foxp3 expression and T regulatory cell function, J. Clin. Immunol. 28 (2008) 640–646. [24] J. Mahalingam, C.Y. Lin, J.M. Chiang, P.J. Su, Y.Y. Chu, H.Y. Lai, J.H. Fang, C.T. Huang, Y.C. Lin, CD4+ T cells expressing latency-associated peptide and Foxp3 are an activated subgroup of regulatory T cells enriched in patients with colorectal cancer, PLoS One 9 (9) (2014) e108554. [25] J.A. Gurr, I.A. Kourides, Regulation of thyrotropin biosynthesis. Discordant effect of thyroid hormone on alpha and beta subunit mRNA levels, J. Biol. Chem. 258 (17) (1983) 10208–10211. [26] M. Luster, F. Lippi, B. Jarzab, rhTSH-aided radioiodine ablation and treatment of differentiated thyroid carcinoma: a comprehensive review, Endocr. Relat. Cancer 12 (1) (2005) 49–64.