Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Jul 2;96(28):11163–11171. doi: 10.1021/acs.analchem.3c05984

Highly Sensitive Spatial Glycomics at Near-Cellular Resolution by On-Slide Derivatization and Mass Spectrometry Imaging

Cécile Cumin †,‡, Lindsay Gee †, Thomas Litfin †, Ropafadzo Muchabaiwa †, Gael Martin †, Oren Cooper †, Viola Heinzelmann-Schwarz ‡,§, Tobias Lange ∥,⊥, Mark von Itzstein †,*, Francis Jacob ‡,*, Arun Everest-Dass †,*
PMCID: PMC11256013  PMID: 38953530

Abstract

graphic file with name ac3c05984_0007.jpg

Glycans on proteins and lipids play important roles in maturation and cellular interactions, contributing to a variety of biological processes. Aberrant glycosylation has been associated with various human diseases including cancer; however, elucidating the distribution and heterogeneity of glycans in complex tissue samples remains a major challenge. Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) is routinely used to analyze the spatial distribution of a variety of molecules including N-glycans directly from tissue surfaces. Sialic acids are nine carbon acidic sugars that often exist as the terminal sugars of glycans and are inherently difficult to analyze using MALDI-MSI due to their instability prone to in- and postsource decay. Here, we report on a rapid and robust method for stabilizing sialic acid on N-glycans in FFPE tissue sections. The established method derivatizes and identifies the spatial distribution of α2,3- and α2,6-linked sialic acids through complete methylamidation using methylamine and PyAOP ((7-azabenzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate). Our in situ approach increases the glycans detected and enhances the coverage of sialylated species. Using this streamlined, sensitive, and robust workflow, we rapidly characterize and spatially localize N-glycans in human tumor tissue sections. Additionally, we demonstrate this method’s applicability in imaging mammalian cell suspensions directly on slides, achieving cellular resolution with minimal sample processing and cell numbers. This workflow reveals the cellular locations of distinct N-glycan species, shedding light on the biological and clinical significance of these biomolecules in human diseases.

Introduction

The phenomenon of cellular heterogeneity, which is frequently observed in biological systems, remains largely elusive. Arising from a variety of genetic, epigenetic, and environmental influences, this heterogeneity manifests in diverse cell morphologies, functions, and pathologies. This underscores the importance of analyzing individual cells and their surroundings to comprehend their biochemical and physiological attributes. Traditional methods typically rely on aggregate data from cell populations, which overlooks the distinct behaviors and variations inherent to individual cells, such as differences in metabolic processes, growth, and proliferation. Glycoproteins and glycolipids are abundant biomolecules on the surface of all cells that form a layer called the glycocalyx.1 Glycomics can directly provide the glycan coding information on single cells that are related to phenotypes without the need to extrapolate from bulk measurements. It can also provide insight into functional processes within single cells. Low abundant and unique glycan coding information can be analyzed by single-cell glycan imaging techniques.2−4 As a result, the glycan study of individual cells can make it possible to reveal the relationship between physiological behavior of cells and their chemical composition and to understand the biochemical basis of cell variation and cell communication.

Sialic acids are a family of sugar units composed of a nine-carbon backbone found mostly attached to the outermost end of these glycoconjugates.5 Sialic acids include N-acetylneuraminic acid (Neu5Ac), N-glycolylneuraminic acid (Neu5Gc), deaminoneuraminic acid (Kdn), and their derivatives with modifications, such as methylation, acetylation, and sulfation at the 4, 7, 8, and 9 positions, generating more than 50 unique species.6 Owning a negative charge on the carboxylic acid at the physiological pH, sialic acids can mediate a wide variety of physiological and pathological processes making them promising biomarkers.7−9 It has been observed for several decades that increased syntheses and expression of sialic acids are associated with cancer development and metastasis.10−12 Significant alterations in sialylation are observed in multiple cancers such as ovarian cancer,13,14 leukemia,15,16 gastric,16 colorectal,17 and breast cancer.18 Changes in sialic acid levels in the tumor microenvironment are primarily attributed to the metabolic flux and aberrant expression of sialyltransferases/sialidases, as tumor cells have an increased uptake of glucose, the raw material for sialic acid synthesis. The aberrant sialylation further facilitates immune escape, enhances tumor proliferation and metastasis, aids tumor angiogenesis, and assists in resisting apoptosis and cancer therapy.7,19 Therefore, characterizing and targeting sialic acids in cancer are attractive diagnostic and therapeutic options.

Development of soft ionization matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) provides a unique advantage to spatially visualize a wide range of analytes from lipids,20 metabolites,21 peptides,22 proteins,23 and more recently glycans23−27 with high sensitivity on a single platform. This technological advance has provided the opportunity to gain unique insights into spatially resolved expression of N-glycan structures from fresh-frozen as well as formalin-fixed paraffin-embedded (FFPE) tissue sections.28 This has led to the characterization of N-glycan MALDI-MSI in several studies, such as diabetes,29 colorectal,30 pancreatic,31 and ovarian cancer.25 Nonetheless, the analysis of the glycome in tissue samples using MALDI-MSI reveals some challenging aspects. Most MALDI-MSI analysis are performed in positive-ion mode, the utilization of the negative-ion mode may serve as an alternative method for the mass spectrometric analysis of sialylated glycans. The negative charge of the sialic acid can lead to a crosswise ionization as well as loss of sialic acids, inducing in-source and postsource decay.32 Moreover, the presence of the carboxylic acid group leads to multiple alkali metal adducts of sialylated glycans during analysis.33

Thus, it is essential to establish a protocol to stabilize α2,3- and α2,6-linked sialic acids for downstream MALDI analysis at cellular resolution. Different derivatization methods have already been established in glycomics34−36 and glycoproteomics37 to convert the sialic acid carboxylate group into ester or amide groups. Initially, the use of DMT-MM together with MeOH was reported to convert α2,3- and α2,6-linked sialic acids into lactones and methyl-esters, respectively, at a temperature of 80 °C. These methods were applicable to release and purify glycans. There have been some studies that have shown the derivatization of sialic acid on-tissue, Wuhrer and colleagues reported on an elegant esterification protocol using EDC, HOBt, dimethylamine, and ammonium hydroxide in DMSO, that requires an incubation time of 3 h to derivitise.24 Similarly, Zhang et al., have taken the approach to neutralize the sialic acid charge after coupling an aromatic group with aniline.38 Lu et al., recently developed a multifunctional isomer-targeted sialic acid derivatization strategy, which also requires long incubation time.39

In this study, we implemented a fast and robust protocol based on the work of Liu et al.,41 aimed at stabilizing sialic acids to facilitate the imaging of N-glycans at cellular resolution from formalin-fixed paraffin-embedded (FFPE) tissues and mammalian cell suspensions. The MALDI imaging workflow used on-slide derivatization using methylamine and PyAOP ((7-azabenzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate) as condensing agents. We tested several incubation times and the effect of temperature and concentration of the derivatizing reagents for reaction completeness and selectivity for sialic acid linkages. We further optimized the subsequent washing steps prior to antigen retrieval and show improved detection of glycans from distinct tissue regions, demonstrating complex N-glycan heterogeneity that is otherwise lost. The entire workflow from tissue dewaxing to MALDI-MSI analysis can be completed in less than 8 h, a crucial factor in the implementation of glycan imaging of FFPE tissues in emerging digital pathology.

Experimental Section

On-Tissue Derivatization and N-Glycan Release

To improve the adherence of the tissue on ITO glass slides, tissue sections were placed on a heating plate at 65 °C for 60 min. The entire protocol was performed under a hood except during the desiccator steps. Paraffin was removed by incubating the slides in xylene (2 × 4 min), and tissue sections were rehydrated by different washes: 100% ethanol (1 × 2 min), 90% ethanol (1 × 2 min), 70% ethanol (1 × 2 min), 50% ethanol (1 × 2 min), water (1 × 2 min), and completely dried in a desiccator. Tissue derivatization (Figure 1A) was performed by incubating the tissue slides in derivatization solution: (1 M methylamide hydrochloride, 0.5 M methylmorphine, and 50 mM PyAOP in DMSO) for 3 × 5 min at ambient temperature. Tissue sections were encircled by using a Hydrophobic (PAP) Barrier Pen. Using a pipet, the tissue was covered by derivatization solution (200–300 μL) until completely submerged, which was renewed after 5 min with fresh solution (200–300 μL) for the following incubation step. The incubation was protected from evaporation by enclosing the slides in a slide chamber. After derivatization, tissue sections were washed in 100% ethanol (2 × 2 min), Carnoy solution (60% ethanol, 30% chloroform, 10% acetic acid, 1 × 5 min), and water (1 × 1 min). Slides were dried in a vacuum desiccator (∼10 min) before antigen retrieval. Citraconic anhydride buffer (∼10 mL, pH = 3)-based antigen retrieval was carried out in a commercial vegetable steamer (Philips All-In-One Cooker, HD2237) for 20 min. Following the antigen retrieval, slides were cooled down for 5 min, rinsed in water, and dried in the vacuum desiccator before enzymatic deglycosylation. PNGaseF (0.1 μg/μL in 25 mM ammonium bicarbonate) was deposited using an HTX TM-Sprayer (HTX Technologies, USA), 15 layer passes at 25 μL/min, velocity of 1200 mm/min, crisscross pattern, 3.0 mm track spacing, and nitrogen gas pressure was set to 10 psi. N-Glycans were released in a 3 h incubation at 37 °C in a humidity chamber protected from evaporation. After incubation, the ITO slides were briefly dried in the vacuum desiccator, and the matrix (10 mg/mL CHCA, 70% ACN and 0.1% TFA) was applied using the HTX TM-Sprayer (5 layers, 0.1 mL/min, 80 °C, velocity of 1300 mm/min, crisscross pattern, 2.5 mm track spacing, nitrogen gas pressure was set to 10 psi).

Figure 1.

Figure 1

Open in a new tab

(A) Schematic representation demonstrating the workflow to stabilize sialic acids on-tissue. (B) Reaction scheme of in situ α2–3 linked and α2–6 linked sialic acid derivatization as previously reported by Liu et al.41 One-step-derivatization using PyAOP and methylamine as reactants for stable amidation of sialylated glycans.

MALDI-TOF/TOF Mass Spectrometry of N-Glycans

MALDI MSI data were acquired using a rapifleX MALDI TOF/TOF mass spectrometer (Bruker, Daltonics) equipped with a Smartbeam 3D 10 kHz laser and operated in a positive-ion reflectron. The laser power was optimized at the start of each run and then held constant during the MALDI–MSI experiment. At each sampling position, 500 shots were used to acquire data at 50 μm pixel resolution for the m/z 920–3500 range. All of the methods used were internally calibrated using commonly present neutral glycans across the mass range. Spectra were normalized against the root-mean-square (RMS) of all data points, unless otherwise stated. Mass spectral analyses and chemical formula calculations were performed in DataAnalysis (Bruker Daltonics, v.4.2). Relative and absolute quantitative analyses were performed using msIQuant48 (v.2.0.1.15).

The MALDI MSI parameters for single-cell glycan imaging were as follows. Laser power was optimized on cells identified outside the region of analysis and then held constant during the MALDI–MSI experiment. At the sampling position, 250 shots were used to acquire data at 10 μm pixel resolution for the m/z 920–3500 range. Similarly, internal calibration was applied prior to method acquisition using commonly present neutral glycans across the mass range.

Other detailed materials and methodology, including mammalian cell culture and suspension on slides protocol, glycan extraction from FFPE tissues and PGC glycomics, tissue preparation, and sectioning is available in the Supporting Information section.

Results and Discussion

Establishment of N-Glycan Methylamidation for Imaging

Derivatization methods have used NH4Cl,40 MeNH2,41 or 2(-2-pyridylamino)ethylamine associated with different condensing agents such as 4-(4,6-dimethoxy-1,3,5-trazin-2-yl)-4-methyl-morpholinium chloride (DMT-MM)42 or (7-azabenzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate (PyAOP)43 previously. However, there are reports on sialic acid isomer linkages that can react differently over time, and some of these methods offer a complete derivatization of α2,6-linked sialic acid but a poor yield for α2,3-linked sialic acid, which forms lactones under acidic conditions. To overcome this problem, a second reaction with a different derivative agent, such as ammonium hydroxide or methylamine, is used to stabilize the α2,3-linked sialic acids. However, these methods require long incubation times and high temperatures of up to 80 °C, which could potentially alter tissue integrity and adversely affect lateral analyte diffusion.

The methylamidation derivatization step (Figure 1) originally established by Liu et al.,41 for sialylated glycans in-solution uses PyAOP, a carboxylic acid activator which is often utilized during peptide synthesis, and has been reported to significantly improves the yield in the amidation of both α2,6 and α2,3-linked sialic acids compared to DMT-MM.44 To methylamidate sialic acid residues, PyAOP was used as a condensing agent with the reactant MeNH2 (Figure 1B). DMSO was used as a solvent with N-methylmorphine to stabilize basic pH (∼8) and avoid the formation of lactones. This induces an increased mass shift of 13 Da corresponding to methylamidated sialic acid. The accuracy of this one step reaction in methylamidating both of these variably linked sialic acids was evaluated in vitro using standards of 3′- and 6′-sialyllactose. Tandem mass spectra analysis (Figure S1) and NMR spectroscopy analysis (Figures S2 and S3) of these standards showed the specific methylamidation of the sialic acid irrespective of their linkages. We further also validated representative derivatized siaylated glycans from tissue sections using PGC-LC-ESI-MS/MS analysis as shown in Figure S4.

Visualization of N-Glycans In Situ after and before Methylamidation

To test the derivatization efficiency and completeness of methylamidation, we monitored the mass shifts of representative sialylated glycans from imaged tissue sections with and without on-tissue derivatization (Figure 2). For example, the monosialylated biantennary N-glycan (Hex5HexNac4NeuAc1) was detected at m/z 1976.6 with an additional neutral proton-sodium exchange. While the derivatized monosialylated biantennary N-glycan detected as m/z 1967.7. Similarly, fucosylated monosialylated biantennary with methylamidated sialic acid is detected at m/z 2113.7 (Hex5HexNac4NeuAc1dHex1) with similar distribution to its native counterpart (m/z 2122.7). While the disialylated biantennary N-glycans m/z 2271.7 (Hex5HexNAc4NeuAc2) and m/z 2417.8 (Hex5HexNac4NeuAc2dHex1) were observed with distinct localization and good signal intensity when derivatized and poorly detected in their native forms.

Figure 2.

Figure 2

Open in a new tab

MALDI-MSI of N-glycans from consecutive FFPE sections of a chemo-naive primary tissue sample derived from a high-grade serous ovarian cancer patient, native (top) and derivatized (bottom) tissues; major peaks are annotated with glycan compositions. Monoisotopic glycan masses were measured in the positive ion reflectron mode as (M+Na) adducts at 50 μm spatial resolution (46 μm raster). Monosaccharide symbols are as defined by Symbol Nomenclature for Glycans (SNFG). Blue square = N-acetylglucosamine; yellow square = N-acetylgalactosamine; red triangle = fucose, purple diamond = N-acetylneuraminic acid; green circle = mannose, yellow circle = galactose.

We also evaluated the effect of the derivatization reactants on the localization of detected neutral glycans and whether there were any observable lateral diffusion effects that can occur and therefore impact spatial resolution significantly. The high mannose N-glycan m/z 1419.5 (Hex6HexNAc2) and complex N-glycan m/z 1809.6 (Hex5HexNAc4dHex1) showed a similar spatial ion intensity pattern in the native and derivatized N-glycan tissues (Figure 2). The slight differences in the ion intensity observed between the two sections could be attributed to changes in cell density or tissue morphology across consecutive tissue sections.

Different solvents including methanol, ethanol, and DMSO were assessed for derivatization efficiency and their impact on tissue integrity. Methanol or ethanol did not allow a complete derivatization of the released N-glycans due to the presence of both m/z 2113.7 (derivatized) and m/z 2122.7 (native) monosialylated species. Under the same conditions, DMSO offered a more complete derivatization of the sialic acids and a negligible delocalization of the analytes. Thus, we decided to continue with DMSO as the solvent for further experiments.

We further optimized the derivatization conditions by slightly altering the incubation time, repetitive derivatization steps, and the concentration of reactants. The efficiency of derivatization was monitored by measuring the abundance of the derivatized sialylated N-glycans and that of their native counterpart (Figure S5). We initially incubated for 1 h at room temperature using the following concentration: 1 M methylamide hydrochloride, 0.5 M methylmorphine, and 50 mM PyAOP in DMSO. However, we observed the presence of the underivatized sialylated N-glycans (Figure S5). Doubling the concentration of each chemical reactant does not improve the reaction within the same amount of time (Figure S5), but a noticeable decrease in the MS intensity of the sialylated glycans. Possibly due to the incomplete removal of the excess derivatization reagent having a detrimental effect on ionization.

In a parallel setup, the solvent was washed away from the tissue after 30 min of incubation, and a freshly made reaction was added to the tissue for another 30 min. N-glycans capped with either α2,6 or α2,3-sialic acids were entirely stabilized with this procedure (Figure S5). Then, to optimize our workflow, we tested different incubation times of the reaction, two times 10 min (Figure S5) and three times 5 min. Comparing between the glycan ion intensities of the derivatized sialylated N-glycans (m/z 1967.7, m/z 2113.7, m/z 2332.7) and their nonderivatized counterparts (m/z 1976.6, m/z 2122.7, m/z 2341.9) revealed optimal derivatization were three times repeated 5 min incubations. Complete derivatization of both neuraminic acid linkage is assumed due to the absence or negligible abundance of the nonderivatized sialylated N-glycan (Figure S5).

Following the methylamidation steps, tissue sections were washed using Carnoy solution (60% ethanol, 30% chloroform, 10% acetic acid) and were found to be effective in removing residual derivatizing reagents.

Holst et al. demonstrated that derivatization by dimethylamidation and subsequent amidation of FFPE tissues did not require antigen retrieval.24 Thus, we examined the effect of the derivatization reagent on epitope unmasking. In our analysis, we found that antigen retrieval was essential to unmask the cross-linking and provide access to the deglycosidase to uniformly release N-glycans across tissue sections. This observed difference from the Holst et al. method could be due to the extended incubation time at high temperature (3 h at 60 °C) that unmasks FFPE cross-linking.

Comparison of Native and Derivatized Glycan Imaging Spectra

In order to evaluate the effect of the methylamidation on global glycans detected from FFPE tissue sections, we compared the N-glycan profile obtained by MALDI-MSI of native and derivatized glycans after complete derivatization of α2,3- and α2,6-linked sialic acids in positive-ion mode (Figure 3). Consecutive serous ovarian cancer from omentum FFPE tissue sections (labeled OC1) were imaged at 50 μm spatial resolution.

Figure 3.

Figure 3

Open in a new tab

Comparison of representative single-pixel mass spectrum of (A) native N-glycans and (B) derivatized N-glycans in the mass range of m/z 920 to m/z 3350 from the consecutive serous ovarian cancer FFPE tissue section. Monoisotopic glycan masses were measured in the positive ion reflectron mode as (M+Na) adducts at 50 μm spatial resolution (46 μm raster). (C) Number of glycans observed in relation to native versus derivatization in the OC1 serous ovarian cancer FFPE tissue section. (D) Number of glycans observed in relation to native versus derivatization in the OC2 serous ovarian cancer FFPE tissue section. An overview of all different N-glycans detected in both derivatized and nonderivatized tissue sections is reported in Table S1.

The masses m/z 1663.5 (Hex5HexNAc4) and m/z 1809.6 (Hex5HexNAc4dHex1) were the highest intensity peaks detected in the native spectra corresponding to non sialylated glycans (Figure 3A) while the sialylated glycans were either not detected or barely detected as monosialylated species in the native tissue section. In contrast, some of the derivatized sialylated N-glycans (m/z 1967.7, m/z 2113.7, m/z 2271.8, and m/z 2478.8) are among the relatively abundant glycans detected after methylamidation (Figure 3B). This indicates that the peak intensity of the neutral N-glycans were biased due to the loss of sialylated species. Di- and trisialylated N-glycans comprising other biologically relevant glycan motifs such as Lewis fucosylation, bisecting, and branched structures are observed which are otherwise lost without derivatization (Figure 3). The spatial localization of all the detected glycan species from tissue OC1 are shown in Figure S6.

We observed a significant increase in the number of glycans detected, especially in the higher mass range of the MALDI-MSI spectra after derivatization. The increased detection of higher glycan masses is due to the effect of an increase in sialylated glycans, while the number of neutral glycans detected marginally changed. In total, 68 N-glycan species were detected after in situ derivatization compared to 48 glycans without methylamidation for this particular FFPE tissue section (Table S1, Figure 3C). Only 4 sialylated structures were detected in the nonderivatized tissue section compared to 24 sialylated structures detected in the derivatized tissue section (Figure 3C).

Similarly, we analyzed consecutive sections from another ovarian cancer patient’s FFPE tissue (labeled OC2). The analysis of the derivatized and native glycans from these consecutive sections revealed a similar pattern. 66 N-glycan species were detected after in situ derivatization compared to 47 glycans without methylamidation (Figure 3D and Table S1). Here as well, only four sialylated structures were detected in the nonderivatized tissue section compared to 23 sialylated structures detected in the derivatized tissue section (Figures 3D and S7). The spatial localization of all the detected glycan species from tissue OC2 are shown in Figure S7. An overview of all different N-glycans detected in both derivatized and nonderivatized tissue sections is reported in Table S1. On average, we observed an increase in ∼30% of glycan structures detected, comprising of about 35% sialylated glycans.

We also tested our workflow on FFPE sections of human colon cancer xenograft tumors grown in SCID mice to test whether the methylamidation protocol enables visualization of Neu5Gc-containing glycans. As shown in Figure S8, the successful derivatization of the monosialylated Neu5Gc structure resulted in m/z 1983.7 while the monosialylated Neu5Ac structure has an expected 16 Da lower mass at m/z 1967.7.

Derivatization Workflow Enables Glycan Imaging at Near-Cellular Resolution in Tissues and Mammalian Cell Suspensions

Mass spectrometry-based single-cell analysis is gaining popularity to quantitate and identify proteins and metabolites. MALDI mass spectrometry imaging has some inherent limitations that restrict spatial localizations such as laser focal diameter, size of matrix crystals, and analyte diffusion during sample preparation.45 Recent technological advancements have now enabled the acquisition of biomolecular information from single eukaryotic cells using commercial instruments if adequate analytes are attainable for sensitive detection. The large diversity and complexity of glycans originating from a single cell represent a significant challenge. Innovative imaging techniques to simultaneously measure and map the localization of glycans from individual cells in clinical tissue samples is a major requirement. In Figure 4, we show the application of our workflow to detect and image N-glycans at a near-single cell resolution (20 μm spatial resolution; 16 μm raster) from a serous ovarian cancer FFPE section. The overlaid H&E-stained image when compared to individual N-glycan spatial intensity maps show discrete morphological patterns and tumor heterogeneity. The oligomannose glycan (m/z 1257.4) is predominantly present in the tumor, while the highly branched complex N-glycan (m/z 2742.8) is also observed in the cancer epithelium but with unique localization. Interestingly, the cancer-associated stroma has both neutral (m/z 1809.6) and sialylated (m/z 2417.8) structures with core fucosylation, while their nonfucosylated counterparts (m/z 1663.5 and m/z 2271.7) are observed at a greater distance away from the cancer epithelium. We also employed techniques for reducing dimensionality and conducted unsupervised clustering based on glycan intensities as a robust method to simplify and make sense of the MALDI-MSI data (Figure 5). A remarkable finding from this approach was that a projection of spatial-agnostic cluster labels on to the tissue area aligned remarkably well with the distinct regions observed in hematoxylin and eosin (H&E) stained samples. Moreover, these clusters revealed unique segmentation that highlighted the intricate tumor microenvironment and the broader tumor landscape, as shown in Figure S9. Although Clusters 1 and 10 represent stroma, as observed from the H&E-stained image in Figure S9C,D, the morphology of these regions is distinctly different and corelates well with the segmentation. Similarly, Clusters 0 and 2 (Figure S9B,E) demonstrate the heterogeneity of tumor regions with unique histomorphological features distinguished by N-glycan segmentation. This is especially significant as it illustrates the potential of using spatial glycomics to gain valuable biological insights. However, one must exercise caution in interpreting these clusters, as we believe their unambiguous identification can only be achieved through alternative, orthogonal approaches. This includes the use of immunohistochemistry to detect specific markers. The various spatial ion intensity maps of diverse N-glycan subtypes observed in Figure 4 illustrate the complex interplay of glycan synthesis and regulation, resulting in unique cellular phenotypes.

Figure 4.

Figure 4

Open in a new tab

On-tissue methylamidation enables near-single cell glycan MALDI-MSI in FFPE tissue sections. Hematoxylin and eosin (H&E) staining of serous ovarian cancer tissue sections (A). Glycan structure-specific panel of images shows the magnified H&E-stained region (black box, B) with the corresponding spatial localization of N-glycans. Note the distinct distribution of individual N-glycan structures in or near the cancerous epithelium (red; C (m/z 1257.4), K (m/z 2742.8)) compared to those in the cancer-associated stroma in close proximity (blue arrows; F (m/z 1809.6), G (m/z 2012.6), H (m/z 2174.7), J (m/z 2417.8)) or at a greater distance from the cancer epithelium (D (m/z 1485.5), E (m/z 1663.5), I (m/z 2271.7)).

Figure 5.

Figure 5

Open in a new tab

Projection of cluster labels derived from glycan intensities. (A) Mapping of cluster labels based on pixel coordinates of glycan MALDI-MSI at 20 μm spatial resolution (16 μm raster). (B) Mapping of cluster labels to low dimensional UMAP of MALDI-MSI spots with cluster numbers corresponding to spatial regions.

Currently, the sequencing or characterization of the glycome of cell culture suspensions is performed by bulk analysis of the same heterogeneous cell-type or of complex mixture of various unique cell types, such as in the glycomic analysis of whole blood. These limitations of bulk analysis can be tremendously improved by performing single-cell analysis. Now, the most developed, scalable, and convenient technique for molecular profiling of cell type diversity is single-cell RNA sequencing (scRNA-seq). When applied at an adequate scale, the technology has the potential to classify cell types46,47 and has shown to trace cell lineages and to find physiologically relevant differences within cells that were previously considered homogeneous. Although this is a remarkable advancement, transcriptomic analyses are insufficient to infer how post-translational modifications such as glycosylation and their subsequent biological functions are regulated as these are influenced by numerous parameters that are beyond transcriptional control.48,49 In contrast to nucleic acids and proteins, the biosynthesis of glycans is not directly template driven but, rather, is a result of a complex network of metabolic and enzymatic reactions that are influenced by many factors, including the genetic profile of the cells in which the glycoconjugates are expressed, epigenetics, and the extracellular environment.50 In consequence, capturing post-transcriptional mechanisms can reliably just be achieved by direct measurements of the target molecules from single cells. With respect to glycosylation, such single-cell measurements have heavily relied on lectin or antibody-based methods,4,51 but to date there are just a handful of lectin/antibodies available that just cover a very limited space of the glycome.4 These limitations can be overcome by a sophisticated combination of leveraging on recent advancements in mass-spectrometry (MS) technology and sample handling/preparation protocols to achieve high throughput single-cell glycomics, including spatial resolution. Angel et al. demonstrated a protocol that adapts tissue N-glycan MSI workflows to cells grown on glass slides in an array format. They further extended this technique through isotopic labeling of N-glycans in cell cultures, thereby enabling the measurement of N-glycan turnover rates following the induction of oxidative stress in human primary aortic endothelial cells.52,53

Our MALDI MSI workflow allows for true molecular glycomics at the single-cell level, as shown in Figure 6 and Table S2. Using a cytocentrifuge, single-cell suspensions are spun onto an ITO slide allowing the cells to adhere to the slide as a monolayer. The glycan ion intensity (Figure 6C) of the triantennary sialylated biantennary glycan (m/z 2478.8) detected from individual mammalian cells can be visually correlated to its corresponding H&E-stained image (Figure 6A). We further explored the heterogeneity in the SH-SY5Y cultured cell population by comparing the complex-type N-glycan profiles of these cells. A total of 75 unique N-glycan compositions were detected in the single-cell MALDI MSI workflow employing derivatization, of which 34 were sialylated N-glycans (Table S2). As expected, we found differing levels of N-glycan expression across cell populations, possibly due to asymmetric cell divisions that lead to different cell fate in a homogeneous microenvironment.54,55 Importantly this demonstrates our workflow’s applicability to answer several open questions in glycomics, such as the associated glycan states to metabolic reprogramming, cell signaling and differentiation, distinct states of cancer cells, and host cell response to viral infections. This methodology enables high throughput, automated glycomics-based diagnostic and prognostic information from single cells.

Figure 6.

Figure 6

Open in a new tab

Single-cell glycan MALDI-MSI of SH-SY5Y mammalian cell suspensions. Hematoxylin and eosin (H and E) staining of the cell suspension (A). Magnified H and E-stained region with the corresponding spatial localization and ion intensity of (B) biantennary core fucosylated N-glycan of m/z 1809.6, (C) oligomannose Man6 glycan of m/z 1419.5, and (D) triantennary sialylated N-glycan of m/z 2478.8. Monoisotopic glycan masses were measured in the positive ion reflectron mode as (M+Na) adducts at 10 μm spatial resolution (6 μm raster).

Conclusions

MALDI imaging is a promising and emerging technology to study the spatial distribution of glycans in human tissue samples. In this study, we demonstrate a rapid and robust method to image N-glycans from FFPE tissues using MALDI-MSI. Altogether, the successful implementation of our workflow facilitates single-cell glycan imaging at a high spatial resolution. The one-step derivatization method, characterized by straightforward preparation steps and a brief processing time, enables the visualization of both neutral and various sialylated species with reliable spatial localization. This method ensures a uniform charge across both neutral and sialylated glycans, thus enhancing efficient ionization in the positive ion mode. A limitation, however, is the inability to differentiate between 2,3- and 2,6-linked sialic acid isomers within a single analysis.

Finally, this workflow will bring new biological insights through the characterization of glycosylation changes associated with human disease.

Acknowledgments

Research was sponsored by the Australian National Health and Medical Research Council ID2009677 and ID1196520 (to M.V.I.), the Wilhelm Sander Stiftung 2018.010.1 and Swiss Cancer League KLS-3841-02-2016 (to F.J.), EACR travel fellowship #687 (to C.C.). The authors acknowledge the International Centre for Cancer Glycomics, established with funding from the Australian Cancer Research Foundation (ACRF) at the Institute for Glycomics for a supportive environment.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.3c05984.

  • Additional experimental procedures including methods and materials, experimental details, data analysis, tandem MS spectra of sialyl-N-acetyllactose standards, NMR spectrum of methyl amidation of 3- and 6-sialyl-N-acetyllactose standards, MALDI MSI of N-linked glycans from OC1 and OC2 tissue sections, MALDI MSI of N-linked glycans from patient-derived xenograft and histomorphology cluster mapping of the H&E-stained tissue section; overview of N-glycans detected in both serous ovarian cancer tissue sections (OC1, OC2), native and derivatized, with detected mass, glycan composition, and proposed structure (Table S1); overview of N-glycans detected in SH-SY5Y single-cell suspension analysis under native and derivatized conditions, with detected mass, glycan composition, and proposed structure (Table S2) (PDF)

Author Contributions

C.C., L.G., R.M., and A.E.-D. developed and optimized the methods. C.C., L.G., T.F., R.M., and A.E.-D. performed the on-sample analysis. G.M. performed cell culture experiments. O.C. performed NMR spectroscopy experiments. C.C. and A.E.-D. performed PGC-LC–MS/MS validation. C.C., M.V.I., T.M., F.J., T.L., and A.E.-D. conceived and performed data analyses. C.C., T.L., M.V.I., T.M., F.J., and A.E.-D. interpreted data. C.C., M.V.I., F.J., and A.E.-D. wrote the paper. A.E.-D., M.V.I., and F.J. supervised and coordinated the work.

The authors declare no competing financial interest.

Supplementary Material

ac3c05984_si_001.pdf (7.3MB, pdf)

References

  1. Paszek M. J.; DuFort C. C.; Rossier O.; Bainer R.; Mouw J. K.; Godula K.; Hudak J. E.; Lakins J. N.; Wijekoon A. C.; Cassereau L.; Rubashkin M. G.; Magbanua M. J.; Thorn K. S.; Davidson M. W.; Rugo H. S.; Park J. W.; Hammer D. A.; Giannone G.; Bertozzi C. R.; Weaver V. M. The cancer glycocalyx mechanically primes integrin-mediated growth and survival. Nature 2014, 511 (7509), 319–325. 10.1038/nature13535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Odaka H.; Ozaki H.; Tateno H. scGR-seq: Integrated analysis of glycan and RNA in single cells. STAR Protoc 2022, 3 (1), 101179. 10.1016/j.xpro.2022.101179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ma T.; McGregor M.; Giron L.; Xie G.; George A. F.; Abdel-Mohsen M.; Roan N. R. Single-cell glycomics analysis by CyTOF-Lec reveals glycan features defining cells differentially susceptible to HIV. Elife 2022, 11, e78870 10.7554/eLife.78870. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Kearney C. J.; Vervoort S. J.; Ramsbottom K. M.; Todorovski I.; Lelliott E. J.; Zethoven M.; Pijpers L.; Martin B. P.; Semple T.; Martelotto L.; Trapani J. A.; Parish I. A.; Scott N. E.; Oliaro J.; Johnstone R. W. SUGAR-seq enables simultaneous detection of glycans, epitopes, and the transcriptome in single cells. Sci. Adv. 2021, 7 (8), eabe3610 10.1126/sciadv.abe3610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Kang H.; Wu Q.; Sun A.; Liu X.; Fan Y.; Deng X. Cancer Cell Glycocalyx and Its Significance in Cancer Progression. Int. J. Mol. Sci. 2018, 19 (9), 2484. 10.3390/ijms19092484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Angata T.; Varki A. Chemical diversity in the sialic acids and related alpha-keto acids: an evolutionary perspective. Chem. Rev 2002, 102 (2), 439–469. 10.1021/cr000407m. [DOI] [PubMed] [Google Scholar]
  7. Pearce O. M.; Laubli H. Sialic acids in cancer biology and immunity. Glycobiology 2016, 26 (2), 111–128. 10.1093/glycob/cwv097. [DOI] [PubMed] [Google Scholar]
  8. Pinho S. S.; Reis C. A. Glycosylation in cancer: mechanisms and clinical implications. Nat. Rev. Cancer 2015, 15 (9), 540–555. 10.1038/nrc3982. [DOI] [PubMed] [Google Scholar]
  9. Munkley J.; Elliott D. J. Hallmarks of glycosylation in cancer. Oncotarget 2016, 7 (23), 35478–35489. 10.18632/oncotarget.8155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. van Beek W. P.; Smets L. A.; Emmelot P. Increased sialic acid density in surface glycoprotein of transformed and malignant cells--a general phenomenon?. Cancer Res. 1973, 33 (11), 2913–2922. [PubMed] [Google Scholar]
  11. Dobie C.; Skropeta D. Insights into the role of sialylation in cancer progression and metastasis. Br. J. Cancer 2021, 124 (1), 76–90. 10.1038/s41416-020-01126-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Murugesan G.; Correia V. G.; Palma A. S.; Chai W.; Li C.; Feizi T.; Martin E.; Laux B.; Franz A.; Fuchs K.; Weigle B.; Crocker P. R. Siglec-15 recognition of sialoglycans on tumor cell lines can occur independently of sialyl Tn antigen expression. Glycobiology 2021, 31 (1), 44–54. 10.1093/glycob/cwaa048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Choi H. J.; Chung T. W.; Choi H. J.; Han J. H.; Choi J. H.; Kim C. H.; Ha K. T. Increased alpha2–6 sialylation of endometrial cells contributes to the development of endometriosis. Exp. Mol. Med. 2018, 50 (12), 1–12. 10.1038/s12276-018-0167-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Anugraham M.; Jacob F.; Everest-Dass A. V.; Schoetzau A.; Nixdorf S.; Hacker N. F.; Fink D.; Heinzelmann-Schwarz V.; Packer N. H. Tissue glycomics distinguish tumour sites in women with advanced serous adenocarcinoma. Mol. Oncol. 2017, 11 (11), 1595–1615. 10.1002/1878-0261.12134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ma H.; Zhou H.; Song X.; Shi S.; Zhang J.; Jia L. Modification of sialylation is associated with multidrug resistance in human acute myeloid leukemia. Oncogene 2015, 34 (6), 726–740. 10.1038/onc.2014.7. [DOI] [PubMed] [Google Scholar]
  16. Liu N.; Zhu M.; Linhai Y.; Song Y.; Gui X.; Tan G.; Li J.; Liu Y.; Deng Z.; Chen X.; Wang J.; Jia L.; He X.; Wang X.; Lin S. Increasing HER2 alpha2,6 sialylation facilitates gastric cancer progression and resistance via the Akt and ERK pathways. Oncol. Rep. 2018, 40 (5), 2997–3005. 10.3892/or.2018.6680. [DOI] [PubMed] [Google Scholar]
  17. Sethi M. K.; Fanayan S. Mass Spectrometry-Based N-Glycomics of Colorectal Cancer. Int. J. Mol. Sci. 2015, 16 (12), 29278–29304. 10.3390/ijms161226165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Ma X.; Dong W.; Su Z.; Zhao L.; Miao Y.; Li N.; Zhou H.; Jia L. Functional roles of sialylation in breast cancer progression through miR-26a/26b targeting ST8SIA4. Cell Death Dis. 2016, 7 (12), e2561 10.1038/cddis.2016.427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. van de Wall S.; Santegoets K. C. M.; van Houtum E. J. H.; Bull C.; Adema G. J. Sialoglycans and Siglecs Can Shape the Tumor Immune Microenvironment. Trends Immunol. 2020, 41 (4), 274–285. 10.1016/j.it.2020.02.001. [DOI] [PubMed] [Google Scholar]
  20. Liu Y.; Chen L.; Qin L.; Han M.; Li J.; Luo F.; Xue K.; Feng J.; Zhou Y.; Wang X. Enhanced in situ detection and imaging of lipids in biological tissues by using 2,3-dicyanohydroquinone as a novel matrix for positive-ion MALDI-MS imaging. Chem. Commun. 2019, 55 (83), 12559–12562. 10.1039/C9CC06961E. [DOI] [PubMed] [Google Scholar]
  21. Dilillo M.; Ait-Belkacem R.; Esteve C.; Pellegrini D.; Nicolardi S.; Costa M.; Vannini E.; Graaf E. L.; Caleo M.; McDonnell L. A. Ultra-High Mass Resolution MALDI Imaging Mass Spectrometry of Proteins and Metabolites in a Mouse Model of Glioblastoma. Sci. Rep. 2017, 7 (1), 603. 10.1038/s41598-017-00703-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Beine B.; Diehl H. C.; Meyer H. E.; Henkel C. Tissue MALDI Mass Spectrometry Imaging (MALDI MSI) of Peptides. Methods Mol. Biol. 2016, 1394, 129–150. 10.1007/978-1-4939-3341-9_10. [DOI] [PubMed] [Google Scholar]
  23. Angel P. M.; Baldwin H. S.; Su Y. R.; Mayer J. E.; Bichell D.; Drake R. R. Advances in MALDI imaging mass spectrometry of proteins in cardiac tissue, including the heart valve. Biochim. Biophys. Acta, Proteins Proteomics 2017, 1865 (7), 927–935. 10.1016/j.bbapap.2017.03.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Holst S.; Heijs B.; de Haan N.; van Zeijl R. J.; Briaire-de Bruijn I. H.; van Pelt G. W.; Mehta A. S.; Angel P. M.; Mesker W. E.; Tollenaar R. A.; Drake R. R.; Bovee J. V.; McDonnell L. A.; Wuhrer M. Linkage-Specific in Situ Sialic Acid Derivatization for N-Glycan Mass Spectrometry Imaging of Formalin-Fixed Paraffin-Embedded Tissues. Anal. Chem. 2016, 88 (11), 5904–5913. 10.1021/acs.analchem.6b00819. [DOI] [PubMed] [Google Scholar]
  25. Everest-Dass A. V.; Briggs M. T.; Kaur G.; Oehler M. K.; Hoffmann P.; Packer N. H. N-glycan MALDI Imaging Mass Spectrometry on Formalin-Fixed Paraffin-Embedded Tissue Enables the Delineation of Ovarian Cancer Tissues. Mol. Cell. Proteomics 2016, 15 (9), 3003–3016. 10.1074/mcp.M116.059816. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Saito T.; Watanabe A.; Nakano M.; Matsuo K. MALDI-TOF mass spectrometry imaging for N-glycans on FFPE tissue sections of mouse NASH liver through Sialic acid Benzylamidation. Glycoconj. J. 2021, 38 (2), 167–175. 10.1007/s10719-021-09984-w. [DOI] [PubMed] [Google Scholar]
  27. Drake R. R.; McDowell C.; West C.; David F.; Powers T. W.; Nowling T.; Bruner E.; Mehta A. S.; Angel P. M.; Marlow L. A.; Tun H. W.; Copland J. A. Defining the human kidney N-glycome in normal and cancer tissues using MALDI imaging mass spectrometry. J. Mass Spectrom 2020, 55 (4), e4490 10.1002/jms.4490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Wisztorski M.; Franck J.; Salzet M.; Fournier I. MALDI direct analysis and imaging of frozen versus FFPE tissues: what strategy for which sample?. Methods Mol. Biol. 2010, 656, 303–322. 10.1007/978-1-60761-746-4_18. [DOI] [PubMed] [Google Scholar]
  29. Miyamoto S.; Hsu C. C.; Hamm G.; Darshi M.; Diamond-Stanic M.; Decleves A. E.; Slater L.; Pennathur S.; Stauber J.; Dorrestein P. C.; Sharma K. Mass Spectrometry Imaging Reveals Elevated Glomerular ATP/AMP in Diabetes/obesity and Identifies Sphingomyelin as a Possible Mediator. EBioMedicine 2016, 7, 121–134. 10.1016/j.ebiom.2016.03.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mirnezami R.; Spagou K.; Vorkas P. A.; Lewis M. R.; Kinross J.; Want E.; Shion H.; Goldin R. D.; Darzi A.; Takats Z.; Holmes E.; Cloarec O.; Nicholson J. K. Chemical mapping of the colorectal cancer microenvironment via MALDI imaging mass spectrometry (MALDI-MSI) reveals novel cancer-associated field effects. Mol. Oncol. 2014, 8 (1), 39–49. 10.1016/j.molonc.2013.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Gruner B. M.; Hahne H.; Mazur P. K.; Trajkovic-Arsic M.; Maier S.; Esposito I.; Kalideris E.; Michalski C. W.; Kleeff J.; Rauser S.; Schmid R. M.; Kuster B.; Walch A.; Siveke J. T. MALDI imaging mass spectrometry for in situ proteomic analysis of preneoplastic lesions in pancreatic cancer. PLoS One 2012, 7 (6), e39424 10.1371/journal.pone.0039424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mechref Y.; Kang P.; Novotny M. V. Differentiating structural isomers of sialylated glycans by matrix-assisted laser desorption/ionization time-of-flight/time-of-flight tandem mass spectrometry. Rapid Commun. Mass Spectrom.: Resour. Chem. Mater. 2006, 20 (8), 1381–1389. 10.1002/rcm.2445. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Nie H.; Li Y.; Sun X. L. Recent advances in sialic acid-focused glycomics. J. Proteomics 2012, 75 (11), 3098–3112. 10.1016/j.jprot.2012.03.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Wheeler S. F.; Domann P.; Harvey D. J. Derivatization of sialic acids for stabilization in matrix-assisted laser desorption/ionization mass spectrometry and concomitant differentiation of alpha(2 --> 3)- and alpha(2 --> 6)-isomers. Rapid Commun. Mass Spectrom. 2009, 23 (2), 303–312. 10.1002/rcm.3867. [DOI] [PubMed] [Google Scholar]
  35. Miura Y.; Shinohara Y.; Furukawa J.; Nagahori N.; Nishimura S. Rapid and simple solid-phase esterification of sialic acid residues for quantitative glycomics by mass spectrometry. Chemistry 2007, 13 (17), 4797–4804. 10.1002/chem.200601872. [DOI] [PubMed] [Google Scholar]
  36. Reiding K. R.; Blank D.; Kuijper D. M.; Deelder A. M.; Wuhrer M. High-throughput profiling of protein N-glycosylation by MALDI-TOF-MS employing linkage-specific sialic acid esterification. Analytical Chemistry 2014, 86 (12), 5784–5793. 10.1021/ac500335t. [DOI] [PubMed] [Google Scholar]
  37. de Haan N.; Reiding K. R.; Haberger M.; Reusch D.; Falck D.; Wuhrer M. Linkage-specific sialic acid derivatization for MALDI-TOF-MS profiling of IgG glycopeptides. Analytical Chemistry 2015, 87 (16), 8284–8291. 10.1021/acs.analchem.5b02426. [DOI] [PubMed] [Google Scholar]
  38. Zhang H.; Shi X.; Liu Y.; Wang B.; Xu M.; Welham N. V.; Li L. On-tissue amidation of sialic acid with aniline for sensitive imaging of sialylated N-glycans from FFPE tissue sections via MALDI mass spectrometry. Anal. Bioanal. Chem. 2022, 414 (18), 5263–5274. 10.1007/s00216-022-03894-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Lu X.; McDowell C. T.; Blaschke C. R. K.; Liu L.; Grimsley G.; Wisniewski L.; Gao C.; Mehta A. S.; Haab B. B.; Angel P. M.; Drake R. R. Bioorthogonal Chemical Labeling Probes Targeting Sialic Acid Isomers for N-Glycan MALDI Imaging Mass Spectrometry of Tissues, Cells, and Biofluids. Anal. Chem. 2023, 95 (19), 7475–7486. 10.1021/acs.analchem.2c04882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sekiya S.; Wada Y.; Tanaka K. Derivatization for stabilizing sialic acids in MALDI-MS. Anal. Chem. 2005, 77 (15), 4962–4968. 10.1021/ac050287o. [DOI] [PubMed] [Google Scholar]
  41. Liu X.; Qiu H.; Lee R. K.; Chen W.; Li J. Methylamidation for sialoglycomics by MALDI-MS: a facile derivatization strategy for both alpha2,3- and alpha2,6-linked sialic acids. Anal. Chem. 2010, 82 (19), 8300–8306. 10.1021/ac101831t. [DOI] [PubMed] [Google Scholar]
  42. Farkas P.; Cizova A.; Bekesova S.; Bystricky S. Comparison of EDC and DMTMM efficiency in glycoconjugate preparation. Int. J. Biol. Macromol. 2013, 60, 325–327. 10.1016/j.ijbiomac.2013.06.014. [DOI] [PubMed] [Google Scholar]
  43. Stawikowski M.; Fields G. B. Introduction to peptide synthesis. Curr. Protoc. Protein Sci. 2002, 69, 18.1.1–18.1.13. 10.1002/0471140864.ps1801s69. [DOI] [PubMed] [Google Scholar]
  44. El-Faham A.; Albericio F. Peptide coupling reagents, more than a letter soup. Chem. Rev. 2011, 111 (11), 6557–6602. 10.1021/cr100048w. [DOI] [PubMed] [Google Scholar]
  45. Taylor M. J.; Lukowski J. K.; Anderton C. R. Spatially Resolved Mass Spectrometry at the Single Cell: Recent Innovations in Proteomics and Metabolomics. J. Am. Soc. Mass Spectrom. 2021, 32 (4), 872–894. 10.1021/jasms.0c00439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Zeisel A.; Munoz-Manchado A. B.; Codeluppi S.; Lonnerberg P.; La Manno G.; Jureus A.; Marques S.; Munguba H.; He L.; Betsholtz C.; Drake R. R. Brain structure. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq. Science 2015, 347 (6226), 1138–1142. 10.1126/science.aaa1934. [DOI] [PubMed] [Google Scholar]
  47. LoVerso P. R.; Cui F. Cell type-specific transcriptome profiling in mammalian brains. Front Biosci. 2016, 21, 973–985. 10.2741/4434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Bathke J.; Konzer A.; Remes B.; McIntosh M.; Klug G. Comparative analyses of the variation of the transcriptome and proteome of Rhodobacter sphaeroides throughout growth. BMC Genomics 2019, 20 (1), 358. 10.1186/s12864-019-5749-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Specht H.; Emmott E.; Petelski A. A.; Huffman R. G.; Perlman D. H.; Serra M.; Kharchenko P.; Koller A.; Slavov N. Single-cell proteomic and transcriptomic analysis of macrophage heterogeneity using SCoPE2. Genome Biol. 2021, 22 (1), 50. 10.1186/s13059-021-02267-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Schjoldager K. T.; Narimatsu Y.; Joshi H. J.; Clausen H. Global view of human protein glycosylation pathways and functions. Nat. Rev. Mol. Cell Biol. 2020, 21 (12), 729–749. 10.1038/s41580-020-00294-x. [DOI] [PubMed] [Google Scholar]
  51. Han J.; Huang X.; Liu H.; Wang J.; Xiong C.; Nie Z. Laser cleavable probes for in situ multiplexed glycan detection by single cell mass spectrometry. Chem. Sci. 2019, 10 (47), 10958–10962. 10.1039/C9SC03912K. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Angel P. M.; Mehta A. S.; Drake R. R. Array-Based N-Glycan Profiling of Cells in Culture. Methods Mol. Biol. 2021, 2271, 331–342. 10.1007/978-1-0716-1241-5_23. [DOI] [PubMed] [Google Scholar]
  53. Angel P. M.; Saunders J.; Clift C. L.; White-Gilbertson S.; Voelkel-Johnson C.; Yeh E.; Mehta A.; Drake R. R. A Rapid Array-Based Approach to N-Glycan Profiling of Cultured Cells. J. Proteome Res. 2019, 18 (10), 3630–3639. 10.1021/acs.jproteome.9b00303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Neumuller R. A.; Knoblich J. A. Dividing cellular asymmetry: asymmetric cell division and its implications for stem cells and cancer. Genes Dev 2009, 23 (23), 2675–2699. 10.1101/gad.1850809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Sato S.; Rancourt A.; Sato Y.; Satoh M. S. Single-cell lineage tracking analysis reveals that an established cell line comprises putative cancer stem cells and their heterogeneous progeny. Sci. Rep. 2016, 6, 23328. 10.1038/srep23328. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ac3c05984_si_001.pdf (7.3MB, pdf)

Articles from Analytical Chemistry are provided here courtesy of American Chemical Society

RESOURCES