N-glycans in lung tissue specimens: a prospective target for enhanced cancer diagnosis and prognosis

Abstract

Background: It is well established that the cancerous transformation of cells is accompanied by profound alterations in glycosylation. In this study, we demonstrate the diagnostic potential of N-glycan profiling in tissue specimens from patients, primarily representing the two major types of lung cancer: non-small cell and small cell lung cancer.

Methods: Lung tissues and biopsies obtained from surgery and bronchoscopy underwent sample processing and enzymatic digestion. After labeling, glycans were analyzed employing matrix-assisted laser desorption/ionization mass spectrometry. Statistical analysis was conducted using methods following principles of compositional data analysis.

Results: Comparison of glycan profiles demonstrated an increase in paucimannose and high mannose glycans in most tumor specimens, including those with inflammation and histological negative for malignancy. Cancerous tissues exhibited more profound changes in glycosylation. Despite the high heterogeneity in profiles, two main groups of not detected glycans in peritumoral tissues, considered as controls, were observed to correlate with cancer progression in patients. One with complex polylactosamine multifucosylated glycans frequently harboring terminal N-acetyl-glucosamine residues. These glycans were present in most tumors, with their numbers and intensities increasing as cancer progressed. In contrast, the second group exhibited polylactosamine glycans sporadically. Instead, the biopsies of several patients with rapid progression displayed a significant presence in a set of tri- and tetra-antennary core fucosylated glycans having mostly unoccupied N-acetyl-glucosamine residues unless carrying additional fucose unit(s).

Conclusions: The results imply distinct glycosylation patterns even in patients with the same histological type of lung cancer, supporting trends toward personalized diagnosis and more tailored therapies. Currently, tissue biopsies remain the gold standard for diagnosing premalignant and malignant lesions in the lung. Expanded knowledge on glycosylation in these lesions could contribute to improved diagnostic accuracy and better monitoring of malignant disease progression in clinical practice.

Keywords: N-glycans; Branched glycans; Fucosylation; Glycosylation; Lung cancer; MALDI-MS.

Fig. 1

N-Glycan profiles of lung tissue from surgery versus biopsy from bronchoscopy. MALDI-MS spectra were recorded from lung metastatic tissues. A m/z 1000–3000 and B m/z 2650–4150 from patient P26 (66yrs, Supplementary Tab. 1); and C m/z 2650–4150 from necrotic biopsy of patient Met-F1 (43yrs, Supplementary Tab. 2). The m/z values in red designate significant increase (p < 0.001) in glycan peaks or represent glycans that were not detected in control tissues. Key symbols: Fuc (red triangle), Gal (yellow circle), GlcNAc (blue square), Man (green circle)

Fig. 2

Statistical comparison of detected glycan in tissue biopsies of LC patients. A Graphical illustration of the number glycans observed in individual samples (represented by dots) in different clinical groups. Error bars represent mean ± SD. B Relative abundances of selected N-glycans (H-Hexose, N-HexNAc, F-Fucose) detected in biopsies of patients with LAC, SqLC and SCLC: (a) H2N2F1 (m/z 1007.3); (b) H6N2 (m/z 1509.5); (c) H3N4F1 (m/z 1575.6); (d) H5N4F2 (m/z 2045.8); (e) H3N6F2 (m/z 2127.8); (f) H3N6F3 (m/z 2273.8). (g) H6N5F3 (m/z 2556.9); (h) H6N6F3 (m/z 2760.0); (i) H7N6F3 (m/z 2922.1); (j) H7N7F2 (m/z 2979.1); (k) H8N7F2 (m/z 3141.2) and (l) H9N8F1 (m/z 3360.2). Box limits denote the 1st and 3rd quartile, central line denotes the median, whiskers extend 1.5 × interquartile range from the limits of the box. Scatterplot points represent individual samples. Statistical comparisons were performed using ANOVA with post-hoc TukeyHSD tests following alr-transformation (*p < 0.05, **p < 0.01, ***p < 0.001)

Fig. 3

Diverse glycan profiles related to progressing cancer. MALDI-MS spectra of N-glycans recorded from the biopsies of patients with SqCC: A 70yrs old man Sq-M29 with glycans detected at range of m/z 1000–3010, B m/z 2690–4110; and C 71yrs old woman Sq-F7 with m/z 3000–5000. In D 62yrs old female patient LAC-F5 with m/z 2690–3750. Peaks with m/z values in red colour suggest malignancy. Blue arrows illustrate an increasing mass 203 u (patients’ subgroup R1). For more peak assignments, see Supplementary Tab. 3

Fig. 4

Glycans associated with poor patient survival. MALDI-MS spectra of N-glycans with mass range 1920–2400 recorded from the biopsies of SqCC patients. A 80yrs old man (Sq-M21, precancerous stage). B 50yrs old woman (Sq-F11; pTx/Gx). C 50yrs old man (Sq-M4, pTx/Gx). Increased intensities in peaks with m/z in red indicate malignancy. Peaks marked with a red cross ( +) were mostly absent in control tissues, and the increase in their intensities consistently correlated with rapid LC progression (patients’ subgroup R2). D Relative abundances of selected N-glycans based on disease progression. Box limits denote the 1st and 3rd quartile, central line denotes the median, whiskers extend 1.5 × interquartile range from the limits of the box. Scatterplot points represent individual samples. Statistical comparisons were performed using ANOVA with post-hoc TukeyHSD tests following alr-transformation (*p < 0.05, **p < 0.01, ***p < 0.001)

Fig. 5

Graphical illustration of this study. A Combination of the histological cross section and detection of differences in N-glycans by MALDI-MS for diagnostics and prognostics of LC. B Schematic summary of N-glycans detected in patients lung tissue specimens. In the blue column are N-glycans commonly detected in control lung tissues (arrow indicates regular increase or decrease of glycan when detected in cancerous tissue). In the yellow columns are N-glycans detected with an increase in patients diagnosed with inflammation and LC: paucimannose (PM) and high-mannose (HM). In the pink column is a group of the complex fully galactosylated multifucosylated polylactosamine glycans (CmF) which increased in LC patients with advanced LC. The violet column displays a group of complex polylactosamine multifucosylated glycans carrying terminal GlcNAc residue/s (CmFG) frequently associated with progressing cancer and metastasis. The last column (dark violet) displays a set of tri- and tetra-antennary complex core fucosylated glycans carrying predominantly terminal GlcNAc units (t-CFG), their significant increase consistently correlated with the highest cancer progression rate. The glycans in two last columns could serve potentially as one of the most reliable markers for early stages of malignancy and monitoring therapy, their significant increase correlated with cancer progression and poor treatment outcome. *Below the columns with glycans are hypothesized increased (↑) or decreased (↓) expressions of some glycoenzymes deduced from detected glycan structures: FUT—fucosyltransferase; GNT—N-acetylglucosaminyltransferase; HEX—N-acetylhexosaminidase (isoenzymes associated with GNT-I dependant truncation pathways); MAN—α-mannosidase; GALT—galactosyltransferase. **At the bottom are examples of patients with LC progression whose tissues specimens showed the highest abundances of glycans listed in the individual groups. More characteristics of glycans detected in samples are in Supplementary Table 3