- Email: [email protected]
Proteomic analysis of the papillary thyroid microcarcinoma
Accepted Manuscript Title: Proteomic analysis of the papillary thyroid microcarcinoma Author: Shan Jin Wuyuntu Bao Yun-Tian Yang Quan Fu Yinbao Bai Yousheng Liu PII: DOI: Reference:
S0003-4266(19)30028-9 https://doi.org/doi:10.1016/j.ando.2019.01.003 ANDO 1109
To appear in:
Annales d’Endocrinologie
Please cite this article as: Jin S, Bao W, Yang Y-Tian, Quan F, Bai Y, Liu Y, Proteomic analysis of the papillary thyroid microcarcinoma, Annales d’Endocrinologie (2019), https://doi.org/10.1016/j.ando.2019.01.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
M
an
us
cr
ip t
Proteomic analysis of the papillary thyroid microcarcinoma Analyse protéomique du microcarcinome papillaire de la thyroïde Shan Jin1, Wuyuntu Bao1, Yun-Tian Yang1, Quan Fu2, Yinbao Bai1, Yousheng Liu1 1 Department of General Surgery, Affiliated Hospital of Inner Mongolia Medical University, Hohhot, Inner Mongolia Autonomous Region, China. 2 Department of Clinical Laboratory, Affiliated Hospital of Inner Mongolia Medical University, Hohhot, Inner Mongolia Autonomous Region, China. Corresponding author: Shan Jin, MD, PhD, Department of General Surgery, Affiliated Hospital of Inner Mongolia Medical University, Tongdao North Rd 1, Hohhot 010050, Inner Mongolia Autonomous Region, China; Tel: +86-471-3451465 Fax: +86-471-3451465 E-mail: [email protected] Wuyuntu Bao E-mail: [email protected]; Yun-Tian Yang E-mail: [email protected]; Quan Fu E-mail: [email protected] ; Yinbao Bai E-mail: [email protected]; Yousheng Liu E-mail: [email protected]
ed
Abstract
Objective: The present study applied iTRAQ and LC-MS/MS techniques for proteome analysis and compared data between specimens of papillary thyroid microcarcinoma
ce pt
(PTMC) vs appropriate controls, in order to investigate the mechanisms underlying the invasion and metastasis process in PTMC development. Materials and methods: Fresh-tissue specimens were collected from 40 patients with
Ac
thyroid disease who underwent surgical treatment. Specimens were divided into four groups: normal histology (NH; n=8), benign thyroid tumor (BTT; n=10), classic PTMC with lymph node metastasis (PTC-LNM(+); n=11), and classic PTMC without lymph node metastasis (PTC-LNM(-); n=11). Proteomic studies were conducted on PTMC tissue samples without capsule invasion and with tumor diameter ranging from 0.5 cm to 1 cm, so as to focus the study on PTMC development excluding metastasis. Results: A total of 8,036 proteins were identified in the four groups. Based on protein function analysis, proteins that might be associated with PTMC invasion and metastasis were screened: alpha-actinin-1, alpha-1-antitrypsin, hepatoma-derived growth factor 1
Page 1 of 23
(HDGF), high-mobility group protein HMGI-C, and carbonic anhydrase 4. In addition, proteins involved in the focal adhesion pathway were examined. Immunohistochemistry confirmed the reliability of the iTRAQ results and the universality of differentially expressed proteins. The data showed that HDGF and high-mobility group protein HMGI-C are upregulated in PTMC and that the focal adhesion pathway that promotes PTMC LNM is
ip t
activated. Conclusions: These findings provide insight into the mechanisms underlying PTMC
cr
invasion and metastasis.
Keywords: invasion; metastasis; papillary thyroid microcarcinoma; proteomics
us
Résumé
Contexte : Nous avons analysé le protéome d'échantillons de microcarcinome papillaire
an
de la thyroïde (MCPT) grâce aux techniques iTRAQ (étiquette isobare pour la quantification relative et absolue) et LC-MS/MS (chromatographie en phase liquide
M
couplée à la spectrométrie de masse) afin d’identifier les mécanismes de l’invasion et de la métastase du MCPT.
ed
Objectifs : Des échantillons de tissu frais étaient pris sur 40 patients atteints d’une maladie de la thyroïde, qui ont subi un traitement initial et étaient ensuite divisés en
ce pt
4 groupes : tissus histologiques normaux (HN ; n = 8), tissus présentant une tumeur de la thyroïde bénigne (TTB ; n = 10), tissus présentant un MCPT classique et des métastases ganglionnaires (CPT-MG(+) ; n = 11), tissus présentant un MCPT classique sans
Ac
métastases ganglionnaires (CPT-MG (-) ; n = 11). Les études des différents protéomes ont été réalisées sur des tissus atteints de MCPT sans invasion capsulaire, prélevés dans des carcinomes uniques de diamètre 0,5-1 cm. Ceci exclut les facteurs anatomiques de métastase, fournissant ainsi des informations sur les changements biologiques au niveau moléculaire impliqués dans le MCPT. Résultats : Au total, 8036 protéines ont été identifiées au sein des 4 groupes. À partir de l’analyse de la fonction de ces protéines, celles pouvant être associées à l’invasion et la métastase du MCPT ont été identifies : alpha-actinine-1, alpha-12
Page 2 of 23
antitrypsine, facteur de croissance dérivé de l’hépatome (HDGF), protéine du groupe de haute mobilité HMGI-C, et l’anhydrase carbonique-4. De plus, la voie d’adhésion focale est peut-être impliquée dans l’invasion et la métastase du MCPT. Les techniques d’immunohistochimie ont confirmé la fiabilité des résultats obtenus par la technique iTRAQ ainsi que l’universalité des protéines à expression
ip t
différentielle. Conclusions : Cette étude démontre que la surexpression du facteur HDGF et de la
cr
protéine du groupe de haute mobilité HMGI-C ainsi que l’activation de la voie d’adhésion focale favorisent l’apparition de métastases ganglionnaires associées au MCPT. Ces
us
résultats pourraient améliorer la compréhension des mécanismes de l’invasion et de la
an
métastase du MCPT.
Mots-clés : invasion ; métastase ; micro-carcinome papillaire de la thyroïde ; protéomique
M
Thyroid carcinoma is a common malignancy of the endocrine system, with incidence increasing every year [1,2]. The most common pathological type is papillary thyroid
ed
carcinoma (PTC), which accounts for >90% of thyroid carcinomas, with papillary thyroid microcarcinoma (PTMC) accounting for about 30–43.1% of PTCs [3-5]. Due to the low-
ce pt
grade malignancy of PTC, post-surgical prognosis and the results of postoperative adjuvant treatment are satisfactory and long-term survival is achievable. Nevertheless, 20– 90% of PTC patients present cervical lymph node metastasis (LNM) [6,7]; the rate in PTMC patients is 16.9–54.1% in the central region of the neck, and 3.7-10.1% in the
Ac
lateral neck [8-11]. Some studies have shown that cervical LNM is observed under a microscope in >60% of the patients with PTMC [12]. Moreover, cervical LNM is a major risk factor for increased recurrence and death in patients with thyroid carcinoma [13-15]. PTMC metastasis is related to tumor diameter, multifocality and capsule invasion [16,17]. Some PTMC patients show LNM without anatomic risk factors such as capsule invasion or multifocality, which indicates that the invasiveness and metastasis of PTMC are determined by its biological behavior. Moreover, the onset, development, invasion and metastasis of thyroid carcinoma are closely related to altered gene expression, disruption in the signal transduction pathway, and dysfunction in the associated proteins [2,18]. The 3
Page 3 of 23
present study therefore performed proteomic analysis based on the iTRAQ and LC-MS/MS techniques to further elucidate the functions and check expression of candidate proteins and shed light on the mechanisms underlying PTMC invasion and metastasis. Material and methods
ip t
Ethics This study conformed to the Declaration of Helsinki regarding ethical principles for medical
cr
research in humans. It was approved by the Biomedical Ethics Committee of Inner Mongolia Medical University (No. YKD2014063).
us
Trial registration
an
ClinicalTrials.gov Protocol Registration System: ClinicalTrials.gov ID NCT02814032, registered on June 24, 2016.
M
General data and sample collection
Forty patients with thyroid disease, initially treated in the General Surgery Department of
ed
the Affiliated Hospital of Inner Mongolia Medical University, were enrolled from June 2016 to May 2017. Standard surgery was performed according to the Chinese version of the
ce pt
guidelines for “diagnosing and managing thyroid nodules and differentiated thyroid cancer” [19]. Forty fresh frozen thyroid specimens were classified as: normal histology (NH, n=10), benign thyroid tumor (BTT, n=8), classic PTMC without LNM (PTC-LNM(-), n=11), and classic PTMC with LNM (PTC-LNM(+), n=11). Inclusion criteria for PTMC patients were as
Ac
follows: (1) complete medical records; (2) operated on by the same medical team; (3) confirmed on pathology as classic PTC without capsule invasion from single carcinoma lesions, with tumor diameter 0.5–1 cm; (4) associated central neck lymph node dissection in the same step; (5) without association of neck and other malignant tumors; and (6) without history of head and neck radiotherapy. Inclusion criteria for benign thyroid tumor were as follows: (1) complete medical records; (2) operated on by the same medical team; and (3) thyroid follicular adenoma or adenomatous hyperplasia confirmed on pathology. Normal sample histology was evaluated by dissecting non-neoplastic tissue from the contralateral lobe of patients with benign disease who had undergone thyroidectomy. In 4
Page 4 of 23
the case of tumors, specimens were obtained by the dissection of neoplastic tissues. Tissue samples were frozen in liquid nitrogen and transferred to -80 °C until further use. To minimize sample-to-sample variation, each group was divided into three repeated biological subgroups: NH, combining 4, 3 and 3 samples; BTT, combining 2, 3 and 3 samples; PTC-LNM(-), combining 4, 4 and 3 samples; and PTC-LNM(+), combining 4, 4 and
ip t
3 samples. Sample preparation
cr
Proteins were extracted using the SDT lysis method with SDT buffer (4% SDS, 1 mM DTT,
us
150 mM Tris-HCl, pH 8.0) and sonicated in an ice bath for 15 min. After centrifugation at 14,000×g for 40 min, the supernatant was filtered with 0.22-µm filters. The filtrate was quantified using the BCA Protein Assay Kit (Bio-Rad, USA). An equivalent of 200 µg
an
protein from each sample was digested using the filter-aided sample preparation (FASP) method as described previously [20], and 100 μg was labeled using iTRAQ reagent
M
according to the manufacturer’s instructions (Applied Biosystems). The iTRAQ-labeled peptide was divided into 15 fractions using the strong cation exchange (SCX)
ed
chromatography method [21], and each fraction was desalted on C18 cartridges (Empore™ SPE Cartridges C18, bed I.D. 7 mm, volume 3 mL, Sigma) and reconstituted in
ce pt
40 µL of 0.1% (v/v) formic acid for subsequent analysis. LC-MS/MS analysis
Each fraction was injected into Easy nLC (Proxeon Biosystems (now Thermo Fisher
Ac
Scientific)) and MS/MS analysis was performed on a Q Exactive mass spectrometer (Thermo Scientific) for 60 min. The mass spectrometer was operated in positive ion mode. MS data were acquired by choosing the 10 most abundant precursor ions from the survey scan (300-1800 m/z) for HCD fragmentation. The automatic gain control (AGC) target was set to 3e6, and maximum injection time to 10 ms. Dynamic exclusion duration was 40.0 s. The survey scans for MS and MS/MS ions were acquired at a resolution of 70,000 at m/z 200 and 17,500 at m/z 200, respectively. Isolation width was 2 m/z. Normalized collision energy was 30 eV, and the underfill ratio, defining a minimum percentage of target value, was similar to the maximum fill time, set at 0.1%. Raw MS/MS spectra data were searched 5
Page 5 of 23
using a MASCOT engine (version 2.2; Matrix Science, London, UK) embedded into a Proteome Discoverer 1.4 (Thermo Fisher Scientific Inc. 2012). The acquired MS/MS spectra were automatically searched against the Uniprot Homo sapiens database (downloaded from the Uniprot database on April 1, 2017, and including 173,050 protein sequences). Search settings were: enzyme=trypsin; max. missed cleavage=2; variable
ip t
modifications: oxidation (M), iTRAQ 4plex (Y); fixed modifications: carbamidomethyl (C), iTRAQ 4plex (K), iTRAQ 4plex (N-term); peptide mass tolerance=±20 ppm; fragment mass
cr
tolerance=0.1 Da; and peptide false discovery rate (FDR) <0.01, protein FDR<0.01.
Protein expression ratios were calculated as the median number of distinct peptides of the
us
protein after normalizing peptide intensity by the median value of the intensity of all the proteins using the Proteome Discoverer 1.4 software. To screen significant protein
an
expression changes, a fold-change of >1.2 or <0.83 and P-value of <0.05 on Student t test were set as cutoff values.
M
Bioinformatics analysis
To determine the functional classification and biological properties of the selected
ed
differentially abundant proteins, the identified protein sequences were mapped onto GO Terms and annotated using BLAST2GO software (version 3.3.5). The annotation
ce pt
configuration comprised: E-value filter 1e-6, default gradual EC weights, GO weight 5, and annotation cutoff 75. Non-annotated sequences were then re-annotated with more permissive parameters. Sequences without BLAST hits and non-annotated sequences were run through an InterProScan10 program against the EBI databases to retrieve
Ac
functional annotations of protein motifs, then the InterProScan GO terms were merged into the annotation set. In addition, the FASTA protein sequences of differentially changed proteins were blasted against the online Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://geneontology.org/) to retrieve their Kos, and were subsequently mapped onto the KEGG pathways. To further explore the impact of the differentially expressed proteins, enrichment analysis was performed. GO enrichment on the three ontologies (biological process, molecular function, and cellular component) and KEGG pathway enrichment analyses were applied, based on Fisher exact test, considering the annotations of the whole quantified protein as the background dataset. Only functional 6
Page 6 of 23
categories and pathways with P-values <0.05 were considered significantly enriched. The protein-protein interaction information was retrieved from Cytoscape5 software (http://www.cytoscape.org/; version 3.2.1) based on the STRING database (http://stringdb.org) (Supplementary Tables S1 and S2). Immunohistochemistry (IHC)
ip t
To verify the reliability of the iTRAQ results and the general expression of the differentially expressed proteins in other PTMC specimens, IHC was performed on the other 16 patient
cr
PTMC tissues (8 PTMC patients with 8 PTMC patients without LNM). The inclusion criteria
us
for PTMC detected by IHC were similar to those described above. The differentially expressed proteins were examined using the streptavidin-peroxidase (SP) method according to the manufacturer’s instructions. Antigens were successively incubated with
an
primary antibodies, biotin-labeled secondary antibodies, and the HRP-labeled streptavidin. PBS rather than primary antibodies was used as negative control. Brown-yellow staining of
M
the nucleus was considered as positive expression in the cells. Images were analyzed using the Image-Pro Plus 6.0 system (Media Cybernetics Inc., USA). IOD (integrated
ed
optical density), area and mean density (IOD (sum)/area (sum)) were measured.
ce pt
Results
Identification results and overall distribution of proteins Fig. 1 shows sample preparation and analysis in each group. Twelve samples of normal thyroid tissue, benign thyroid tumor tissue, and PTMC tissues with and without LNM were
Ac
identified by iTRAQ. Finally, 49,037 peptide segments and 8,036 proteins (Supplementary Tables S3 and S4) were identified. Quantitative analysis of proteins Proteins were screened based on >1.2-fold (up- or down-regulation) difference in expression, and those with P-value <0.05 were considered differentially expressed. The number of differentially expressed proteins in the BTT and NH, PTC-LNM(-) and NH, PTCLNM(+) and NH, PTC-LNM(-) and BTT, PTC-LNM(+) and BTT, and PTC-LNM(+) and PTCLNM(-) groups are shown in Table 1. Comparison between PTC-LNM(-) and PTC-LNM(+) 7
Page 7 of 23
included 67 upregulated and 82 downregulated proteins. Identified proteins were further screened according to factors such as differentially expression, differential expression multiples and protein function analysis in the various groups. Finally, 26 differential proteins were obtained that might be related to PTMC metastasis, of which 13 were upregulated and 13 downregulated. The 13 upregulated proteins were: alpha-1-antitrypsin,
ip t
tenascin C (hexabrachion), microtubule-actin cross-linking factor 1, nuclear respiratory factor 1, 14-3-3 protein sigma, annexin A1, thrombospondin-1, high-mobility group protein
cr
HMGI-C (HMGA2), hepatoma-derived growth factor (HDGF), immunoglobulin heavy
variable 5-10-1, Ras-related protein Ral-B, pro-cathepsin H, and thrombospondin-2. The
us
13 downregulated proteins were: sushi-domain-containing protein 2, nostrin, ubiquitin carboxyl-terminal hydrolase 7, proto-oncogene tyrosine-protein kinase Src, vascular
an
endothelial growth-factor receptor 2 (VEGFR2), IQ motif-containing GTPase-activating protein 2 (IQGAP2), alpha-actinin 1 (ACTN1), RNA-binding protein 4B, carbonic
M
anhydrase 4 (CA4), thyroid peroxidase, pyruvate dehydrogenase E1, cellular retinoic-acidbinding protein 1, and myosin-reactive immunoglobulin light-chain variable region (Table
Bioinformatics analysis
ce pt
Cluster analysis
ed
2).
Clustering revealed high similarity within groups and low similarity between groups. Cluster analyses of four groups of samples demonstrated that the differentially expressed proteins identified experimentally were rational and accurate; the X and Y axes represented the
Ac
samples and the differentially expressed proteins, respectively (Supplemental Fig. 1). Functional annotation and analysis by GO The differentially expressed proteins in each group were annotated based on three aspects: the biological processes involved, the molecular functions, and the cell components. Significant enrichment analysis of GO annotations was evaluated by Fisher’s exact test (P<0.01). The differential proteins associated with PTMC metastasis involved 23 biological processes, 11 molecular functions, and 16 cell components (Fig. 2). In addition, the biological processes following enrichment mainly comprised chemical homeostasis 8
Page 8 of 23
within a tissue, zonula adherens maintenance, surfactant homeostasis, single-multicellular organism process, and multicellular organism process, while molecular functions comprised transforming growth-factor beta binding, and cell components mainly comprised zonula adherens (Supplemental Fig. 2). KEGG pathway analysis
ip t
All types of regulatory protein exerted a role in the network by interacting, co-operating, and promoting each other. Thus, a complex regulatory network was formed that executed
cr
the occurrence, development and progression of PTC. In addition, the KEGG pathway
us
systematically and comprehensively analyzed the target protein set to determine the affected signal transduction pathway. The KEGG pathway analysis was conducted on the differential proteins listed above, indicating that 13 pathways were associated with PTMC
an
metastasis: 11 upregulated pathways comprising focal adhesion, Rap1 signaling, ECMreceptor interaction, microRNAs in cancer, VEGF signaling, adherens junction, EGFR
M
tyrosine kinase inhibitor resistance, PI3K-Akt signaling, proteoglycans in cancer, actin cytoskeleton regulation, and Ras signaling, and 2 downregulated pathways comprising
ed
FoxO signaling and HIF-1 signaling (Supplemental Fig. 3). The above-mentioned KEGG pathways were enriched and analyzed based on P-values. The focal adhesion pathway
ce pt
exhibited a minimal P-value, and 6 differentially expressed proteins were identified: tenascin C, thrombospondin-1, thrombospondin-2, proto-oncogene tyrosine-protein kinase Src, VEGFR2, and ACTN1.
Ac
IHC verification
To check the reliability of the iTRAQ results and the general expression of the differentially expressed proteins in other PTMC specimens, 4 differential proteins were selected for IHC assessment. Results further confirmed that the expression of HDGF and HMGA2 was significantly higher in the LNM group than in the non-LNM group (P<0.01), while expression of IQGAP2 and CA4 was significantly lower in the LNM group than in the nonLNM group (P<0.01) (Table 3) (Fig. 3). These IHC results were in agreement with those of iTRAQ and were confirmed in other PTMC samples, which in turn substantiated the reliability of the iTRAQ results and the ubiquitous differential expression of the proteins. 9
Page 9 of 23
Discussion Currently, PTMC is defined as a tumor with size ≤1cm; however, invasiveness and metastatic capacity are not clearly understood. The necessity and scope of surgery for PTMC remain controversial. Some investigators advocate “over-diagnosis” and “overtreatment” for PTMC [22-27]. Thus, a small number of PTMCs might occur in a subclinical
ip t
state of non-progression. However, local invasion, LNM and distant metastasis might also occur in the early stage of PTMC [17,22,23,28]. In addition, any advanced PTC has
cr
progressed from PTMC, rather than having late onset. These phenomena suggest that PTMC exhibits different biological behaviors, despite pathological and anatomical factors
us
being identical. These different biological behaviors can provide a basis for individualized treatment of PTMC, which is crucial for determining the reasonable scope of surgery and
an
reducing “overtreatment.” We therefore conducted proteomic analyses using PTMC tissue without capsule invasion from single carcinoma lesions with tumor diameter 0.5–1 cm.
M
This excluded the above anatomical factors, enabling understanding of the molecular biological changes involved in PTMC, in order to provide novel information on the
ed
mechanism underlying PTC invasion and metastasis. Proteomic studies compared normal thyroid tissue, benign thyroid tumor tissue, PTC
ce pt
tissue with LNM, and PTC tissue without LNM, and some of the screened proteins were verified. Results demonstrated that the differentially expressed proteins identified by comparing the NH and BTT groups with the PTC-LNM(-) group independently, might be associated with cancerization, while those identified by comparing the NH and BTT groups
Ac
with the PTC-LNM(+) group independently might be associated with cancerization and metastasis, and those identified by comparing the PTC-LNM(-) and PTC-LNM(+) groups might be associated with metastasis. The identified proteins were further screened according to the differentially expressed proteins, differential expression multiplex, protein function analysis and background factors of thyroid carcinoma, with some screened proteins obtained from the other PTMC specimens serving as candidates for verification. Finally, the differential proteins and signaling pathways that might be closely associated with PTMC metastasis were found to be ACTN1, alpha-1-antitrypsin, HMGA2, HDGF, 10
Page 10 of 23
carbonic anhydrase 4, immunoglobulin heavy variable 5-10-1, myosin-reactive immunoglobulin light-chain variable region, and focal adhesion pathway. ACTN1 plays a critical role in cell adhesion, cell-shape regulation, and cell movement [29]. Moreover, tumor onset and development are dependent on the alpha-actinin involved in the cytoskeleton structure. Hirooka et al. found that the expression level of ACTN1 mRNA
ip t
in lung adenocarcinoma tissue was significantly lower than in normal lung tissue [30]. Glück et al. confirmed that increased ACTN1 expression in 3T3 cells reduced cell
cr
migration, while inhibition of expression by antisense transfection enhanced cell movement [31]. These phenomena prompted the downregulation of ACTN1 that is involved in
us
skeleton assembly and cell adhesion activity in transformed cell lines and tumor tissues. The current study found that ACTN1 expression in PTMC tissues with LNM was
an
downregulated and not only localized at the critical nodes of the PPI network but also closely related to the focal adhesion pathway. Thus, we speculated that ACTN1 may be a
M
significant regulator of PTMC metastasis.
Alpha-1-antitrypsin (alpha 1-AT) plays a major role in the stability of local tissues and
ed
tumorigenesis [32]. Alpha 1-AT content is significantly elevated in several tumor tissues and in serum from liver, pancreatic or gastric cancer [33-35]. Tzonou et al. confirmed that
ce pt
elevated serum α1-AT levels are associated with short-term survival in liver cancer patients [36]. Poblete et al. evaluated α1-AT expression in PTC tissue [37]. The present study demonstrated expression of α1-AT in PTMC tissue, and that upregulation promoted
Ac
PTMC LNM.
HMGI-C, also known as HMGA2, is associated with a variety of diseases, such as developmental disorders, malignant tumor, benign stromal tumor, obesity, diabetes, atherosclerosis and arthritis [38]. Kumar et al. developed a series of HMGA2 expression structures and compared the expression types of Let-7 and HMGA2; interaction between HMGA2 mRNA and let-7 was identified as the key step in carcinogenesis [39]. In addition, Jun et al. confirmed that HMGA2 was associated with onset of gastric cancer and promoted tumor progression, and might thus be used as a predictor of tumor recurrence [40]. Similarly, overexpression of HMGA2 was associated with growth and invasion in 11
Page 11 of 23
breast cancer [41]. In addition, microRNA Let-7b inhibited PTC by downregulating expression of HMGA2 [42]. In the present study, HMGA2 expression was found to be upregulated in PTMC tissue with LNM. We therefore speculated that HMGA2 could be a predictor of invasion, metastasis and prognosis in PTC. HDGF is related to the onset and development of several types of tumor: liver cancer, lung
ip t
cancer, cholangiocarcinoma, gastric cancer and prostate cancer; increased expression in malignant tumors is also associated with prognosis [43-45]. Targeted inhibition of HDGF
cr
effectively reduced hepatocellular carcinoma growth in in vitro and in vivo experiments [46]. Yang et al. speculated that the role of HDGF in hepatocellular carcinoma might be
us
related to the ERK signaling pathway [47]. A previous study using NIH/3T3 cells demonstrated that HDGF promotes cytoskeleton remodeling and migration by activating
an
the PI3K/Akt signaling pathway [48]; however, correlation studies between HDGF and thyroid carcinoma are still lacking. Our data showed that HDGF was expressed in PTMC
M
tissues, and that expression was significantly higher in PTMC with than without LNM. The difference in the expression multiplex of HDGF was the greatest of all differentially
ed
expressed proteins, indicating that HDGF plays a critical role in PTC invasion and metastasis. However, the KEGG pathway analysis could not identify the HDGF-associated
ce pt
signaling pathways or the specific upstream and downstream signaling molecules. Thus, HDGF might serve as an independent risk factor for PTC invasion and metastasis and could be a target for physiological treatment. The focal adhesion pathway directly or indirectly regulates cytoskeleton remodeling,
Ac
epithelial-mesenchymal transition and changes in a variety of related signaling pathways [49,50]. MicroRNA-218 inhibits renal carcinoma cell invasion and metastasis by regulating the focal adhesion pathway [51]. Furthermore, Talin1 promoted prostate carcinoma cell invasion and metastasis via the focal adhesion pathway in in vivo and in vitro experiments [52]. Also, in the present study, the activated signaling pathway associated with PTMC metastasis included focal adhesion, Rap1 signaling pathway, ECM-receptor interaction, microRNAs in cancer, VEGF signaling pathway, adherens junction, EGFR tyrosine kinase inhibitor resistance, PI3K-Akt signaling pathway, proteoglycans in cancer, regulation of actin cytoskeleton, and Ras signaling pathway, of which the enriched focal adhesion 12
Page 12 of 23
pathway might be a vital regulatory signaling pathway. Focal adhesion pathway activation not only promotes tumor cell invasion and metastasis but also regulates the PI3K-Akt signaling pathway, actin cytoskeleton, ECM-receptor interaction, VEGF signaling pathway, and Rap1 signaling pathway. Additionally, differentially expressed proteins such as tenascin C, thrombospondin-1, thrombospondin-2 and alpha-actinin-1 were closely
ip t
associated with the focal adhesion pathway. These results suggest that the focal adhesion pathway might be a critical upstream signaling pathway that regulates PTMC invasion and
cr
metastasis. In conclusion, these findings may provide deep insight into the mechanisms underlying PTMC invasion and metastasis. Further analysis of the present experimental
us
results concluded that upregulated expression of hepatoma-derived growth factor and high-mobility group protein HMGI-C and activation of the focal adhesion pathway promote
an
lymph node metastasis in PTMC. Data Availability Statement
M
The proteomics data used to support the findings of this study have been deposited in the
Disclosure Summary
ed
iProX (ww.iprox.org) repository (ID IPX0001247000 /PXD010302).
Funding
ce pt
The authors have no conflicts of interest to disclose.
The study was supported by The Inner Mongolia Autonomous Region “Prairie Outstanding
Ac
Ability Person” Project.
13
Page 13 of 23
References [1] Siegel RL, et al. Cancer Statistics, 2017. CA Cancer J Clin.2017; 67:7-30. [2] Jin S, et al. Signaling Pathways in Thyroid Cancer. Vitam Horm. 2018; 106:501-515. [3] Tang J, et al. Clinical-pathological Characteristics and Prognostic Factors for Papillary Thyroid Microcarcinoma in the Elderly. J Cancer. 2018; 9:256-262.
ip t
[4] Davies L, Welch HG. Current thyroid cancer trends in the United States. JAMA Otolaryngol Head Neck Surg. 2014; 140:317-322.
cr
[5] Lee JH, Shin SW. Overdiagnosis and screening for thyroid cancer in Korea. Lancet. 2014; 22:1848.
us
[6] Mutalib NS, et al. MicroRNAs and lymph node metastasis in papillary thyroid cancers. Asian Pacific Journal of Cancer Prevention. 2016; 17:25-35.
an
[7] Sun W, et al. Risk Factors for central lymph node metastasis in CN0 papillary thyroid carcinoma: a systematic review and meta-analysis. PLoS One. 2015; 10:e0139021. [8] Wu AW, et al. What is the best treatment for papillary thyroid microcarcinoma? Laryngoscope. 2011; 121:1828-1829.
M
[9] Woo CG, et al. Clinicopathological Significance of Minimal Extrathyroid Extension in Solitary Papillary Thyroid Carcinomas. Ann Surg Oncol. 2015; 22:S728-733.
ed
[10] Zhang L, et al. The clinical prognosis of patients with cN0 papillary thyroid microcarcinoma by central neck dissection. World J Surg Oncol. 2015; 7:138.
ce pt
[11] Yu X, et al. Independent Risk Factors Predicting Central Lymph Node Metastasis in Papillary Thyroid Microcarcinoma. Horm Metab Res. 2017;49; :201-207. [12] Takano T. Natural history of thyroid cancer. Endocr J. 2017; 64:237-244. [13] Grogan RH, et al. A study of recurrence and death from papillary thyroid cancer with 27 years of median follow-up. Surgery. 2013; 154:1436-1447.
Ac
[14] Yang J, et al. Risk factors for level V lymph node metastases in solitary papillary thyroid carcinoma with clinically lateral lymph node metastases. Cancer Medicine. 2016; 5:2161-2168. [15] Chéreau N, et al. Recurrence of papillary thyroid carcinoma with lateral cervical node metastases: Predictive factors and operative management. Surgery. 2016; 159:755-762. [16] Siddiqui S, et al. Clinical and Pathologic Predictors of Lymph Node Metastasis and Recurrence in Papillary Thyroid Microcarcinoma. Thyroid. 2016; 26:807-815. [17] Xu Y, et al. Clinical predictors of lymph node metastasis and survival rate in papillary thyroid microcarcinoma: analysis of 3607 patients at a single institution. J Surg Res. 2018; 221:128-134. 14
Page 14 of 23
[18] Jin S, et al. Signaling pathways in thyroid cancer and their therapeutic implications. J Cli Med Res. 2016; 8:284-296. [19] Gao M, et al. 2016 Chinese expert consensus and guidelines for the diagnosis and treatment of papillary thyroid microcarcinoma. Cancer Biol Med. 2017; 14:203-211. [20] Wiśniewski JR, et al. Universal sample preparation method for proteome analysis. Nat Methods. 2009; 6:359-362.
ip t
[21] Chan KC, Issaq HJ. Fractionation of peptides by strong cation-exchange liquid chromatography. Methods Mol Biol. 2013; 1002:311-315.
cr
[22] Jeon MJ, et al. Features of papillary thyroid microcarcinoma associated with lateral cervical lymph node metastasis. Clin Endocrinol (Oxf). 2017; 86:845-851.
us
[23] Ito Y, et al. Patient age is significantly related to the progression of papillary microcarcinoma of the thyroid under observation. Thyroid. 2014; 24:27-34.
an
[24] Gschwandtner E, et al. Increase of papillary thyroid microcarcinoma and a plea for restrictive treatment: A retrospective study of 1,391 prospective documented patients. Surgery. 2016; 159:503-511.
M
[25] Ahn HS, et al. Korea's thyroid-cancer "epidemic"--screening and overdiagnosis. N Engl J Med. 2014; 371:1765-1767.
ed
[26] Kim TY, Shong YK. Active Surveillance of Papillary Thyroid Microcarcinoma: A MiniReview from Korea. Endocrinol Metab (Seoul). 2017; 32:399-406. [27] Weber T, et al. Surgery of papillary thyroid microcarcinoma. Der Chirurg.2018; 1:1-6.
ce pt
[28] Li M, et al. Risk factors for predicting central lymph node metastasis in papillary thyroid microcarcinoma (CN0): a study of 273 resections. Eur Rev Med Pharmacol Sci. 2017; 21:3801-3807.
Ac
[29] Hamill KJ, et al. Alpha actinin-1 regulates cell-matrix adhesion organization in keratinocytes: consequences for skin cell motility. J Invest Dermatol. 2015; 135:10431052. [30] Hirooka S, et al. Localization of the invadopodia-related proteins actinin-1 and cortactin to matrix-contact-side cytoplasm of cancer cells in surgically resected lung adenocarcinomas. Pathobiology. 2011; 78:10-23. [31] Glück U, Ben-Ze'ev A. Modulation of alpha-actinin levels affects cell motility and confers tumorigenicity on 3T3 cells. J Cell Sci. 1994; 107:1773-1782. [32] De Serres F, Blanco I. Role of alpha-1 antitrypsin in human health and disease. Journal of Internal Medicine.2014; 276:311-335. [33] Topic A, et al. Alpha-1-antitrypsin in pathogenesis of hepatocellular carcinoma. Hepat Mon. 2012; 12:e7042. 15
Page 15 of 23
[34] Wang Y, et al. Screening for serological biomarkers of pancreatic cancer by twodimensional electrophoresis and liquid chromatography-tandem mass spectrometry. Oncol Rep. 2011; 26:287-292. [35] Wu W, et al. S100A9, GIF and AAT as potential combinatorial biomarkers in gastric cancer diagnosis and prognosis. Proteomics Clin Appl. 2012; 6:152-162.
ip t
[36] Tzonou A, et al. Alpha 1-antitrypsin and survival in hepatocellular carcinoma. Br J Cancer. 1990; 61:72-73. [37] Poblete MT, et al. Alpha 1-antitrypsin expression in human thyroid papillary carcinoma. Am J Surg Pathol. 1996; 20:956-63.
cr
[38] Young AR, Narita M. Oncogenic HMGA2: short or small? Genes Dev. 2007; 21:10051009.
us
[39] Kumar MS, et al. HMGA2 functions as a competing endogenous RNA to promote lung cancer progression. Nature. 2014; 505:212-217.
an
[40] Jun KH, et al. HMGA1/HMGA2 protein expression and prognostic implications in gastric cancer. Int J Surg. 2015; 505:39-44.
M
[41] Zou Q, et al. RKIP suppresses the proliferation and metastasis of breast cancer cell lines through up-regulation of miR-185 targeting HMGA2. Arch Biochem Biophys. 2016; 610:25-32.
ed
[42] Li H, et al. Roles of microRNA let-7b in papillary thyroid carcinoma by regulating HMGA2. Tumour Biol. 2017; 39(10): 1010428317719274.
ce pt
[43] Enomoto H, et al. Hepatoma-Derived Growth Factor: Its Possible Involvement in the Progression of Hepatocellular Carcinoma. Int J Mol Sci. 2015; 16:14086-14097. [44] Bao CH, et al. Prognostic role of hepatoma-derived growth factor in solid tumors of Eastern Asia: a systematic review and meta- analysis. Asian Pac J Cancer Prev. 2015; 16:1803-1811.
Ac
[45] Shetty A, et al. Hepatoma-derived growth factor: A survival-related protein in prostate oncogenesis and a potential target for vitamin K2. Urol Oncol. 2016; 34:483.e1-483.e8. [46] Enomoto H, et al. Down regulation of HDGF Inhibits the Growth of Hepatocellular Carcinoma Cells In Vitro and In Vivo. Anticancer Res. 2015; 35:6475-6479. [47] Yang GY, et al. Hepatoma-derived growth factor promotes growth and metastasis of hepatocellular carcinoma cells. Cell Biochem Funct. 2016; 34:274-285. [48] Kung ML, et al. Hepatoma-derived growth factor stimulates podosome rosettes formation in NIH/3T3 cells through the activation of phosphatidylinositol 3-kinase/Akt pathway. Biochem Biophys Res Commun. 2012; 425:169-176.
16
Page 16 of 23
[49] Eke I, Cordes N. Focal adhesion signaling and therapy resistance in cancer. Semin Cancer Biol. 2015; 31:65-75. [50] Maziveyi M, Alahari SK. Cell matrix adhesions in cancer: The proteins that form the glue. Oncotarget. 2017; 8:48471-48487.
ip t
[51] Yamasaki T, et al. MicroRNA-218 inhibits cell migration and invasion in renal cell carcinoma through targeting caveolin-2 involved in focal adhesion pathway. J Urol. 2013; 190:1059-1068. [52] Sakamoto S, et al. Talin1 promotes tumor invasion and metastasis via focal adhesion signaling and anoikis resistance. Cancer Res. 2010; 70:1885-1895.
cr
Figure Legends
us
Figure 1. Schematic of the experimental procedure.
ce pt
Figure 2. GO annotation.
ed
M
an
Total protein was extracted from freshly frozen tissues from NH, BTT, PTC-LNM(-) and PTC-LNM(+) samples using the SDT lysis method and digested using the FASP method; The peptide mixture of each sample was labeled using iTRAQ reagent. The iTRAQlabeled peptide was further divided using SCX chromatography. Finally, LC-MS/MS was used for high-throughput screening of the sample. Data were analyzed by bioinformatics tools from different aspects. Top check the reliability of the results, IHC was performed for other patient tissues of PTMC. (NH: normal histology; BTT: benign thyroid tumor; PTCLNM(-): classic papillary thyroid microcarcinoma without LNM; PTC-LNM(+): classic papillary thyroid microcarcinoma with LNM; FASP: filter-aided sample preparation; SCX: strong cation exchange; IHC: immunohistochemistry).
The differentially expressed proteins in each group were annotated based on the three aspects: the biological processes involved, the molecular functions, and the cell components.
Ac
Figure 3. Validation of protein expression in PTMC. To check the reliability of the iTRAQ results and the general expression of the differentially expressed proteins in other PTMC specimens, 4 differential proteins were selected for IHC. Cells in which the cytoplasm was stained brown-yellow were considered to be HDGF, HMGA2-, and IQGAP2-positive cells. If the nucleus was stained brown, the cells were considered to be CA4-positive cells. A: HDGF expression (LNM group); B: HDGF expression (non-LNM group); C: HMGA2 expression (LNM group); D: HMGA2 expression (non-LNM group); E: CA4 expression (LNM group); F: CA4 expression (non-LNM group); G: IQGAP2 expression (LNM group); H: IQGAP2 expression (non-LNM group). (HDGF: hepatoma-derived growth factor; HMGA2: high-mobility group protein HMGI-C; CA4: carbonic anhydrase 4; IQGAP2: GTPase-activating protein 2). Table1.Overview of the number of differentially expressed proteins 17
Page 17 of 23
Up-
Down-
All-
BTT/NH
458
563
1021
PTC-LNM(-)/NH
323
346
669
PTC-LNM(+)/ NH
595
598
1193
PTC-LNM(-)/BTT
369
275
644
PTC-LNM(+)/BTT
499
383
882
PTC-LNM(+)/PTC-LNM(-)
67
82
149
cr
ip t
Comparisons
us
Comparisons. Comparison groups: Up-: Upregulated differentially expressed proteins; Down-: Downregulated differentially expressed proteins; All: All differentially expressed proteins. Table 2. Differentially expressed proteins associated with PTMC metastasis
Covera Peptide MW ge s (kDa)
Protein
M
Protein ID
an
Single
A0A024R 6I7 Alpha-1-antitrypsin
58.61 2
46.68
Fold t-test Calc. pI change P-value 5.59
3.92018 0.00492 3 1
240.8
4.92
1.54226 0.00922 6 7
Microtubule-actin crossE9PLY5 linking factor 1
159.8
5.66
1.22338 0.00998 6 5
ed
A0A024R 884 Tenascin C (hexabrachion) 30.35 2
ce pt
16.51 1 1
22.68
5.95
1.24943 0.01126 4 4
47.98 7
27.76
4.74
1.46705 0.01323 4 1
7.02
1.33810 0.01410 3 2
4.94
1.40608 0.02414 6 9
8.27
2.55509 0.02597 7 0
10.6
2.15147 0.02873 2 7
C9JP85 Nuclear respiratory factor 1 4.31
P04083 P07996
14-3-3 protein sigma
Ac
P31947
Annexin A1
Thrombospondin-1
A0A0J9Y Immunoglobulin heavy XX1 variable 5-10-1 P52926
67.63 23 23.25 22 26.5
High-mobility group protein HMGI-C 57.8
1 5
38.69 129.3 12.7 11.83
18
Page 18 of 23
A0A087X 0D5 Pro-cathepsin H P35442
26.7
4
16.72 6
Thrombospondin-2
Sushi-domain-containing Q9UGT4 protein 2
90.15
6.28
P35968
Vascular endothelial growth factor receptor 2 3.32
ed
9.51
2 2
ce pt
A0A024R IQ motif containing GTPase AJ8 activating protein 2 33.71 43
Ac
Carbonic anhydrase 4
62.44 2 18.66 1 25.96 8
0.81471 0.01449 7 5
9.36
0.79401 0.01785 6 5
8.15
0.73934 0.01903 5 4
128.2
5.55
0.78006 0.01540 5 7
59.8
7.42
0.75258 0.01900 7 3
151.4
5.85
0.79197 0.03597 6 3
180.5
5.58
0.81172 0.03805 2 7
5.41
0.80465 0.04123 6 5
6.74
0.67211 0.04114 9 1
7.83
0.59967 0.00485 1 3
11.62
an
1
M
Proto-oncogene tyrosineprotein kinase Src
P22748
1.43338 0.04403 5 2
13.02 7
P12931
Q9BQ04 RNA-binding protein 4B
8.1
1.42785 0.04389 1 4
4.83
6.12
Alpha-actinin-1
6.62
129.9
Ubiquitin carboxyl-terminal Q93009 hydrolase 7 32.76 3
P12814
36.25
4.62343 0.03871 3 1 1.33063 0.04320 8 5
15.78 14
Myosin-reactive Q9UL70 37.96 2 immunoglobulin light chain variable region A0A0G2J M79 Nostrin (Fragment)
23.39
4.67
ip t
Ras-related protein Ral-B
26.79
cr
P11234
Hepatoma-derived growth 79.17 1 factor (high-mobility group protein 1-like)
us
B2RDE8
44.81
103 40.12 35.01
A0M8X1 Thyroid peroxidase
40.41 1
102.9
6.76
0.48434 0.00838 1 1
Pyruvate dehydrogenase Q53GE3 E1
38.21 1
43.19
8.06
0.18339 0.02038 3 4
5.38
0.65274 5 0.01492
P29762
Cellular retinoic acidbinding protein 1
57.66 8
15.56
19
Page 19 of 23
Table 3. Mean density of expression of HDGF, HMGA2, CA4, and IQGAP2 proteins in PTMC Group (n=8)
PTMC without lymph node metastasis
0.280±0.019
0.051±0.002＊
HMGA2
0.234±0.014
0.069±0.003＊
CA4
0.049±0.003
0.147±0.019＊
IQGAP2
0.067±0.006
0.198±0.021＊
M
an
us
cr
HDGF
ip t
PTMC with lymph node metastasis
*P<0.01,
Ac
ce pt
ed
vs. PTMC with lymph node metastasis group; HDGF, hepatoma-derived growth factor; HMGA2, high-mobility group protein HMGI-C; CA4, carbonic anhydrase 4; IQGAP2, IQ motif-containing GTPase-activating protein 2
20
Page 20 of 23
ip t cr us an M ed ce pt Ac
Figr-1
21
Page 21 of 23
ip t cr us an M Ac
ce pt
ed
Figr-2
22
Page 22 of 23
ip t cr Ac
ce pt
ed
M
an
us
Figr-3
23
Page 23 of 23
S0003-4266(19)30028-9 https://doi.org/doi:10.1016/j.ando.2019.01.003 ANDO 1109
To appear in:
Annales d’Endocrinologie
Please cite this article as: Jin S, Bao W, Yang Y-Tian, Quan F, Bai Y, Liu Y, Proteomic analysis of the papillary thyroid microcarcinoma, Annales d’Endocrinologie (2019), https://doi.org/10.1016/j.ando.2019.01.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
M
an
us
cr
ip t
Proteomic analysis of the papillary thyroid microcarcinoma Analyse protéomique du microcarcinome papillaire de la thyroïde Shan Jin1, Wuyuntu Bao1, Yun-Tian Yang1, Quan Fu2, Yinbao Bai1, Yousheng Liu1 1 Department of General Surgery, Affiliated Hospital of Inner Mongolia Medical University, Hohhot, Inner Mongolia Autonomous Region, China. 2 Department of Clinical Laboratory, Affiliated Hospital of Inner Mongolia Medical University, Hohhot, Inner Mongolia Autonomous Region, China. Corresponding author: Shan Jin, MD, PhD, Department of General Surgery, Affiliated Hospital of Inner Mongolia Medical University, Tongdao North Rd 1, Hohhot 010050, Inner Mongolia Autonomous Region, China; Tel: +86-471-3451465 Fax: +86-471-3451465 E-mail: [email protected] Wuyuntu Bao E-mail: [email protected]; Yun-Tian Yang E-mail: [email protected]; Quan Fu E-mail: [email protected] ; Yinbao Bai E-mail: [email protected]; Yousheng Liu E-mail: [email protected]
ed
Abstract
Objective: The present study applied iTRAQ and LC-MS/MS techniques for proteome analysis and compared data between specimens of papillary thyroid microcarcinoma
ce pt
(PTMC) vs appropriate controls, in order to investigate the mechanisms underlying the invasion and metastasis process in PTMC development. Materials and methods: Fresh-tissue specimens were collected from 40 patients with
Ac
thyroid disease who underwent surgical treatment. Specimens were divided into four groups: normal histology (NH; n=8), benign thyroid tumor (BTT; n=10), classic PTMC with lymph node metastasis (PTC-LNM(+); n=11), and classic PTMC without lymph node metastasis (PTC-LNM(-); n=11). Proteomic studies were conducted on PTMC tissue samples without capsule invasion and with tumor diameter ranging from 0.5 cm to 1 cm, so as to focus the study on PTMC development excluding metastasis. Results: A total of 8,036 proteins were identified in the four groups. Based on protein function analysis, proteins that might be associated with PTMC invasion and metastasis were screened: alpha-actinin-1, alpha-1-antitrypsin, hepatoma-derived growth factor 1
Page 1 of 23
(HDGF), high-mobility group protein HMGI-C, and carbonic anhydrase 4. In addition, proteins involved in the focal adhesion pathway were examined. Immunohistochemistry confirmed the reliability of the iTRAQ results and the universality of differentially expressed proteins. The data showed that HDGF and high-mobility group protein HMGI-C are upregulated in PTMC and that the focal adhesion pathway that promotes PTMC LNM is
ip t
activated. Conclusions: These findings provide insight into the mechanisms underlying PTMC
cr
invasion and metastasis.
Keywords: invasion; metastasis; papillary thyroid microcarcinoma; proteomics
us
Résumé
Contexte : Nous avons analysé le protéome d'échantillons de microcarcinome papillaire
an
de la thyroïde (MCPT) grâce aux techniques iTRAQ (étiquette isobare pour la quantification relative et absolue) et LC-MS/MS (chromatographie en phase liquide
M
couplée à la spectrométrie de masse) afin d’identifier les mécanismes de l’invasion et de la métastase du MCPT.
ed
Objectifs : Des échantillons de tissu frais étaient pris sur 40 patients atteints d’une maladie de la thyroïde, qui ont subi un traitement initial et étaient ensuite divisés en
ce pt
4 groupes : tissus histologiques normaux (HN ; n = 8), tissus présentant une tumeur de la thyroïde bénigne (TTB ; n = 10), tissus présentant un MCPT classique et des métastases ganglionnaires (CPT-MG(+) ; n = 11), tissus présentant un MCPT classique sans
Ac
métastases ganglionnaires (CPT-MG (-) ; n = 11). Les études des différents protéomes ont été réalisées sur des tissus atteints de MCPT sans invasion capsulaire, prélevés dans des carcinomes uniques de diamètre 0,5-1 cm. Ceci exclut les facteurs anatomiques de métastase, fournissant ainsi des informations sur les changements biologiques au niveau moléculaire impliqués dans le MCPT. Résultats : Au total, 8036 protéines ont été identifiées au sein des 4 groupes. À partir de l’analyse de la fonction de ces protéines, celles pouvant être associées à l’invasion et la métastase du MCPT ont été identifies : alpha-actinine-1, alpha-12
Page 2 of 23
antitrypsine, facteur de croissance dérivé de l’hépatome (HDGF), protéine du groupe de haute mobilité HMGI-C, et l’anhydrase carbonique-4. De plus, la voie d’adhésion focale est peut-être impliquée dans l’invasion et la métastase du MCPT. Les techniques d’immunohistochimie ont confirmé la fiabilité des résultats obtenus par la technique iTRAQ ainsi que l’universalité des protéines à expression
ip t
différentielle. Conclusions : Cette étude démontre que la surexpression du facteur HDGF et de la
cr
protéine du groupe de haute mobilité HMGI-C ainsi que l’activation de la voie d’adhésion focale favorisent l’apparition de métastases ganglionnaires associées au MCPT. Ces
us
résultats pourraient améliorer la compréhension des mécanismes de l’invasion et de la
an
métastase du MCPT.
Mots-clés : invasion ; métastase ; micro-carcinome papillaire de la thyroïde ; protéomique
M
Thyroid carcinoma is a common malignancy of the endocrine system, with incidence increasing every year [1,2]. The most common pathological type is papillary thyroid
ed
carcinoma (PTC), which accounts for >90% of thyroid carcinomas, with papillary thyroid microcarcinoma (PTMC) accounting for about 30–43.1% of PTCs [3-5]. Due to the low-
ce pt
grade malignancy of PTC, post-surgical prognosis and the results of postoperative adjuvant treatment are satisfactory and long-term survival is achievable. Nevertheless, 20– 90% of PTC patients present cervical lymph node metastasis (LNM) [6,7]; the rate in PTMC patients is 16.9–54.1% in the central region of the neck, and 3.7-10.1% in the
Ac
lateral neck [8-11]. Some studies have shown that cervical LNM is observed under a microscope in >60% of the patients with PTMC [12]. Moreover, cervical LNM is a major risk factor for increased recurrence and death in patients with thyroid carcinoma [13-15]. PTMC metastasis is related to tumor diameter, multifocality and capsule invasion [16,17]. Some PTMC patients show LNM without anatomic risk factors such as capsule invasion or multifocality, which indicates that the invasiveness and metastasis of PTMC are determined by its biological behavior. Moreover, the onset, development, invasion and metastasis of thyroid carcinoma are closely related to altered gene expression, disruption in the signal transduction pathway, and dysfunction in the associated proteins [2,18]. The 3
Page 3 of 23
present study therefore performed proteomic analysis based on the iTRAQ and LC-MS/MS techniques to further elucidate the functions and check expression of candidate proteins and shed light on the mechanisms underlying PTMC invasion and metastasis. Material and methods
ip t
Ethics This study conformed to the Declaration of Helsinki regarding ethical principles for medical
cr
research in humans. It was approved by the Biomedical Ethics Committee of Inner Mongolia Medical University (No. YKD2014063).
us
Trial registration
an
ClinicalTrials.gov Protocol Registration System: ClinicalTrials.gov ID NCT02814032, registered on June 24, 2016.
M
General data and sample collection
Forty patients with thyroid disease, initially treated in the General Surgery Department of
ed
the Affiliated Hospital of Inner Mongolia Medical University, were enrolled from June 2016 to May 2017. Standard surgery was performed according to the Chinese version of the
ce pt
guidelines for “diagnosing and managing thyroid nodules and differentiated thyroid cancer” [19]. Forty fresh frozen thyroid specimens were classified as: normal histology (NH, n=10), benign thyroid tumor (BTT, n=8), classic PTMC without LNM (PTC-LNM(-), n=11), and classic PTMC with LNM (PTC-LNM(+), n=11). Inclusion criteria for PTMC patients were as
Ac
follows: (1) complete medical records; (2) operated on by the same medical team; (3) confirmed on pathology as classic PTC without capsule invasion from single carcinoma lesions, with tumor diameter 0.5–1 cm; (4) associated central neck lymph node dissection in the same step; (5) without association of neck and other malignant tumors; and (6) without history of head and neck radiotherapy. Inclusion criteria for benign thyroid tumor were as follows: (1) complete medical records; (2) operated on by the same medical team; and (3) thyroid follicular adenoma or adenomatous hyperplasia confirmed on pathology. Normal sample histology was evaluated by dissecting non-neoplastic tissue from the contralateral lobe of patients with benign disease who had undergone thyroidectomy. In 4
Page 4 of 23
the case of tumors, specimens were obtained by the dissection of neoplastic tissues. Tissue samples were frozen in liquid nitrogen and transferred to -80 °C until further use. To minimize sample-to-sample variation, each group was divided into three repeated biological subgroups: NH, combining 4, 3 and 3 samples; BTT, combining 2, 3 and 3 samples; PTC-LNM(-), combining 4, 4 and 3 samples; and PTC-LNM(+), combining 4, 4 and
ip t
3 samples. Sample preparation
cr
Proteins were extracted using the SDT lysis method with SDT buffer (4% SDS, 1 mM DTT,
us
150 mM Tris-HCl, pH 8.0) and sonicated in an ice bath for 15 min. After centrifugation at 14,000×g for 40 min, the supernatant was filtered with 0.22-µm filters. The filtrate was quantified using the BCA Protein Assay Kit (Bio-Rad, USA). An equivalent of 200 µg
an
protein from each sample was digested using the filter-aided sample preparation (FASP) method as described previously [20], and 100 μg was labeled using iTRAQ reagent
M
according to the manufacturer’s instructions (Applied Biosystems). The iTRAQ-labeled peptide was divided into 15 fractions using the strong cation exchange (SCX)
ed
chromatography method [21], and each fraction was desalted on C18 cartridges (Empore™ SPE Cartridges C18, bed I.D. 7 mm, volume 3 mL, Sigma) and reconstituted in
ce pt
40 µL of 0.1% (v/v) formic acid for subsequent analysis. LC-MS/MS analysis
Each fraction was injected into Easy nLC (Proxeon Biosystems (now Thermo Fisher
Ac
Scientific)) and MS/MS analysis was performed on a Q Exactive mass spectrometer (Thermo Scientific) for 60 min. The mass spectrometer was operated in positive ion mode. MS data were acquired by choosing the 10 most abundant precursor ions from the survey scan (300-1800 m/z) for HCD fragmentation. The automatic gain control (AGC) target was set to 3e6, and maximum injection time to 10 ms. Dynamic exclusion duration was 40.0 s. The survey scans for MS and MS/MS ions were acquired at a resolution of 70,000 at m/z 200 and 17,500 at m/z 200, respectively. Isolation width was 2 m/z. Normalized collision energy was 30 eV, and the underfill ratio, defining a minimum percentage of target value, was similar to the maximum fill time, set at 0.1%. Raw MS/MS spectra data were searched 5
Page 5 of 23
using a MASCOT engine (version 2.2; Matrix Science, London, UK) embedded into a Proteome Discoverer 1.4 (Thermo Fisher Scientific Inc. 2012). The acquired MS/MS spectra were automatically searched against the Uniprot Homo sapiens database (downloaded from the Uniprot database on April 1, 2017, and including 173,050 protein sequences). Search settings were: enzyme=trypsin; max. missed cleavage=2; variable
ip t
modifications: oxidation (M), iTRAQ 4plex (Y); fixed modifications: carbamidomethyl (C), iTRAQ 4plex (K), iTRAQ 4plex (N-term); peptide mass tolerance=±20 ppm; fragment mass
cr
tolerance=0.1 Da; and peptide false discovery rate (FDR) <0.01, protein FDR<0.01.
Protein expression ratios were calculated as the median number of distinct peptides of the
us
protein after normalizing peptide intensity by the median value of the intensity of all the proteins using the Proteome Discoverer 1.4 software. To screen significant protein
an
expression changes, a fold-change of >1.2 or <0.83 and P-value of <0.05 on Student t test were set as cutoff values.
M
Bioinformatics analysis
To determine the functional classification and biological properties of the selected
ed
differentially abundant proteins, the identified protein sequences were mapped onto GO Terms and annotated using BLAST2GO software (version 3.3.5). The annotation
ce pt
configuration comprised: E-value filter 1e-6, default gradual EC weights, GO weight 5, and annotation cutoff 75. Non-annotated sequences were then re-annotated with more permissive parameters. Sequences without BLAST hits and non-annotated sequences were run through an InterProScan10 program against the EBI databases to retrieve
Ac
functional annotations of protein motifs, then the InterProScan GO terms were merged into the annotation set. In addition, the FASTA protein sequences of differentially changed proteins were blasted against the online Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://geneontology.org/) to retrieve their Kos, and were subsequently mapped onto the KEGG pathways. To further explore the impact of the differentially expressed proteins, enrichment analysis was performed. GO enrichment on the three ontologies (biological process, molecular function, and cellular component) and KEGG pathway enrichment analyses were applied, based on Fisher exact test, considering the annotations of the whole quantified protein as the background dataset. Only functional 6
Page 6 of 23
categories and pathways with P-values <0.05 were considered significantly enriched. The protein-protein interaction information was retrieved from Cytoscape5 software (http://www.cytoscape.org/; version 3.2.1) based on the STRING database (http://stringdb.org) (Supplementary Tables S1 and S2). Immunohistochemistry (IHC)
ip t
To verify the reliability of the iTRAQ results and the general expression of the differentially expressed proteins in other PTMC specimens, IHC was performed on the other 16 patient
cr
PTMC tissues (8 PTMC patients with 8 PTMC patients without LNM). The inclusion criteria
us
for PTMC detected by IHC were similar to those described above. The differentially expressed proteins were examined using the streptavidin-peroxidase (SP) method according to the manufacturer’s instructions. Antigens were successively incubated with
an
primary antibodies, biotin-labeled secondary antibodies, and the HRP-labeled streptavidin. PBS rather than primary antibodies was used as negative control. Brown-yellow staining of
M
the nucleus was considered as positive expression in the cells. Images were analyzed using the Image-Pro Plus 6.0 system (Media Cybernetics Inc., USA). IOD (integrated
ed
optical density), area and mean density (IOD (sum)/area (sum)) were measured.
ce pt
Results
Identification results and overall distribution of proteins Fig. 1 shows sample preparation and analysis in each group. Twelve samples of normal thyroid tissue, benign thyroid tumor tissue, and PTMC tissues with and without LNM were
Ac
identified by iTRAQ. Finally, 49,037 peptide segments and 8,036 proteins (Supplementary Tables S3 and S4) were identified. Quantitative analysis of proteins Proteins were screened based on >1.2-fold (up- or down-regulation) difference in expression, and those with P-value <0.05 were considered differentially expressed. The number of differentially expressed proteins in the BTT and NH, PTC-LNM(-) and NH, PTCLNM(+) and NH, PTC-LNM(-) and BTT, PTC-LNM(+) and BTT, and PTC-LNM(+) and PTCLNM(-) groups are shown in Table 1. Comparison between PTC-LNM(-) and PTC-LNM(+) 7
Page 7 of 23
included 67 upregulated and 82 downregulated proteins. Identified proteins were further screened according to factors such as differentially expression, differential expression multiples and protein function analysis in the various groups. Finally, 26 differential proteins were obtained that might be related to PTMC metastasis, of which 13 were upregulated and 13 downregulated. The 13 upregulated proteins were: alpha-1-antitrypsin,
ip t
tenascin C (hexabrachion), microtubule-actin cross-linking factor 1, nuclear respiratory factor 1, 14-3-3 protein sigma, annexin A1, thrombospondin-1, high-mobility group protein
cr
HMGI-C (HMGA2), hepatoma-derived growth factor (HDGF), immunoglobulin heavy
variable 5-10-1, Ras-related protein Ral-B, pro-cathepsin H, and thrombospondin-2. The
us
13 downregulated proteins were: sushi-domain-containing protein 2, nostrin, ubiquitin carboxyl-terminal hydrolase 7, proto-oncogene tyrosine-protein kinase Src, vascular
an
endothelial growth-factor receptor 2 (VEGFR2), IQ motif-containing GTPase-activating protein 2 (IQGAP2), alpha-actinin 1 (ACTN1), RNA-binding protein 4B, carbonic
M
anhydrase 4 (CA4), thyroid peroxidase, pyruvate dehydrogenase E1, cellular retinoic-acidbinding protein 1, and myosin-reactive immunoglobulin light-chain variable region (Table
Bioinformatics analysis
ce pt
Cluster analysis
ed
2).
Clustering revealed high similarity within groups and low similarity between groups. Cluster analyses of four groups of samples demonstrated that the differentially expressed proteins identified experimentally were rational and accurate; the X and Y axes represented the
Ac
samples and the differentially expressed proteins, respectively (Supplemental Fig. 1). Functional annotation and analysis by GO The differentially expressed proteins in each group were annotated based on three aspects: the biological processes involved, the molecular functions, and the cell components. Significant enrichment analysis of GO annotations was evaluated by Fisher’s exact test (P<0.01). The differential proteins associated with PTMC metastasis involved 23 biological processes, 11 molecular functions, and 16 cell components (Fig. 2). In addition, the biological processes following enrichment mainly comprised chemical homeostasis 8
Page 8 of 23
within a tissue, zonula adherens maintenance, surfactant homeostasis, single-multicellular organism process, and multicellular organism process, while molecular functions comprised transforming growth-factor beta binding, and cell components mainly comprised zonula adherens (Supplemental Fig. 2). KEGG pathway analysis
ip t
All types of regulatory protein exerted a role in the network by interacting, co-operating, and promoting each other. Thus, a complex regulatory network was formed that executed
cr
the occurrence, development and progression of PTC. In addition, the KEGG pathway
us
systematically and comprehensively analyzed the target protein set to determine the affected signal transduction pathway. The KEGG pathway analysis was conducted on the differential proteins listed above, indicating that 13 pathways were associated with PTMC
an
metastasis: 11 upregulated pathways comprising focal adhesion, Rap1 signaling, ECMreceptor interaction, microRNAs in cancer, VEGF signaling, adherens junction, EGFR
M
tyrosine kinase inhibitor resistance, PI3K-Akt signaling, proteoglycans in cancer, actin cytoskeleton regulation, and Ras signaling, and 2 downregulated pathways comprising
ed
FoxO signaling and HIF-1 signaling (Supplemental Fig. 3). The above-mentioned KEGG pathways were enriched and analyzed based on P-values. The focal adhesion pathway
ce pt
exhibited a minimal P-value, and 6 differentially expressed proteins were identified: tenascin C, thrombospondin-1, thrombospondin-2, proto-oncogene tyrosine-protein kinase Src, VEGFR2, and ACTN1.
Ac
IHC verification
To check the reliability of the iTRAQ results and the general expression of the differentially expressed proteins in other PTMC specimens, 4 differential proteins were selected for IHC assessment. Results further confirmed that the expression of HDGF and HMGA2 was significantly higher in the LNM group than in the non-LNM group (P<0.01), while expression of IQGAP2 and CA4 was significantly lower in the LNM group than in the nonLNM group (P<0.01) (Table 3) (Fig. 3). These IHC results were in agreement with those of iTRAQ and were confirmed in other PTMC samples, which in turn substantiated the reliability of the iTRAQ results and the ubiquitous differential expression of the proteins. 9
Page 9 of 23
Discussion Currently, PTMC is defined as a tumor with size ≤1cm; however, invasiveness and metastatic capacity are not clearly understood. The necessity and scope of surgery for PTMC remain controversial. Some investigators advocate “over-diagnosis” and “overtreatment” for PTMC [22-27]. Thus, a small number of PTMCs might occur in a subclinical
ip t
state of non-progression. However, local invasion, LNM and distant metastasis might also occur in the early stage of PTMC [17,22,23,28]. In addition, any advanced PTC has
cr
progressed from PTMC, rather than having late onset. These phenomena suggest that PTMC exhibits different biological behaviors, despite pathological and anatomical factors
us
being identical. These different biological behaviors can provide a basis for individualized treatment of PTMC, which is crucial for determining the reasonable scope of surgery and
an
reducing “overtreatment.” We therefore conducted proteomic analyses using PTMC tissue without capsule invasion from single carcinoma lesions with tumor diameter 0.5–1 cm.
M
This excluded the above anatomical factors, enabling understanding of the molecular biological changes involved in PTMC, in order to provide novel information on the
ed
mechanism underlying PTC invasion and metastasis. Proteomic studies compared normal thyroid tissue, benign thyroid tumor tissue, PTC
ce pt
tissue with LNM, and PTC tissue without LNM, and some of the screened proteins were verified. Results demonstrated that the differentially expressed proteins identified by comparing the NH and BTT groups with the PTC-LNM(-) group independently, might be associated with cancerization, while those identified by comparing the NH and BTT groups
Ac
with the PTC-LNM(+) group independently might be associated with cancerization and metastasis, and those identified by comparing the PTC-LNM(-) and PTC-LNM(+) groups might be associated with metastasis. The identified proteins were further screened according to the differentially expressed proteins, differential expression multiplex, protein function analysis and background factors of thyroid carcinoma, with some screened proteins obtained from the other PTMC specimens serving as candidates for verification. Finally, the differential proteins and signaling pathways that might be closely associated with PTMC metastasis were found to be ACTN1, alpha-1-antitrypsin, HMGA2, HDGF, 10
Page 10 of 23
carbonic anhydrase 4, immunoglobulin heavy variable 5-10-1, myosin-reactive immunoglobulin light-chain variable region, and focal adhesion pathway. ACTN1 plays a critical role in cell adhesion, cell-shape regulation, and cell movement [29]. Moreover, tumor onset and development are dependent on the alpha-actinin involved in the cytoskeleton structure. Hirooka et al. found that the expression level of ACTN1 mRNA
ip t
in lung adenocarcinoma tissue was significantly lower than in normal lung tissue [30]. Glück et al. confirmed that increased ACTN1 expression in 3T3 cells reduced cell
cr
migration, while inhibition of expression by antisense transfection enhanced cell movement [31]. These phenomena prompted the downregulation of ACTN1 that is involved in
us
skeleton assembly and cell adhesion activity in transformed cell lines and tumor tissues. The current study found that ACTN1 expression in PTMC tissues with LNM was
an
downregulated and not only localized at the critical nodes of the PPI network but also closely related to the focal adhesion pathway. Thus, we speculated that ACTN1 may be a
M
significant regulator of PTMC metastasis.
Alpha-1-antitrypsin (alpha 1-AT) plays a major role in the stability of local tissues and
ed
tumorigenesis [32]. Alpha 1-AT content is significantly elevated in several tumor tissues and in serum from liver, pancreatic or gastric cancer [33-35]. Tzonou et al. confirmed that
ce pt
elevated serum α1-AT levels are associated with short-term survival in liver cancer patients [36]. Poblete et al. evaluated α1-AT expression in PTC tissue [37]. The present study demonstrated expression of α1-AT in PTMC tissue, and that upregulation promoted
Ac
PTMC LNM.
HMGI-C, also known as HMGA2, is associated with a variety of diseases, such as developmental disorders, malignant tumor, benign stromal tumor, obesity, diabetes, atherosclerosis and arthritis [38]. Kumar et al. developed a series of HMGA2 expression structures and compared the expression types of Let-7 and HMGA2; interaction between HMGA2 mRNA and let-7 was identified as the key step in carcinogenesis [39]. In addition, Jun et al. confirmed that HMGA2 was associated with onset of gastric cancer and promoted tumor progression, and might thus be used as a predictor of tumor recurrence [40]. Similarly, overexpression of HMGA2 was associated with growth and invasion in 11
Page 11 of 23
breast cancer [41]. In addition, microRNA Let-7b inhibited PTC by downregulating expression of HMGA2 [42]. In the present study, HMGA2 expression was found to be upregulated in PTMC tissue with LNM. We therefore speculated that HMGA2 could be a predictor of invasion, metastasis and prognosis in PTC. HDGF is related to the onset and development of several types of tumor: liver cancer, lung
ip t
cancer, cholangiocarcinoma, gastric cancer and prostate cancer; increased expression in malignant tumors is also associated with prognosis [43-45]. Targeted inhibition of HDGF
cr
effectively reduced hepatocellular carcinoma growth in in vitro and in vivo experiments [46]. Yang et al. speculated that the role of HDGF in hepatocellular carcinoma might be
us
related to the ERK signaling pathway [47]. A previous study using NIH/3T3 cells demonstrated that HDGF promotes cytoskeleton remodeling and migration by activating
an
the PI3K/Akt signaling pathway [48]; however, correlation studies between HDGF and thyroid carcinoma are still lacking. Our data showed that HDGF was expressed in PTMC
M
tissues, and that expression was significantly higher in PTMC with than without LNM. The difference in the expression multiplex of HDGF was the greatest of all differentially
ed
expressed proteins, indicating that HDGF plays a critical role in PTC invasion and metastasis. However, the KEGG pathway analysis could not identify the HDGF-associated
ce pt
signaling pathways or the specific upstream and downstream signaling molecules. Thus, HDGF might serve as an independent risk factor for PTC invasion and metastasis and could be a target for physiological treatment. The focal adhesion pathway directly or indirectly regulates cytoskeleton remodeling,
Ac
epithelial-mesenchymal transition and changes in a variety of related signaling pathways [49,50]. MicroRNA-218 inhibits renal carcinoma cell invasion and metastasis by regulating the focal adhesion pathway [51]. Furthermore, Talin1 promoted prostate carcinoma cell invasion and metastasis via the focal adhesion pathway in in vivo and in vitro experiments [52]. Also, in the present study, the activated signaling pathway associated with PTMC metastasis included focal adhesion, Rap1 signaling pathway, ECM-receptor interaction, microRNAs in cancer, VEGF signaling pathway, adherens junction, EGFR tyrosine kinase inhibitor resistance, PI3K-Akt signaling pathway, proteoglycans in cancer, regulation of actin cytoskeleton, and Ras signaling pathway, of which the enriched focal adhesion 12
Page 12 of 23
pathway might be a vital regulatory signaling pathway. Focal adhesion pathway activation not only promotes tumor cell invasion and metastasis but also regulates the PI3K-Akt signaling pathway, actin cytoskeleton, ECM-receptor interaction, VEGF signaling pathway, and Rap1 signaling pathway. Additionally, differentially expressed proteins such as tenascin C, thrombospondin-1, thrombospondin-2 and alpha-actinin-1 were closely
ip t
associated with the focal adhesion pathway. These results suggest that the focal adhesion pathway might be a critical upstream signaling pathway that regulates PTMC invasion and
cr
metastasis. In conclusion, these findings may provide deep insight into the mechanisms underlying PTMC invasion and metastasis. Further analysis of the present experimental
us
results concluded that upregulated expression of hepatoma-derived growth factor and high-mobility group protein HMGI-C and activation of the focal adhesion pathway promote
an
lymph node metastasis in PTMC. Data Availability Statement
M
The proteomics data used to support the findings of this study have been deposited in the
Disclosure Summary
ed
iProX (ww.iprox.org) repository (ID IPX0001247000 /PXD010302).
Funding
ce pt
The authors have no conflicts of interest to disclose.
The study was supported by The Inner Mongolia Autonomous Region “Prairie Outstanding
Ac
Ability Person” Project.
13
Page 13 of 23
References [1] Siegel RL, et al. Cancer Statistics, 2017. CA Cancer J Clin.2017; 67:7-30. [2] Jin S, et al. Signaling Pathways in Thyroid Cancer. Vitam Horm. 2018; 106:501-515. [3] Tang J, et al. Clinical-pathological Characteristics and Prognostic Factors for Papillary Thyroid Microcarcinoma in the Elderly. J Cancer. 2018; 9:256-262.
ip t
[4] Davies L, Welch HG. Current thyroid cancer trends in the United States. JAMA Otolaryngol Head Neck Surg. 2014; 140:317-322.
cr
[5] Lee JH, Shin SW. Overdiagnosis and screening for thyroid cancer in Korea. Lancet. 2014; 22:1848.
us
[6] Mutalib NS, et al. MicroRNAs and lymph node metastasis in papillary thyroid cancers. Asian Pacific Journal of Cancer Prevention. 2016; 17:25-35.
an
[7] Sun W, et al. Risk Factors for central lymph node metastasis in CN0 papillary thyroid carcinoma: a systematic review and meta-analysis. PLoS One. 2015; 10:e0139021. [8] Wu AW, et al. What is the best treatment for papillary thyroid microcarcinoma? Laryngoscope. 2011; 121:1828-1829.
M
[9] Woo CG, et al. Clinicopathological Significance of Minimal Extrathyroid Extension in Solitary Papillary Thyroid Carcinomas. Ann Surg Oncol. 2015; 22:S728-733.
ed
[10] Zhang L, et al. The clinical prognosis of patients with cN0 papillary thyroid microcarcinoma by central neck dissection. World J Surg Oncol. 2015; 7:138.
ce pt
[11] Yu X, et al. Independent Risk Factors Predicting Central Lymph Node Metastasis in Papillary Thyroid Microcarcinoma. Horm Metab Res. 2017;49; :201-207. [12] Takano T. Natural history of thyroid cancer. Endocr J. 2017; 64:237-244. [13] Grogan RH, et al. A study of recurrence and death from papillary thyroid cancer with 27 years of median follow-up. Surgery. 2013; 154:1436-1447.
Ac
[14] Yang J, et al. Risk factors for level V lymph node metastases in solitary papillary thyroid carcinoma with clinically lateral lymph node metastases. Cancer Medicine. 2016; 5:2161-2168. [15] Chéreau N, et al. Recurrence of papillary thyroid carcinoma with lateral cervical node metastases: Predictive factors and operative management. Surgery. 2016; 159:755-762. [16] Siddiqui S, et al. Clinical and Pathologic Predictors of Lymph Node Metastasis and Recurrence in Papillary Thyroid Microcarcinoma. Thyroid. 2016; 26:807-815. [17] Xu Y, et al. Clinical predictors of lymph node metastasis and survival rate in papillary thyroid microcarcinoma: analysis of 3607 patients at a single institution. J Surg Res. 2018; 221:128-134. 14
Page 14 of 23
[18] Jin S, et al. Signaling pathways in thyroid cancer and their therapeutic implications. J Cli Med Res. 2016; 8:284-296. [19] Gao M, et al. 2016 Chinese expert consensus and guidelines for the diagnosis and treatment of papillary thyroid microcarcinoma. Cancer Biol Med. 2017; 14:203-211. [20] Wiśniewski JR, et al. Universal sample preparation method for proteome analysis. Nat Methods. 2009; 6:359-362.
ip t
[21] Chan KC, Issaq HJ. Fractionation of peptides by strong cation-exchange liquid chromatography. Methods Mol Biol. 2013; 1002:311-315.
cr
[22] Jeon MJ, et al. Features of papillary thyroid microcarcinoma associated with lateral cervical lymph node metastasis. Clin Endocrinol (Oxf). 2017; 86:845-851.
us
[23] Ito Y, et al. Patient age is significantly related to the progression of papillary microcarcinoma of the thyroid under observation. Thyroid. 2014; 24:27-34.
an
[24] Gschwandtner E, et al. Increase of papillary thyroid microcarcinoma and a plea for restrictive treatment: A retrospective study of 1,391 prospective documented patients. Surgery. 2016; 159:503-511.
M
[25] Ahn HS, et al. Korea's thyroid-cancer "epidemic"--screening and overdiagnosis. N Engl J Med. 2014; 371:1765-1767.
ed
[26] Kim TY, Shong YK. Active Surveillance of Papillary Thyroid Microcarcinoma: A MiniReview from Korea. Endocrinol Metab (Seoul). 2017; 32:399-406. [27] Weber T, et al. Surgery of papillary thyroid microcarcinoma. Der Chirurg.2018; 1:1-6.
ce pt
[28] Li M, et al. Risk factors for predicting central lymph node metastasis in papillary thyroid microcarcinoma (CN0): a study of 273 resections. Eur Rev Med Pharmacol Sci. 2017; 21:3801-3807.
Ac
[29] Hamill KJ, et al. Alpha actinin-1 regulates cell-matrix adhesion organization in keratinocytes: consequences for skin cell motility. J Invest Dermatol. 2015; 135:10431052. [30] Hirooka S, et al. Localization of the invadopodia-related proteins actinin-1 and cortactin to matrix-contact-side cytoplasm of cancer cells in surgically resected lung adenocarcinomas. Pathobiology. 2011; 78:10-23. [31] Glück U, Ben-Ze'ev A. Modulation of alpha-actinin levels affects cell motility and confers tumorigenicity on 3T3 cells. J Cell Sci. 1994; 107:1773-1782. [32] De Serres F, Blanco I. Role of alpha-1 antitrypsin in human health and disease. Journal of Internal Medicine.2014; 276:311-335. [33] Topic A, et al. Alpha-1-antitrypsin in pathogenesis of hepatocellular carcinoma. Hepat Mon. 2012; 12:e7042. 15
Page 15 of 23
[34] Wang Y, et al. Screening for serological biomarkers of pancreatic cancer by twodimensional electrophoresis and liquid chromatography-tandem mass spectrometry. Oncol Rep. 2011; 26:287-292. [35] Wu W, et al. S100A9, GIF and AAT as potential combinatorial biomarkers in gastric cancer diagnosis and prognosis. Proteomics Clin Appl. 2012; 6:152-162.
ip t
[36] Tzonou A, et al. Alpha 1-antitrypsin and survival in hepatocellular carcinoma. Br J Cancer. 1990; 61:72-73. [37] Poblete MT, et al. Alpha 1-antitrypsin expression in human thyroid papillary carcinoma. Am J Surg Pathol. 1996; 20:956-63.
cr
[38] Young AR, Narita M. Oncogenic HMGA2: short or small? Genes Dev. 2007; 21:10051009.
us
[39] Kumar MS, et al. HMGA2 functions as a competing endogenous RNA to promote lung cancer progression. Nature. 2014; 505:212-217.
an
[40] Jun KH, et al. HMGA1/HMGA2 protein expression and prognostic implications in gastric cancer. Int J Surg. 2015; 505:39-44.
M
[41] Zou Q, et al. RKIP suppresses the proliferation and metastasis of breast cancer cell lines through up-regulation of miR-185 targeting HMGA2. Arch Biochem Biophys. 2016; 610:25-32.
ed
[42] Li H, et al. Roles of microRNA let-7b in papillary thyroid carcinoma by regulating HMGA2. Tumour Biol. 2017; 39(10): 1010428317719274.
ce pt
[43] Enomoto H, et al. Hepatoma-Derived Growth Factor: Its Possible Involvement in the Progression of Hepatocellular Carcinoma. Int J Mol Sci. 2015; 16:14086-14097. [44] Bao CH, et al. Prognostic role of hepatoma-derived growth factor in solid tumors of Eastern Asia: a systematic review and meta- analysis. Asian Pac J Cancer Prev. 2015; 16:1803-1811.
Ac
[45] Shetty A, et al. Hepatoma-derived growth factor: A survival-related protein in prostate oncogenesis and a potential target for vitamin K2. Urol Oncol. 2016; 34:483.e1-483.e8. [46] Enomoto H, et al. Down regulation of HDGF Inhibits the Growth of Hepatocellular Carcinoma Cells In Vitro and In Vivo. Anticancer Res. 2015; 35:6475-6479. [47] Yang GY, et al. Hepatoma-derived growth factor promotes growth and metastasis of hepatocellular carcinoma cells. Cell Biochem Funct. 2016; 34:274-285. [48] Kung ML, et al. Hepatoma-derived growth factor stimulates podosome rosettes formation in NIH/3T3 cells through the activation of phosphatidylinositol 3-kinase/Akt pathway. Biochem Biophys Res Commun. 2012; 425:169-176.
16
Page 16 of 23
[49] Eke I, Cordes N. Focal adhesion signaling and therapy resistance in cancer. Semin Cancer Biol. 2015; 31:65-75. [50] Maziveyi M, Alahari SK. Cell matrix adhesions in cancer: The proteins that form the glue. Oncotarget. 2017; 8:48471-48487.
ip t
[51] Yamasaki T, et al. MicroRNA-218 inhibits cell migration and invasion in renal cell carcinoma through targeting caveolin-2 involved in focal adhesion pathway. J Urol. 2013; 190:1059-1068. [52] Sakamoto S, et al. Talin1 promotes tumor invasion and metastasis via focal adhesion signaling and anoikis resistance. Cancer Res. 2010; 70:1885-1895.
cr
Figure Legends
us
Figure 1. Schematic of the experimental procedure.
ce pt
Figure 2. GO annotation.
ed
M
an
Total protein was extracted from freshly frozen tissues from NH, BTT, PTC-LNM(-) and PTC-LNM(+) samples using the SDT lysis method and digested using the FASP method; The peptide mixture of each sample was labeled using iTRAQ reagent. The iTRAQlabeled peptide was further divided using SCX chromatography. Finally, LC-MS/MS was used for high-throughput screening of the sample. Data were analyzed by bioinformatics tools from different aspects. Top check the reliability of the results, IHC was performed for other patient tissues of PTMC. (NH: normal histology; BTT: benign thyroid tumor; PTCLNM(-): classic papillary thyroid microcarcinoma without LNM; PTC-LNM(+): classic papillary thyroid microcarcinoma with LNM; FASP: filter-aided sample preparation; SCX: strong cation exchange; IHC: immunohistochemistry).
The differentially expressed proteins in each group were annotated based on the three aspects: the biological processes involved, the molecular functions, and the cell components.
Ac
Figure 3. Validation of protein expression in PTMC. To check the reliability of the iTRAQ results and the general expression of the differentially expressed proteins in other PTMC specimens, 4 differential proteins were selected for IHC. Cells in which the cytoplasm was stained brown-yellow were considered to be HDGF, HMGA2-, and IQGAP2-positive cells. If the nucleus was stained brown, the cells were considered to be CA4-positive cells. A: HDGF expression (LNM group); B: HDGF expression (non-LNM group); C: HMGA2 expression (LNM group); D: HMGA2 expression (non-LNM group); E: CA4 expression (LNM group); F: CA4 expression (non-LNM group); G: IQGAP2 expression (LNM group); H: IQGAP2 expression (non-LNM group). (HDGF: hepatoma-derived growth factor; HMGA2: high-mobility group protein HMGI-C; CA4: carbonic anhydrase 4; IQGAP2: GTPase-activating protein 2). Table1.Overview of the number of differentially expressed proteins 17
Page 17 of 23
Up-
Down-
All-
BTT/NH
458
563
1021
PTC-LNM(-)/NH
323
346
669
PTC-LNM(+)/ NH
595
598
1193
PTC-LNM(-)/BTT
369
275
644
PTC-LNM(+)/BTT
499
383
882
PTC-LNM(+)/PTC-LNM(-)
67
82
149
cr
ip t
Comparisons
us
Comparisons. Comparison groups: Up-: Upregulated differentially expressed proteins; Down-: Downregulated differentially expressed proteins; All: All differentially expressed proteins. Table 2. Differentially expressed proteins associated with PTMC metastasis
Covera Peptide MW ge s (kDa)
Protein
M
Protein ID
an
Single
A0A024R 6I7 Alpha-1-antitrypsin
58.61 2
46.68
Fold t-test Calc. pI change P-value 5.59
3.92018 0.00492 3 1
240.8
4.92
1.54226 0.00922 6 7
Microtubule-actin crossE9PLY5 linking factor 1
159.8
5.66
1.22338 0.00998 6 5
ed
A0A024R 884 Tenascin C (hexabrachion) 30.35 2
ce pt
16.51 1 1
22.68
5.95
1.24943 0.01126 4 4
47.98 7
27.76
4.74
1.46705 0.01323 4 1
7.02
1.33810 0.01410 3 2
4.94
1.40608 0.02414 6 9
8.27
2.55509 0.02597 7 0
10.6
2.15147 0.02873 2 7
C9JP85 Nuclear respiratory factor 1 4.31
P04083 P07996
14-3-3 protein sigma
Ac
P31947
Annexin A1
Thrombospondin-1
A0A0J9Y Immunoglobulin heavy XX1 variable 5-10-1 P52926
67.63 23 23.25 22 26.5
High-mobility group protein HMGI-C 57.8
1 5
38.69 129.3 12.7 11.83
18
Page 18 of 23
A0A087X 0D5 Pro-cathepsin H P35442
26.7
4
16.72 6
Thrombospondin-2
Sushi-domain-containing Q9UGT4 protein 2
90.15
6.28
P35968
Vascular endothelial growth factor receptor 2 3.32
ed
9.51
2 2
ce pt
A0A024R IQ motif containing GTPase AJ8 activating protein 2 33.71 43
Ac
Carbonic anhydrase 4
62.44 2 18.66 1 25.96 8
0.81471 0.01449 7 5
9.36
0.79401 0.01785 6 5
8.15
0.73934 0.01903 5 4
128.2
5.55
0.78006 0.01540 5 7
59.8
7.42
0.75258 0.01900 7 3
151.4
5.85
0.79197 0.03597 6 3
180.5
5.58
0.81172 0.03805 2 7
5.41
0.80465 0.04123 6 5
6.74
0.67211 0.04114 9 1
7.83
0.59967 0.00485 1 3
11.62
an
1
M
Proto-oncogene tyrosineprotein kinase Src
P22748
1.43338 0.04403 5 2
13.02 7
P12931
Q9BQ04 RNA-binding protein 4B
8.1
1.42785 0.04389 1 4
4.83
6.12
Alpha-actinin-1
6.62
129.9
Ubiquitin carboxyl-terminal Q93009 hydrolase 7 32.76 3
P12814
36.25
4.62343 0.03871 3 1 1.33063 0.04320 8 5
15.78 14
Myosin-reactive Q9UL70 37.96 2 immunoglobulin light chain variable region A0A0G2J M79 Nostrin (Fragment)
23.39
4.67
ip t
Ras-related protein Ral-B
26.79
cr
P11234
Hepatoma-derived growth 79.17 1 factor (high-mobility group protein 1-like)
us
B2RDE8
44.81
103 40.12 35.01
A0M8X1 Thyroid peroxidase
40.41 1
102.9
6.76
0.48434 0.00838 1 1
Pyruvate dehydrogenase Q53GE3 E1
38.21 1
43.19
8.06
0.18339 0.02038 3 4
5.38
0.65274 5 0.01492
P29762
Cellular retinoic acidbinding protein 1
57.66 8
15.56
19
Page 19 of 23
Table 3. Mean density of expression of HDGF, HMGA2, CA4, and IQGAP2 proteins in PTMC Group (n=8)
PTMC without lymph node metastasis
0.280±0.019
0.051±0.002＊
HMGA2
0.234±0.014
0.069±0.003＊
CA4
0.049±0.003
0.147±0.019＊
IQGAP2
0.067±0.006
0.198±0.021＊
M
an
us
cr
HDGF
ip t
PTMC with lymph node metastasis
*P<0.01,
Ac
ce pt
ed
vs. PTMC with lymph node metastasis group; HDGF, hepatoma-derived growth factor; HMGA2, high-mobility group protein HMGI-C; CA4, carbonic anhydrase 4; IQGAP2, IQ motif-containing GTPase-activating protein 2
20
Page 20 of 23
ip t cr us an M ed ce pt Ac
Figr-1
21
Page 21 of 23
ip t cr us an M Ac
ce pt
ed
Figr-2
22
Page 22 of 23
ip t cr Ac
ce pt
ed
M
an
us
Figr-3
23
Page 23 of 23