Thyroid nodules are one of the most common endocrine diseases, of which only 5-15% turn into thyroid cancer. Thyroid cancer (TC) is the most common endocrine malignancy, accounting for 1% of all malignancies in the body and 91.5% of head and neck malignancies [1]. Differentiated thyroid carcinoma (DTC) accounts for more than 90% of all thyroid cancers, including papillary thyroid carcinoma (PTC) and follicular thyroid carcinoma (FTC). Surveys have shown that there are approximately 122,800 new cases of thyroid cancer worldwide each year, with a particular increase in the incidence of PTC [2]. The prognosis of differentiated thyroid carcinoma is closely related to the stage, so early diagnosis is of great importance.
Many diagnostic techniques are currently used for the early clinical diagnosis of thyroid cancer: ultrasound, CT, MRI, PET/CT, fine needle aspiration biopsy (FNAB). Usually, ultrasound-guided fine needle aspiration biopsy is considered the most effective method for preoperative diagnosis of thyroid cancer, and FNAB has been reported to have 93% sensitivity and 75% specificity [3]. However, FNAB has limitations: it is difficult to obtain accurate lesion tissue when the nodule is less than 1 cm in diameter; the amount of cells or tissue is limited; and it is an invasive test. In postoperative pathological findings in patients with indeterminate or suspected malignant lesions, only 20%-25% are thyroid cancer, and another 75%-80% undergo unnecessary thyroid surgery [4]. How to diagnose DTC accurately and specifically by noninvasive means at an early stage has been a clinical hotspot and difficulty. In recent years, many studies have attempted to improve the early diagnosis of DTC by detecting specific protein expression alterations in the serum of DTC patients and searching for DTC tumor markers, and some of these findings have demonstrated promising applications.
1. Target-free serum proteomics study of differentiated thyroid cancer
Many studies have found aberrant gene expression in DTC, including BRAFV600E mutation, RET/PTC1 gene rearrangement, RAS mutation and TRK in papillary carcinoma [5]; NRASQ61R mutation, BRAFK601E mutation, PAX8/PPARG rearrangement, RAS mutation and RET/PTC gene rearrangement in follicular carcinoma etc [6]. These mutated genes not only cause abnormalities in the encoded proteins, but also can cause alterations in the activity of related signaling pathways downstream, directly or indirectly affecting the translation and post-translational processing modifications of pathway-related proteins. Finding and analyzing tumor protein differences has the potential to identify DTC-specific tumor markers. Currently, the common proteomics techniques are matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS), surface enhanced laser desorption/ionization time-of-flight mass spectrometry (SALDI-TOF-MS), and surface enhanced laser desorption/ionization time-of-flight mass spectrometry (SALDI-TOF-MS). mass spectrometry (SELDI-TOF-MS), and surface enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS).
Sofiadis A [7] (2010) and Becker S [8] (2012) applied SELDI-TOF-MS and two-dimensional gel electrophoresis to analyze the differences in protein expression between DTC tissue samples and control samples, respectively, as a model for DTC diagnosis. However, protein differences in tissues are not equivalent to protein differences in serum, while the content and nature of tissues obtained for early tumor diagnosis are variable and prone to false-negative results, so tissue proteomics is not applicable to early preoperative diagnosis. Serum samples are abundant, and serum extraction is convenient and fast in clinical practice. Serum proteins are abundant (60C80 mg/ml) and diverse (hundreds of species) [9]. Lu Xubo et al [10] (2006) then found that each of the three groups of thyroid cancer, benign thyroid nodules and healthy individuals had unique differences in serum protein expression, and that serum protein fingerprinting models with high specificity and sensitivity could be established using serum proteomics techniques. This suggests that proteomics can analyze the differences between tumor and benign lesion sera, so serum proteomics may be more advantageous in the early diagnosis of tumors.
Since the first report of using proteomics technology to screen for serum-related protein markers in ovarian cancer in 2002 [11], proteomics technology has been widely used in the study of many human tumors, such as lung cancer [12], pancreatic cancer [13], ovarian cancer [14], breast cancer [15], and colon cancer [16]. So far, domestic and foreign researchers have studied serum specimens from patients with differentiated thyroid cancer using proteomic approaches and screened some potential serum protein markers or established serum protein profiling models with high specificity and sensitivity for the diagnosis of DTC.
1.1 Matrix-assisted laser-resolved ionization time-of-flight mass spectrometry (MALDI-TOF-MS)
MALDI-TOF-MS is a traditional technique for protein separation and identification. It is based on the principle of using a nitrogen-source pulsed laser to cause the matrix to absorb the energy of the laser, thereby ionizing the peptide sample. The peptide sample then enters the mass analyzer, where ionization occurs due to the difference in mass-to-charge ratio, and the relevant parameters of the peptide ion can be measured: peptide mass fingerprint, peptide sequence label or partial amino acid sequence, and finally the corresponding software is used to search the proteomic database for qualitative identification or quantitative analysis of the protein.
Andrew Martorella et al [17] (2007), on the other hand, used MALDI-TOF-MS analysis to identify 98 significantly different serum peptides in 27 cases of thyroid cancer and 32 cases of serum to form a serum peptide fingerprint, and this model was able to identify the differences between the two groups of serum peptides with a high statistical accuracy. However, the experimentally discovered peptide model was not subjected to blind selection experiments, and the experimental results are yet to be confirmed and the experimental methods need to be improved. Compared with MALDI-TOF-MS, it can directly use raw unprocessed samples (e.g. serum, tissues, body fluids, etc.) for detection, and can be used for large-scale, ultra-micro, high-throughput and fully automated screening of proteins. It can also be used for diagnosis and research of related diseases by combining protein profiles in different ways, and is currently widely studied in basic and clinical tumor research.
1.2 Surface-enhanced laser-resolved ionization time-of-flight mass spectrometry (SELDI-TOF-MS)
SELDI-TOF-MS, also known as protein fingerprinting technique, consists of 3 parts: protein microarray, time-of-flight mass spectrometry and analysis software. According to the different mass-to-charge ratios of proteins with different flight times in the instrument field, the position of the measured protein group in the map depends on the flight time and is presented in the form of peaks, which are plotted by computer processing to form a mass spectrometry map, directly showing the molecular weight and content of various proteins in the sample. By comparing the spectra of the experimental group group with the control group, it is possible to identify and capture the disease-specific relevant proteins.
Yuxia Fan et al [1] (2009) analyzed and screened three differential serum protein peaks: haptoglobin a1 chain (9190 Da), apolipoprotein C-I (6631 Da), and apolipoprotein C-III (6631 Da) in patients with PTC and benign thyroid lesions using SELDI-TOF-MS. High expression of haptoglobin a1 chain and low expression of apolipoprotein C-I and apolipoprotein C-III were established as the diagnostic model of PTC, and its diagnostic sensitivity was 98%. The specificity was 97%. Wang et al [18] (2006) and William H et al [19] (2008) used the same method to establish a tumor diagnostic model consisting of differential protein peaks. Their PTC diagnostic sensitivity and specificity are above 80% or even reach 100%. SELDI-TOF-MS technique can accurately analyze and screen the proteins or peptides that differ between malignant tumors and benign tumors and normal serum, and can find the subtle differences between them; meanwhile, the serum is easy to collect and preserve with sufficient quantity. Its can be used for early diagnosis of tumors and finding tumor-specific markers. However, this method has no target protein, large measurement works and poor reproducibility of experimental results, which still needs to be further validated.
In summary, the differential proteins obtained in the above experiments are far apart, with different degrees of sensitivity and specificity, and such data cannot be used in clinical applications. The reasons for this can be: 1, the uneven level of measurement at different institutions, Alexander [20] (2009) sent 20 highly purified recombinant human proteins to 27 liquid chromatography-based proteomics research laboratories, and only 7 laboratories reported 20 proteins correctly.2, the protein or peptide composition and quantity in serum is in dynamic change. The reproducibility of the results of such experiments is poor. The method of detecting protein differences in serum without a target does not seem to work.
2. Study of abnormal protein glycosylation modifications in differentiated thyroid cancer
In fact, the proteomic study of tumor consists of two main aspects: (1) expression proteomics, which focuses on the screening of differential proteins from quantitative and qualitative charge ratios to establish tumor diagnostic models. Functional proteomics focuses on the study of tumorigenesis and development at the level of protein structure, function and mechanism of action. Therefore, protein differences in not only stem from differences in abundance expression, but also respond to post-synthesis modifications of proteins, including phosphorylation, glycosylation, and acetylation [21]. Among them, glycosylation modifications are one of the most important post-translational modifications of proteins and are the most studied in scientific research.
Protein glycosylation is the covalent joining of oligosaccharide fragments and side chains of amino acids together, followed by further folding of polypeptide chains with a certain amino acid arrangement to form protein molecules with a certain spatial structure, which is one of the important post-translational modifications of proteins, and 2/3 of the proteins in eukaryotic cells are glycosylation modified. There are two main types of protein glycosylation: one is N-glycosylation, in which oligosaccharide fragments are covalently attached to side chains containing asparagine (Asn), and N-glycosylation commonly occurs in proteins in the extracellular environment, including membrane proteins, secreted proteins, and proteins in body fluids; the other is O-glycosylation, in which oligosaccharide fragments are attached to side chains containing serine (Ser) or threonine (Thr) residues [ 22].
Protein glycosylation is involved in the regulation of many physiological and pathological events in cells, such as cell growth, migration, differentiation, and tumor metastasis. Cell membrane surface glycoproteins are located in the outermost layer of the cell, communicating between the internal and external environment, and are involved in many important biological processes such as receptor activation and signal transduction [23]. Two-thirds of the proteins in eukaryotic cells are glycosylation-modified, and many cell surface receptors belong to glycoproteins, such as EGFR, integrins and TGFBR are N-glycosylation-modified [21].
Arcinas [24] in 2009 comparatively analyzed the expression profiles of cell surface proteins and secreted proteins of five different human thyroid cancer cell lines and identified a total of 333 glycosylation-modified proteins. In a study of differentiated thyroid cancer cell lines (TPC-1, FTC-133), five proteins, cell membrane proteins, botulin, cell adhesion molecule-1, trophoblast cell glycoprotein and discoid integrin α-5 protein chain, were expressed only in DTC. Miyoshi et al [21] (2010) further confirmed that in the glycosylation-modified tumor cells of differentiated thyroid cancer glycosylation-modifying transferase FUT8 was significantly upregulated in the early stage and GnT-V was significantly upregulated in the later stage of expression.
From the above studies, it is evident that there is enhanced glycosylation modification of thyroid histone cell proteins. The current study also found that in patients with hepatocellular carcinoma (N-acetylglucosaminyltransferase V, GnT-V; N-acetylglucosaminyltransferase III, GnT-III; α1-6 inkhorn glycosyltransferase,α1-6FT) [25], breast cancer (N-acetylglucosaminyltransferase V , GnT-V) [26] and prostate cancer (B-binding bead protein) [27] Hidenori [28] detected the expression of Galectin-3 in the sera of patients with thyroid cancer. At present, there are few studies on serum protein glycosylation modifications in differentiated thyroid cancer for early diagnosis, but this approach still shows some prospects for development and provides novel research ideas, for example, several DTC differentially glycosylated proteins can be screened and combined with more specific thyroglobulin as a diagnostic model, so further research is needed to study abnormal protein glycosylation modifications in differentiated thyroid cancer.
Conclusion and outlook
In conclusion, the study of early diagnosis of differentiated thyroid cancer serologically is still in the experimental stage. Target-free serum proteomics studies provide a very clear research idea, but serum proteomics has not been applied to the early clinical diagnosis of differentiated thyroid cancer so far, for the following reasons:1, A careful analysis of the above serological test results shows that the differential proteins obtained from different research institutions vary greatly, and the reproducibility of experimental data is poor and cannot be clinically applied [20].2,Serum proteins 99% of serum proteins are composed of 22 high molecular weight proteins, such as: albumin, transferrin, and conjugated globulin; the remaining 1% is only composed of hundreds of low molecular weight proteins. The number of proteins in serum is diverse and often in dynamic change [9]. The current traditional approach of trying to find tumor markers through untargeted serum/plasma with low abundance of small molecule peptides does not seem to work.3, A single-minded study of protein quantity alterations. The differences between normal and tumor tissues due to genetic alterations are reacted to proteins, but the functional differences of proteins in not only stem from the differences in abundance expression, but more in post-synthesis modifications of proteins, including phosphorylation, glycosylation, acetylation, etc. The study of protein modification in differentiated thyroid cancer began to develop, especially the study of differentiated thyroid cancer-specific glycosylated proteins, as seen in the above study, there is enhanced glycosylation modification of thyroid tissue cell proteins, but the abnormal serum/plasma protein glycosylation modification in thyroid cancer is less studied and needs to be carried out further.