This review discusses the current status of proteomics technology in endometrial cancer diagnosis, treatment and prognosis. The first part of this review focuses on recently identified biomarkers for endometrial cancer, their importance in clinical use as well as the proteomic methods used in their discovery. The second part highlights some of the emerging mass spectrometry based proteomic technologies that promise to contribute to a better understanding of endometrial cancer by comparing the abundance of hundreds or thousands of proteins simultaneously.
Tumorigenesis involves a complex interplay between genetically modified cancer cells and their adjacent normal tissue, the stroma. We used an established breast cancer mouse model to investigate this inter-relationship. Conditional activation of Rho-associated protein kinase (ROCK) in a model of mammary tumorigenesis enhances tumor growth and progression by educating the stroma and enhancing the production and remodeling of the extracellular matrix. We used peptide matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) to quantify the proteomic changes occurring within tumors and their stroma in their regular spatial context. Peptides were ranked according to their ability to discriminate between the two groups, using a receiver operating characteristic tool. Peptides were identified by liquid chromatography tandem mass spectrometry, and protein expression was validated by quantitative immunofluorescence using an independent set of tumor samples. We have identified and validated four key proteins upregulated in ROCK-activated mammary tumors relative to those expressing kinase-dead ROCK, namely, collagen I, α-SMA, Rab14, and tubulin-β4. Rab14 and tubulin-β4 are expressed within tumor cells, whereas collagen I is localized within the stroma. α-SMA is predominantly localized within the stroma but is also expressed at higher levels in the epithelia of ROCK-activated tumors. High expression of COL1A, the gene encoding the pro-α 1 chain of collagen, correlates with cancer progression in two human breast cancer genomic data sets, and high expression of COL1A and ACTA2 (the gene encoding α-SMA) are associated with a low survival probability (COLIA, p = 0.00013; ACTA2, p = 0.0076) in estrogen receptor-negative breast cancer patients. To investigate whether ROCK-activated tumor cells cause stromal cancer-associated fibroblasts (CAFs) to upregulate expression of collagen I and α-SMA, we treated CAFs with medium conditioned by primary mammary tumor cells in which ROCK had been activated. This led to abundant production of both proteins in CAFs, clearly highlighting the inter-relationship between tumor cells and CAFs and identifying CAFs as the potential source of high levels of collagen 1 and α-SMA and associated enhancement of tissue stiffness. Our research emphasizes the capacity of MALDI-MSI to quantitatively assess tumor-stroma inter-relationships and to identify potential prognostic factors for cancer progression in human patients, using sophisticated mouse cancer models.
It is well established that cell surface glycans play a vital role in biological processes and their altered form can lead to carcinogenesis. Mass spectrometry-based techniques have become prominent for analysing N-linked glycans, for example using matrix-assisted laser desorption/ionization mass spectrometry (MALDI MS). Additionally, MALDI MS can be used to spatially map N-linked glycans directly from cancer tissue using a technique termed MALDI MS imaging (MALDI MSI). This powerful technique combines mass spectrometry and histology to visualise the spatial distribution of N-linked glycans on a single tissue section. Here, we performed N-glycan MALDI MSI on six endometrial cancer (EC) formalin-fixed paraffin-embedded (FFPE) tissue sections and tissue microarrays (TMA) consisting of eight EC patients with lymph node metastasis (LNM) and twenty without LNM. By doing so, several putative N-linked glycan compositions were detected that could significantly distinguish normal from cancerous endometrium. Furthermore, a complex core-fucosylated N-linked glycan was detected that could discriminate a primary tumour with and without LNM. Structural identification of these putative N-linked glycans was performed using porous graphitized carbon liquid chromatography tandem mass spectrometry (PGC-LC-MS/MS). Overall, we observed higher abundance of oligomannose glycans in tumour compared to normal regions with AUC ranging from 0.85-0.99, and lower abundance of complex N-linked glycans with AUC ranges from 0.03-0.28. A comparison of N-linked glycans between primary tumours with and without LNM indicated a reduced abundance of a complex core-fucosylated N-glycan (Hex)2(HexNAc)2(Deoxyhexose)1+(Man)3(GlcNAc)2, in primary tumour with associated lymph node metastasis. In summary, N-linked glycan MALDI MSI can be used to differentiate cancerous endometrium from normal, and endometrial cancer with LNM from endometrial cancer without.
Although acetylation is regarded as a common protein modification, a detailed proteome-wide profile of this post-translational modification may reveal important biological insight regarding differential acetylation of individual proteins. Here we optimized a novel peptide IEF fractionation method for use prior to LC-MS/MS analysis to obtain a more in depth coverage of N-terminally acetylated proteins from complex samples. Application of the method to the analysis of the serous ovarian cancer cell line OVCAR-5 identified 344 N-terminally acetylated proteins, 12 of which are previously unreported. The protein peptidyl-prolyl cis-trans isomerase A (PPIA) was detected in both the N-terminally acetylated and unmodified forms and was further analyzed by data-independent acquisition in carboplatin-responsive parental OVCAR-5 cells and carboplatin-resistant OVCAR-5 cells. This revealed a higher ratio of unacetylated to acetylated N-terminal PPIA in the parental compared with the carboplatin-resistant OVCAR-5 cells and a 4.1-fold increase in PPIA abundance overall in the parental cells relative to carboplatin-resistant OVCAR-5 cells (P = 0.015). In summary, the novel IEF peptide fractionation method presented here is robust, reproducible, and can be applied to the profiling of N-terminally acetylated proteins. All mass spectrometry data is available as a ProteomeXchange repository (PXD003547).
Protein glycosylation, particularly N-linked glycosylation, is a complex posttranslational modification (PTM), which plays an important role in protein folding and conformation, regulating protein stability and activity, cell-cell interaction, and cell signaling pathways. This review focuses on analytical techniques, primarily MS-based techniques, to qualitatively and quantitatively assess N-glycosylation while successfully characterizing compositional, structural, and linkage features with high specificity and sensitivity. The analytical techniques explored in this review include LC-ESI-MS/MS and MALDI time-of-flight MS (MALDI-TOF-MS), which have been used to analyze clinical samples, such as serum, plasma, ascites, and tissue. Targeting the aberrant N-glycosylation patterns observed in MALDI-MS imaging (MSI) offers a platform to visualize N-glycans in tissue-specific regions. The studies on the intra-patient (i.e., a comparison of tissue-specific regions from the same patient) and inter-patient (i.e., a comparison of tissue-specific regions between different patients) variation of early- and late-stage ovarian cancer (OC) patients identify specific N-glycan differences that improve understanding of the tumor microenvironment and potentially improve therapeutic strategies for the clinic.
Metastasis is a crucial step of malignant progression and is the primary cause of death from endometrial cancer. However, clinicians presently face the challenge that conventional surgical-pathological variables, such as tumour size, depth of myometrial invasion, histological grade, lymphovascular space invasion or radiological imaging are unable to predict with accuracy if the primary tumour has metastasized. In the current retrospective study, we have used primary tumour samples of endometrial cancer patients diagnosed with (n = 16) and without (n = 27) lymph node metastasis to identify potential discriminators. Using peptide matrix assisted laser desorption/ionisation mass spectrometry imaging (MALDI-MSI), we have identified m/z values which can classify 88% of all tumours correctly. The top discriminative m/z values were identified using a combination of in situ sequencing and LC-MS/MS from digested tumour samples. Two of the proteins identified, plectin and α-Actin-2, were used for validation studies using LC-MS/MS data independent analysis (DIA) and immunohistochemistry. In summary, MALDI-MSI has the potential to identify discriminators of metastasis using primary tumour samples.
The prediction of lymph node metastasis using clinic-pathological data and molecular information from endometrial cancers lacks accuracy and is therefore currently not routinely used in patient management. Consequently, although only a small percentage of patients with endometrial cancers suffer from metastasis, the majority undergo radical surgery including removal of pelvic lymph nodes. Upon analysis of publically available data and published research, we compiled a list of 60 proteins having the potential to display differential abundance between primary endometrial cancers with versus those without lymph node metastasis. Using data dependent acquisition LC-ESI-MS/MS we were able to detect 23 of these proteins in endometrial cancers, and using data independent LC-ESI-MS/MS the differential abundance of five of those proteins was observed. The localization of the differentially expressed proteins, was visualized using peptide MALDI MSI in whole tissue sections as well as tissue microarrays of 43 patients. The proteins identified were further validated by immunohistochemistry. Our data indicate that annexin A2 protein level is upregulated, whereas annexin A1 and α actinin 4 expression are downregulated in tumours with lymph node metastasis compared to those without lymphatic spread. Moreover, our analysis confirmed the potential of these markers, to be included in a statistical model for prediction of lymph node metastasis. The predictive model using highly ranked m/z values identified by MALDI MSI showed significantly higher predictive accuracy than the model using immunohistochemistry data. In summary, using publicly available data and complementary proteomics approaches, we were able to improve the prediction model for lymph node metastasis in EC.
Matrix assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) can determine the spatial distribution of analytes such as protein distributions in a tissue section according to their mass-to-charge ratio. Here, we explored the clinical potential of machine learning (ML) applied to MALDI MSI data for cancer diagnostic classification using tissue microarrays (TMAs) on 302 colorectal (CRC) and 257 endometrial cancer (EC)) patients. ML based on deep neural networks discriminated colorectal tumour from normal tissue with an overall accuracy of 98% in balanced cross-validation (98.2% sensitivity and 98.6% specificity). Moreover, our machine learning approach predicted the presence of lymph node metastasis (LNM) for primary tumours of EC with an accuracy of 80% (90% sensitivity and 69% specificity). Our results demonstrate the capability of MALDI MSI for complementing classic histopathological examination for cancer diagnostic applications.
Colorectal cancer (CRC) is the second leading cause of cancer-related deaths worldwide. CRC liver metastases (CRLM) are often resistant to conventional treatments, with high rates of recurrence. Therefore, it is crucial to identify biomarkers for CRLM patients that predict cancer progression. This study utilised matrix-assisted laser desorption/ionisation mass spectrometry imaging (MALDI-MSI) in combination with liquid chromatography-tandem mass spectrometry (LC-MS/MS) to spatially map the CRLM tumour proteome. CRLM tissue microarrays (TMAs) of 84 patients were analysed using tryptic peptide MALDI-MSI to spatially monitor peptide abundances across CRLM tissues. Abundance of peptides was compared between tumour vs stroma, male vs female and across three groups of patients based on overall survival (0-3 years, 4-6 years, and 7+ years). Peptides were then characterised and matched using LC-MS/MS. A total of 471 potential peptides were identified by MALDI-MSI. Our results show that two unidentified m/z values (1589.876 and 1092.727) had significantly higher intensities in tumours compared to stroma. Ten m/z values were identified to have correlation with biological sex. Survival analysis identified three peptides (Histone H4, Haemoglobin subunit alpha, and Inosine-5'-monophosphate dehydrogenase 2) and two unidentified m/z values (1305.840 and 1661.060) that were significantly higher in patients with shorter survival (0-3 years relative to 4-6 years and 7+ years). This is the first study using MALDI-MSI, combined with LC-MS/MS, on a large cohort of CRLM patients to identify the spatial proteome in this malignancy. Further, we identify several protein candidates that may be suitable for drug targeting or for future prognostic biomarker development.