Time-Resolved and Comprehensive Analysis of Surface Glycoproteins Reveals Distinct Responses of Monocytes and Macrophages to Bacterial Infection

Abstract

Glycoproteins on the surface of immune cells play extremely important roles in response to pathogens. Yet, a systematic and time-resolved investigation of surface glycoproteins during the immune response remains to be explored. Integrating selective enrichment of surface glycoproteins with multiplexed proteomics, we globally and site-specifically quantified the dynamics of surface glycoproteins on THP-1 monocytes and macrophages in response to bacterial infection and during the monocyte-to-macrophage differentiation. The time-resolved analysis reveals transient changes and differential remodeling of surface glycoproteins on both cell types, and potential upstream regulators and downstream effects of the regulated glycoproteins. Besides, we identified novel surface glycoproteins participating in the immune response such as APMAP, and site-specific changes of glycoproteins. This study provides unprecedented information to deepen our understanding of glycoproteins and cellular activities.

Introduction

The immune system protects the body from harmful pathogens such as bacteria and viruses that can cause adverse health effects. Innate immunity is the first and immediate mechanism that the body employs to respond to these pathogens in a non-specific manner, including physical barriers such as the epithelium that prevents pathogens from entering the body, and leukocytes such as macrophages and natural killer cells that may be recruited to the infection site.^[1] Adaptive immunity, however, is the delayed and specific response to the infection, during which B and T cells are critical.^[2] Cell-surface glycoproteins regulate many cellular events in the immune system, for example, sensing pathogens and functioning as receptors or transporters for cellular communication to recruit different immune cells to the infection location.^[3] The innate and adaptive immune systems are also interconnected by antigen-presenting cells, including macrophages and dendritic cells that have major histocompatibility complex (MHC) proteins on their surface.^[4]

Gram-negative bacteria contain lipopolysaccharides (LPS) on the cell wall that are released into the environment when the cell integrity is disrupted.^[5] LPS binds to pattern recognition receptors (PRRs) on the cell surface, particularly those in the toll-like receptor (TLR) family through pathogen-associated molecular pattern (PAMP) recognition. LPS triggers signaling cascades that activate transcription factors including NF-κB, AP-1, and IRF3. This eventually results in the secretion of pro-inflammatory cytokines such as TNF-α or chemokines that attract other cells into the vicinity.^[6] Monocytes and macrophages are important cells in the innate immune system that respond to LPS, and thus several cell-culture models such as THP-1 and U937 human monocytes or RAW264.7 mouse macrophages have been widely used for in vitro studies of the immune response to avoid the donor-to-donor variation issue with peripheral blood mononuclear cells (PBMCs).^[7] Despite the importance of surface glycoproteins, their responses in the immune response process remain to be explored particularly in a time-resolved and site-specific manner, which can pinpoint the transient and important events that may not be observed in a typical study.

In this work, combining metabolic labeling, bioorthogonal chemistry, and multiplexed mass spectrometry (MS)-based proteomics, we investigated the time-resolved and site-specific responses of surface glycoproteins on THP-1 monocytes and macrophages upon the LPS stimulation and during the monocyte-to-macrophage differentiation. Differential remodeling of the surface glycoproteomes was observed among the cells, including the expression of new glycoproteins to the surface and the removal/internalization of existing surface glycoproteins. Some surface glycoproteins transiently altered their abundances before returning to the normal state, which would not be observed in a typical experiment. Nonetheless, some glycoproteins were minimally affected. A comparison of the surface glycoproteome changes in response to LPS between monocytes and macrophages revealed the similarities and differences, and the priming of monocytes for the response during the differentiation process. Apart from previously reported markers, novel surface glycoproteins quantified in the immune response process in this work may serve as potential markers. Furthermore, selective and site-specific protein glycosylation was observed in different processes. This work results in a better understanding of the important roles of cell-surface glycoproteins in the immune response during the bacterial infection and potentially the identification of surface glycoproteins as disease biomarkers and drug targets.

Results and Discussion

Modern mass spectrometry-based proteomics provides a possibility to globally analyze protein glycosylation.^[8] To target surface glycoproteins, we need to selectively separate them prior to MS analysis.^[9] This study systematically investigated the dynamics of glycoproteins on the surface of THP-1 monocytes and macrophages in response to LPS in time-resolved and site-specific manners by combining metabolic labeling, bioorthogonal chemistry, and multiplexed proteomics (Figure 1).^[10] THP-1 cells were labeled with 100 μM N-azidoacetylgalactosamine-tetraacylated (Ac₄GalNAz). Surface glycoproteins containing the functional azido group were then tagged with 100 μM dibenzocyclooctyne (DBCO)-biotin. After cell lysis, protein extraction and digestion, the tagged glycopeptides from surface glycoproteins were selectively enriched using NeutrAvidin beads, followed by deglycosylation with PNGase F in heavy-oxygen water (H₂¹⁸O) for MS analysis. The glycopeptides from the quantification experiments were also labeled with the TMT reagents to determine the protein dynamics.

Figure 1.

Experimental procedure for global quantification of the dynamics of surface glycoproteins. A) Experiments with THP-1 monocytes and macrophages, including the LPS treatment and the differentiation induced by PMA. B) General time-resolved experiment setup to determine the dynamics of cell-surface glycoproteins.

Overall, 2205 unique glycopeptides from 764 cell-surface glycoproteins were detected from all identification and quantification experiments (Figure 2A; Supporting Information, Table S1). Over 1000 glycopeptides and >400 glycoproteins were found commonly in both monocytes and macrophages. Gene ontology (GO)-based clustering of all the identified glycoproteins based on biological process showed that immune system process (P=4.99 × 10⁻⁵⁷), cell adhesion (P=4.93 × 10⁻⁵¹), cell-surface receptor signaling pathway (P=3.84 × 10⁻⁴¹), and transmembrane transport (P=9.60 × 10⁻¹¹) are highly enriched, which is in excellent agreement with the known functions of cell-surface glycoproteins (Figure 2B). Based on cellular component, over 600 glycoproteins are annotated to the membrane part (P=5.27 × 10⁻¹¹⁶) while about 500 are on the plasma membrane (P=9.71 × 10⁻¹⁰¹), demonstrating that the current experiments are effective for specific detection of glycoproteins on the cell surface. The reproducibility of the results from the biological duplicate experiments is reasonably high, such as those from the identification experiments with monocytes (Figure 2C).

Figure 2.

Glycoproteins identified from all experiments and some functional analysis results. A) The numbers of all detected glycoproteins and glycopeptides. B) GO clusters of the identified glycoproteins based on biological process. The full list is in the Supporting Information, Table S1D. C) Overlaps of glycoproteins and glycopeptides on monocytes from the duplicate identification experiments. D) Clustering of surface glycoproteins found exclusively in monocytes (n=180) or macrophages (n=107). The x-axis represents the number of surface glycoproteins in each cluster or the -log(P) value, and both share the same x-axis. The P-values for protein clustering were determined using DAVID.

Some glycoproteins were found only in a specific cell type despite the identical genome backgrounds of THP-1 monocytes and macrophages. Clustering of the 180 proteins identified exclusively in monocytes revealed highly enriched processes such as humoral immune response mediated by circulating immunoglobulin (P=9.30 × 10⁻¹⁰) and regulation of complement activation (P=3.2 × 10⁻⁸) (Figure 2D). For the 107 proteins found only in macrophages, those involved in the regulation of cytokine secretion (P=1.30 × 10⁻⁸), response to molecules of bacterial origin (P=2.20 × 10⁻⁷), and regulation of adaptive immune response (P=8.00 × 10⁻⁴) are highly enriched (Figure 2D). Some example proteins are described in detail in Supporting Information.

In the experiments to identify and quantify surface glycoproteins from THP-1 monocytes, more than 1500 unique glycopeptides from over 600 glycoproteins were detected (Figure 3A). In the quantification experiments, the cells were pre-labeled with Ac₄GalNAz for 12 h before the treatment with LPS (1.0 μgmL⁻¹) or PBS (as a control) to allow sufficient labeling of surface glycoproteins. The cells were collected at six time points, and the glycopeptides were labeled with the TMT 6-plex reagents and analyzed with LC-MS/MS (Figure 1B). In one example, the glycopeptide RGPECSQN#YTTPSGVIK (# refers to the glycosylation site) from neuropilin-1 (NRP1) was confidently identified and quantified (Figure 3B). While the abundance of this glycopeptide stayed relatively similar across 24 h in the control group, its abundance in the cells treated with LPS increased rapidly at 12 and 24 h (Figure 3C).

Figure 3.

Quantification of cell-surface glycoproteins on monocytes in response to LPS. A) Overlaps of glycoproteins and glycopeptides from the identification and quantification experiments. B) An example MS/MS spectrum of a quantified glycopeptide (see text). The highlighted blue region shows the TMT reporter ions. C) The TMT ratios of the peptide from B) demonstrating the difference between the control (green) and the treatment (blank) groups. D)–F) Time-resolved plots of surface glycoprotein abundances over 24 h from the PBS-treated (control, D) or LPS-treated E) cells, and the normalized abundances F). G) Total intensities of glycoproteins that changed their abundances in four different patterns. The plot shows glycoproteins that increased their abundances gradually over time (Type A), increased abruptly and returned to normal (Type B), decreased (Type C), and were not affected (Type D). H)–K) Example glycoproteins from Type D (H), Type A (I), Type B (J), and Type C (K). The raw and normalized abundances are displayed. The error bars represent one standard deviation.

In the control experiments, the abundances of cell-surface glycoproteins generally increased over time due to cell growth (Figure 3D), in contrast to the cells treated with LPS where dramatic changes were observed (Figure 3E). The abundance at each time point from the treatment group was then normalized by the abundance of the corresponding surface glycoproteins from the control group (Figure 3F), resulting in 247 glycoproteins in the final dataset with high reproducibility (Supporting Information, Figures S1 and S2A, and Table S2). The quantified glycoproteins were classified into four categories based on their abundance changes (Figure 3G; details are in the Supporting Information): 12 glycoproteins (5%, Type A) increased their abundances gradually and did not return to the baseline; 18 (7%, Type B) increased their abundances abruptly and returned to the baseline; 92 (37%, Type C) decreased their abundances; and 64 (26%, Type D) were minimally affected. Type A glycoproteins are involved in the regulation of inflammatory response, regulation of programmed cell death, and cell development. Those involved in protein transport and ion sequestering (especially calcium), and cytokine production (IL-2, IL-4, IL-10, and IFN-γ) are enriched in Type B. Glycoproteins involved in extracellular matrix organization, positive regulation of cell proliferation, interleukin production (IL-5 and IL-13), and cell-cell adhesion are overrepresented among Type C glycoproteins. Nonetheless, some glycoproteins did not meet the criteria for any above patterns, but instead their abundances fluctuated over the treatment period, and hence the type cannot be assigned (Supporting Information, Figure S2B).

One protein that stood out in the profile plot is CD83, a Type B glycoprotein that has the highest abundance compared with other glycoproteins at 3 h (Figure 3J). CD83 is a marker for mature dendritic cells and involved in antigen presentation.^[11] The previous report showed that the protein can be pre-formed inside immune cells and exported to the surface once cells were stimulated with LPS.^[12] In contrast, CD40 gradually increased its abundance over time (Figure 3I). CD40 is typically expressed on B cells and binds to the CD40L ligand on T cells under inflammatory conditions^[13] Some glycoproteins, such as integrin alpha-L (IT-GAL; CD11a), decreased their abundances over time (Figure 3K). These proteins may be internalized and degraded by cells or shed into the extracellular environment by ectodomain shedding.^[14] Other glycoproteins were not affected by LPS, such as CD172a (Figure 3H), and they may not participate in the response towards LPS or may be regulated through their interactions with other molecules rather than changing their abundances. Other example proteins are displayed in Table 1 and the Supporting Information, Figure S3. Some of these proteins, to our knowledge, have never been linked to the cellular response towards LPS and may play important roles in the immune response process, such as adipocyte plasma membrane-associated protein (APMAP) that has arylesterase activity. The protein increased its abundance abruptly to 1.8-fold at 3 h after the LPS challenge before returning to normal.

Table 1:

Example surface glycoproteins quantified from THP-1 monocytes and macrophages In response to LPS or during the monocyte-to-macrophage differentiation.

Experiment	UniProt ID	Gene Symbol	Type	Annotation
Monocytes	Q9BZZ2	SIGLEC1	A	Sialoadhesin
	Q99571	P2RX4	B	P2X purinoceptor 4
	P55899	FCGRT	C	IgG receptor FcRn large subunit p51
	P20701	ITGAL	C	Integrin alpha-L
	Q15262	PTPRK	D	Receptor-type tyrosine-protein phosphatase kappa
Macrophages	P28907	CD38	A	ADP-ribosyl cyclase/cyclic ADP-ribose hydrolase 1
	P13726	F3	B	Tissue factor
	Q9UKQ2	ADAM28	B	Disintegrin and metalloproteinase domain-containing protein 28
	P20138	CD33	C	Myeloid cell surface antigen CD33
	P15260	IFNGR1	C	Interferon gamma receptor 1
	A6NGU5	GGT3P	C	Putative glutathione hydrolase 3 proenzyme
Differentiation	O14786	NRP1	A	Neuropilin-1
	P18084	ITGB5	A	Integrin beta-5
	Q9Y6M7	SLC4A7	A	Sodium bicarbonate cotransporter 3
	Q01151	CD83	B	CD83 antigen
	P33527	ABCC1	B	Multidrug resistance-associated protein 1
	P02786	TFRC	C	Transferrin receptor protein 1
	P10586	PTPRF	C	Receptor-type tyrosine-protein phosphatase F

Over 1600 glycopeptides from >500 surface glycoproteins were detected in THP-1-derived M0 macrophages (Supporting Information, Figure S4A). The differentiated macrophages were similarly treated with either LPS (at the same concentration as that in the monocyte experiment) or PBS (Supporting Information, Figures S4B and S4C). Eventually, 223 glycoproteins were quantified commonly (Figure 4A; Supporting Information, Figures S4A and S4D, and Table S3). The quantified proteins were classified into the four response types as above, that is, 24 glycoproteins belonging to Type A (11%), 36 Type B (16%), 68 Type C (30%), and 56 Type D (25%). Type A glycoproteins are related to the nitric oxide metabolic process, negative regulation of programmed cell death, and positive regulation of reactive oxygen species biosynthetic process. Type B glycoproteins are involved in the positive regulation of adaptive immune response, sequestering of ions, including calcium similar to that of monocytes, detection of bacterium, and positive regulation of chemotaxis. Antigen processing and presentation of endogenous peptide antigen were also enriched in this group. Those that decreased their abundances (Type C) participate in cell migration and the regulation of cell-cell adhesion.

Figure 4.

Quantification of cell-surface glycoproteins on macrophages in response to LPS. A) Normalized abundances of cell-surface glycoproteins from macrophages in response to LPS. B) The abundance changes of example surface glycoproteins. The control (PBS), treatment (LPS), and normalized intensities are shown similar to Figure 3. The error bars represent one standard deviation. C) Comparison of the abundance changes of some surface glycoproteins on monocytes and macrophages in response to LPS. D),E) Abundance changes of ICAM1 D) and CD38 E) on the cell surface showing the priming of the cells during the monocyte-to-macrophage differentiation for their specific functions.

Examples of glycoproteins that significantly changed their abundances on the surface include tumor necrosis factor receptor superfamily member 9 (TNFRSF9; CD137). The protein reached its highest abundance at 6 h after the treatment, which is consistent with two previous MS-independent studies (Figure 4B).^[15] HLA-A, a protein that presents antigens to the cell surface, had a peak abundance at 12 h, suggesting the activation or connection of the innate and adaptive immune systems. T-lymphocyte activation antigen CD80, a marker for the M1 subtype, increased its abundance on the cell surface, showing that the LPS treatment might polarize them toward the M1 pro-inflammatory state as reported previously.^[16] It was not detected in the control experiment, likely due to the higher expression induced by LPS, with an increased abundance over 13 times after 24 h (Supporting Information, Figure S5). Other quantified glycoproteins are included in Table 1 and the Supporting Information, Figure S6 and Table S3. Similar to monocytes for glycoproteins that have never been associated with the response to LPS, APMAP increased its abundance to 2.7-fold after 6 h in response to LPS.

A total of 116 glycoproteins were quantified commonly in both THP-1 monocytes and macrophages with a determined response type (A, B, C, or D). Among those, 61 glycoproteins (53%) involved in general processes such as heterotypic cell-cell adhesion and cell mobility have the same response types, while the responses from 55 glycoproteins (47%) participating in the regulation of defense response and detection of bacterium are different. Nevertheless, glycoproteins involved in leukocyte activation and innate immune response may have the same or different response types. For instance, the abundances of NRP1 and CD115 in monocytes increased gradually while those in macrophages stayed relatively similar or decreased, respectively. On the contrary, HLA-A has a greater abundance in macrophages after the LPS challenge (Type B in macrophages) but is not affected (Type D) in monocytes, corresponding well with previous reports regarding the antigen-presenting capability of macrophages.^[7a] Some glycoproteins responded similarly in both cell types, such as NPTN (Type B) and multidrug resistance-associated protein 1 (ABCC1; Type B) (Figure 4C).

Comparative studies of the two cell types previously showed that macrophages typically secreted more cytokines and chemokines than monocytes after the LPS treatment.^{[7b, 17]} Interestingly, some glycoprotein abundances did not change as expected under the LPS treatment. For example, intercellular adhesion molecule 1 (ICAM1; CD54) was up-regulated in several previous studies after the LPS treatment or inflammation.^[18] In this work, its abundance increased by over 8.4 times in monocytes but only 1.9 times in macrophages after 6 h (Figure 4D). On the other hand, the abundance of ADP-ribosyl cyclase/cyclic ADP-ribose hydrolase 1 (CD38) decreased in monocytes but increased in macrophages (Figure 4E). These responses may be explained by the plasticity of the cells and the differentiation process (discussed further in the next section) that leads to the expression of some glycoproteins on the surface required for their proper functions depending on the cell type. Previous reports demonstrated that the differentiation of monocytes into macrophages prepared the cells for their functions in the immune system. For example, the phorbol myristate acetate (PMA) treatment was found to increase the abundance of bound NF-κB in the cytoplasm (inactive form). Once the cells were challenged with LPS, more free NF-κB (active form) migrated into the nucleus, resulting in enhanced secretion of TNF-α and greater response to LPS.^[19] In the case of surface glycoproteins, the differentiation causes the surfaceome remodeling and hence primes them for promptly functioning in the immune response process similar to the increased abundance of the inactive NF-κB. For ICAM1, its surface abundance on monocytes is low before the LPS treatment and increases to >8 times after six hours so that they can participate in the response to LPS. The differentiation of monocytes into macrophages causes the abundance of ICAM1 to increase gradually to >5 times after 48 h. Since its surface abundance is already high, macrophages do not need to express more ICAM1 to the surface as occurred in monocytes, and thus its abundance change in macrophages responded to the infection is relatively lower.

To better understand the different responses of THP-1 monocytes and macrophages, we further quantified the surface glycoproteome changes during the monocyte-to-macrophage differentiation. As reported previously,^[17] the monocyte-to-macrophage differentiation was induced by PMA (details in Supporting Information), and surface glycoproteins were quantified as above. More than 900 glycopeptides from >300 surface glycoproteins were detected in the differentiation and the control experiments, with over 200 glycoproteins quantified commonly in both experiments (Supporting Information, Figure S7A and Table S4). Similarly, 27 Type A (12%), 28 Type B (12%), 117 Type C (51%), and 32 Type D (14%) glycoproteins were found during the differentiation (Figure 5). Type A glycoproteins function in the response to lipoprotein particle, positive regulation of cytokine production, cell differentiation, and negative regulation of programmed cell death, indicating that the cells were transformed into the phenotype to respond to the infection. Type B proteins are involved in the regulation of ion transport, receptor internalization, and wound healing. Other proteins that decreased their abundances are involved in processes such as protein maturation and positive regulation of leukocyte proliferation. Based on previously reported markers for the monocyte-to-macrophage differentiation, the current results clearly demonstrated that macrophages were obtained through the differentiation, including integrin alpha-M (ITGAM; CD11b), integrin alpha-X (ITGAX; CD11c), and ICAM1 that increased their abundances, CD280 and integrin alpha-4 (ITGA4; CD49d) that decreased their abundances, and CD36 and CD44 that increased their abundances abruptly at 6 h (Figure 5C).^{[7b, 20]} Other affected surface glycoproteins are shown in the Supporting Information, Table S4 and Figure S8.

Figure 5.

Surface glycoprotein analysis during the monocyte-to-macrophage differentiation. A) Normalized abundances of cell-surface glycoproteins during the monocyte-to-macrophage differentiation. B) Heat map based on hierarchical clustering with log₂- and Z-score-transformed normalized abundances showing changes of cell-surface glycoproteins during the differentiation. C) Surface glycoprotein markers for the monocyte-to-macrophage differentiation. The control (PBS), treatment (LPS), and normalized intensities are displayed similar to Figure 3. The error bars represent one standard deviation.

Through global and time-resolved quantification of surface glycoproteins in cells treated with LPS, we found the abundance changes of many surface glycoproteins including the transient changes. Potential upstream regulators of these regulated glycoproteins and their downstream influences were further explored through ingenuity pathway analysis (IPA), and the results are in Figure 6 and the Supporting Information, Figure S9 and Table S5.^[21] These results aid in a better understanding of the roles of these surface glycoproteins in the immune response. For instance, in macrophages after the LPS treatment for 6 h, the transcription factors EGR1, SP1, RELA, and FOS, the enzymes NOS2 and KRAS, and other proteins involved in the inflammatory process and in response to molecules of bacterial origin were predicted to be activated, resulting in the expression of more ICAM1, CD40, CD44, and CD55 on the cell surface as observed in the study.^[22] This eventually causes downstream effects including cell survival, synthesis of reactive oxygen species, and invasion of cells. At 6 h after the LPS treatment, the majority of upstream regulators and downstream influences were also predicted to be in effect in macrophages similar to the maximum response time in a transcriptomic study.^[17]

Figure 6.

IPA analysis of upstream regulators and downstream effects of the regulated surface glycoproteins on macrophages treated with LPS for 6 h. The full list is included in the Supporting Information, Table S5.

The LPS treatment clearly triggers the TLR4 signaling pathway as seen by the activation of NF-κB, MAPK3 (ERK1), MAPK1 (ERK2), and P38 MAPK in both cell types. This resulted in changes in surface glycoproteins that may signal and recruit other types of immune cells into the vicinity, shown as the downstream effects including chemotaxis, infection of cells, interaction of mononuclear leukocytes, and infection of T-lymphocytes. Nevertheless, the IPA results demonstrate some different outcomes of the LPS challenge among the cell types. The majority of regulators and downstream influences involved in the immune response were predicted at 12 h in monocytes, and the production of reactive oxygen species was not predicted as a possible downstream effect in monocytes. This response was found to occur as fast as 6 h in macrophages, which possibly results from the differentiation that primes the cell for the response to bacteria (Supporting Information, Table S5).

For the PMA-induced monocyte-to-macrophage differentiation, most upstream regulators were predicted at 12 h (Supporting Information, Table S5). PMA was among the upstream regulators as expected. Known proteins involved in the differentiation of leukocytes as upstream regulators including F2, PAX5, and LDB1 were predicted. The IPA analysis demonstrates that the regulated surface glycoproteins are responsible for cell attachment, reorganization of cytoskeleton, activation of antigen-presenting cells, and cell movement of phagocytes, indicating the change of phenotypes from monocytes to macrophages (Supporting Information, Table S5).

A total of 1869 sites with the N-X-S/T/C motif were detected from all experiments, with several hundred sites identified exclusively in each cell type (Figure 7B; Supporting Information, Table S1). Distinct responses of different glycosylation sites on the same proteins were observed (Figure 7A; Supporting Information, Figure S10, Table S6). For instance, glycosylation on N229 of granulocyte-macrophage colony-stimulating factor receptor subunit alpha (CSF2RA) increased to 4.9 times in monocytes after the LPS treatment while the sites of N195 and N223 remained relatively similar throughout the treatment (Figure 7A). Some glycoproteins did not show site-specific changes, such as N150 and N522 on NRP1, which may result from the increased protein abundance on the surface. For ICAM1, glycosylation on N267 increased more rapidly up to 6 h (16.7 times) compared with that on N145 and N183 (Figure 7C). The differentiation resulted in an increased glycosylation level on the site 267 by 21.9 times, and only 4.0 and 2.2 times for the other two sites, while the glycosylation level in macrophages for these sites are similar and even lower than those from monocytes, suggesting that N267 may be crucial for ICAM1 to function properly during the response. Different ICAM1 proteoforms can arise from alternative splicing, and specific glycosylation event further increases the complexity of the proteoforms.^[23] For ICAM1, it was reported to increase its binding affinity with lymphocyte function-associated antigen 1 (LFA-1).^[24] Thus, systematic and site-specific analysis of protein glycosylation facilitates a better understanding of glycoprotein functions and cellular immune responses.

Figure 7.

Site-specific quantification of surface glycoproteins. A) Examples of glycoproteins with multiple quantified glycosylation sites showing site-specific changes. The normalized fold changes are displayed in the plots with the glycosylation sites annotated below the plot. B) Overlap of glycosylation sites identified. C) Site-specific glycosylation changes of ICAM1 from different processes in both monocytes and macrophages.

Bacterial infection can lead to inflammation,^[25] fever, endotoxemia, septic shock,^[26] liver damage,^[27] and autoimmune diseases such as Rheumatoid arthritis.^[28] LPS has also been associated with diabetes^[29] and neurological diseases including Parkinson’s and Alzheimer’s.^[30] Depending on the environment and cell types, various surface glycoproteins have an altered expression after the infection, as demonstrated in this work. Moreover, the expression changes of some surface glycoproteins such as receptors or transporters may also indicate specific infection within the body.^[31] A previous report showed that the inflammatory monocyte subset is a valuable biomarker for human inflammatory diseases, including cardiovascular diseases, and may provide a potential therapeutic target for inflammatory monocytosis.^[32] The up- or down-regulation of surface glycoproteins are crucial for the cellular response towards LPS/inflammation. Manipulation of these glycoproteins, the pathways regulating their expression, or the transcription factors responsible for their translation will inhibit the inflammation and affect the production of downstream cytokines.^[33] The unprecedented and valuable information about the cell surface glycoprotein changes in the immune response can deepen our understanding of glycoprotein functions and cellular activities. More studies will allow us to identify surface glycoproteins as disease biomarkers, which can be further developed as a clinical assay, and targets for drug development.

Conclusion

Cell-surface glycoproteins play crucial roles in the immune response process. These glycoproteins are very dynamic to allow cells to adapt to ever-changing environment. However, it is highly challenging to globally quantify the dynamics of glycoproteins only on the cell surface. Here, coupling selective enrichment with multiplexed proteomics, we comprehensively and site-specifically quantified the dynamics of surface glycoproteins in monocytes and macrophages in response to LPS. The quantification results demonstrate cell-surface glycoprotein changes in monocytes and macrophages during the response, including gradual or transient changes. The responses are found to be different for both monocytes and macrophages. Besides the well-documented glycoprotein changes in response to LPS, we also identified some new surface glycoproteins participating in the immune response such as APMAP, TSPAN3, and IGSF8. Moreover, the current work provides site-specific information regarding protein glycosylation changes during the LPS treatment, for instance, the site N229 of CSF2RA but not N195 and N223 in monocytes. The quantification of glycoproteins in the differentiation experiment from monocyte to macrophage revealed that the different responses were at least partly attributed to the priming of monocytes during the differentiation. Systematic investigation of the dynamics of surface glycoproteins results in a better understanding of glycoprotein functions and cellular immune responses, leading to the identification of surface glycoproteins as biomarkers and drug targets.