RENBP inhibition amplifies metabolic glycan labeling efficiency of antigen-presenting cells in vitro and in vivo

SUMMARY

Metabolic glycoengineering of unnatural sugars provides a powerful tool to introduce unique chemical tags onto cell membrane for subsequent conjugation of cargos. However, the metabolic glycan labeling efficiency of antigen-presenting cells (APCs), the key mediators of adaptive immunity, is often low. Here, we report that APCs upregulate GlcNAc 2-epimerase (RENBP) and that RENBP inhibition leads to improved labeling efficiency of tetraacetyl-N-azidoacetylmannosamine (AAM) in APCs, including dendritic cells (1.2-fold), macrophages (1.3-fold), and B cells (1.4-fold) in vitro. RENBP inhibition can preferentially enhance AAM labeling efficiency in APCs than in non-APCs and selectively enhance the labeling efficiency of AAM over azido-galactosamine. We further demonstrate that RENBP inhibitors can improve AAM-mediated labeling of B cells and other APCs in vivo, with the largest enhancement for B cells (>3-fold) for 7days. Our study uncovers a facile approach to improving metabolic glycan labeling of APCs, enabling the development of APC-targeted immunotherapies.

In brief

Liu et al. show that antigen-presenting cells (APCs) upregulate GlcNAc 2-epimerase (RENBP) and its inhibition leads to enhanced labeling efficiency of tetraacetyl-N-azidoacetylmannosamine in APCs. RENBP inhibitors can improve azido labeling of B cells and other APCs both in vitro and in vivo, providing a promising strategy to develop APC-targeted immunotherapies.

Graphical Abstract

INTRODUCTION

Metabolic glycan engineering (MGE), which utilizes the glycan synthetic machinery to incorporate unnatural sugars into glycol-proteins and glycolipids, has been extensively explored in both eukaryotic and prokaryotic cells.^1–8 This approach enables the metabolic labeling of various types of cells, including bacteria,^4–8 cancer cells,^9–11 and stem cells^12,13 with unnatural chemical tags (e.g., azide, ketone, alkyne, and alkene) for cell tracking and targeting purposes.^14–16 For example, azido-sugars, such as N-azidoacetyl-mannosamine (ManNAz),¹⁷ N-azidoacetylgalactosamine (GalNAz),¹⁸ and N-azidoacetyl-glucosamine (GlcNAz)^19,20 can metabolically label cancer cells with azido groups, for subsequent targeted conjugation of dibenzocyclooctyne (DBCO)-bearing anticancer drugs via click chemistry between azide and DBCO.^21–25 Recent advances in MGE have also encouraged its application in the field of immunoengineering and immunotherapy, by metabolically tagging and specifically modulating immune cells of interest.^25–28 Among them, the targeted modulation of antigen-presenting cells (APCs), key mediators of adaptive immunity in the body, holds great promise for the development of robust vaccines and immunotherapies.²⁹ However, compared to cancer cells and bacteria, the labeling efficiency of APCs, such as dendritic cells, macrophages, and B cells remains low,^25,30,31 motivating the development of facile strategies to boost the metabolic glycan labeling efficiency of unnatural sugars toward APCs.

In mammalian cells, N-acetylmannosamine (ManNAc) analog bearing unnatural chemical tags can be internalized by cells and phosphorylated by ManNAc kinase (UDP-GlcNAc 2-epimerase [GNE]) to form N-acetylmannosamine-6-phosphate, which is then converted to N-acetylneuraminic acid-9-phosphate.^32–34 Further, N-acetylneuraminic acid-9-phosphate is dephosphorylated to form N-acetylneuraminic acid (Neu5Ac) which is subsequently converted to CMP-Neu5Ac via CMP-sialic acid synthetase. CMP-Neu5Ac analogs are then conjugated to proteins or lipids in the presence of sialyltransferase in the Golgi apparatus, resulting in the expression of chemically tagged glycoproteins and glycolipids on the cell membrane (Figure 1A).^32–34 Several factors affect the labeling efficiency of ManNAc analogs, including cell type-specific metabolic pathways,³⁵ expression levels of participating enzymes,^36,37 and variations in cellular uptake efficiency and pathways of ManNAc analogs.³⁵ While optimizing the incubation time, concentration, and cellular uptake rate of ManNAc analogs can increase their metabolic labeling efficiency to some extent, it is still insufficient for the labeling of APCs. Alternatively, the modulation of key enzymes could also potentially enhance the metabolic labeling efficiency. Prior research has identified two enzymes involved in sialic acid synthetic pathways, i.e., GNE and GlcNAc 2-epimerase (RENBP).^33,38 GNE regulates the phosphorylation of ManNAc analogs, while RENBP catalyzes the reversible conversion between GlcNAc analogs and ManNAc analogs with an equilibrium ratio of 3.9:1.^38,39 While the impact of RENBP on the labeling efficiency of ManNAc analogs was demonstrated using RENBP-overexpressing Jurket cells,³⁸ the ability of RENBP and GNE inhibitors to tune the labeling efficiency of ManNAc analogs in wild-type cells including APCs remains unexplored.

Figure 1. RENBP and GNE are upregulated in immune cells and APCs.

(A) Schematic illustration of sialic acid biosynthesis pathway in mammalian cells, which enables the metabolic incorporation of unnatural mannosamine analogs into glycans.

(B and C) Analysis of RENBP mRNA expression in different types of cells. Single-cell mRNA data were obtained from the Human Protein Atlas. (B) Expression levels of RENBP in individual cell types, showing data with nTPM >5.

(C) Proportion of RENBP expression in grouped cell types relative to total RENBP expression.

(D and E) Analysis of GNE mRNA expression in different types of cells. Single-cell mRNA data were obtained from the Human Protein Atlas. (D) Expression levels of GNE in individual cell types, showing data with nTPM >5. (E) Proportion of GNE expression in grouped cell types relative to total GNE expression.

(F and G) Western blot analysis of RENBP protein expression in DC2.4, A549, A20, and Jurkat cells, using β-tubulin as a loading control. (F) Representative western blot images. (G) Quantification of chemiluminescence signals for RENBP (normalized to β-tubulin).

(H) Flow cytometry analysis of intracellular RENBP in DC2.4, A549, A20, and Jurkat cells using PE-conjugated anti-RENBP antibodies. Expression levels were normalized to cell volume for cross-cell type comparisons.

(I and J) Western blot analysis of RENBP in DC2.4, RAW264.7, A20, NIT-1, 4T1, and EG7-OVA cells, using GAPDH as a loading control. (I) Representative western blot images. (J) Quantification of chemiluminescence signals for RENBP (normalized to GAPDH).

Data are presented as mean ± SD (one-way ANOVA with post hoc Fisher’s LSD test was used; 0.01 < *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001).

Here, we show that RENBP inhibition can result in significantly improved labeling efficiency of ManNAc analogs in APCs, including dendritic cells, macrophages, and B cells. Compared to GNE inhibitors that induce high cytotoxicity, the inhibition of RENBP provides a safer and efficient strategy to enhance the metabolic labeling efficiency of ManNAc analogs (e.g., tetraacetyl-N-azidoacetylmannosamine [AAM]) in APCs. RENBP inhibition also prolongs the membrane retention of cell-surface azido tags. We further show that the co-administration of AAM and RENBP inhibitor can enhance the azido labeling efficiency of monocytes, granulocytes, and B cells in vivo, with the largest enhancement for B cells. Our study uncovers a facile approach to enhancing the metabolic glycan labeling efficiency of APCs in vitro and in vivo, opening new opportunities for the development of APC-targeted immunotherapies.^25,30,31

RESULTS

APCs exhibit high expression levels of RENBP and GNE

We first analyzed the expression levels of RENBP and GNE mRNAs in human cells based on the RNA sequencing database (the Human Protein Atlas).⁴⁰ As a result, blood and immune cells tend to express the highest levels of RENBP mRNA among all types of cells, followed by epithelial cells and pigment cells (Figures 1B and 1C). For GNE, glandular epithelial cells express the highest level of GNE mRNA, followed by specialized epithelial cells and blood and immune cells (Figures 1D and 1E). Among the blood and immune cells, APCs, such as macrophages, granulocytes, and monocytes are among the RENBP- and GNE-overexpressing cells (Figures 1B and 1D). We also analyzed murine RENBP expression levels based on the single-cell RNA sequencing data from the EMBL-EBL single cell expression atlas,⁴¹ which shows a similar expression profile to human RENBP (Figures S1A and S1B). To validate the overexpression of RENBP in APCs, we performed the western blot and flow cytometry analysis of RENBP in DC2.4 cells (a murine dendritic cell line), A549 lung cancer cells, A20 B cell lymphoma cells, and Jurkat T cells. It is noteworthy that the anti-human RENBP we used is able to detect both human and murine RENBP, as a result of the highly conserved structure of human and murine RENBP (Figures S1C and S1D). Compared to A549 lung cancer cells and Jurkat T cells, both DC2.4 and A20 cells, two APC lineage cells, show significantly higher expression levels of RENBP (Figures 1F and 1G). Flow cytometry analysis also confirmed the much higher expression level of RENBP in DC2.4 and A20 cells than A549 lung cancer cells and Jurkat T cells (Figure 1H). We also extended the western blot analysis to more murine cell lines. Consistently, APCs (DC2.4, RAW264.7, and A20) express a higher level of RENBP than non-APCs (4T1, EG7-OVA, and NIT-1) (Figures 1I and 1J). These data confirmed the upregulated expression of RENBP in APCs.

Inhibition of RENBP and GNE amplifies labeling efficiency of AAM in DC2.4 cells

We next studied whether the inhibition of RENBP or GNE can enhance the labeling efficiency of AAM in DC2.4 cells. Three GNE inhibitors (I1, I2, and I3) and one RENBP inhibitor (Ac₅GlutolNAc) were studied (Figures 2A, S2A, and S2B). DC2.4 cells were incubated with PBS or 25 μM AAM in the presence of varied concentrations of inhibitors for 24 h, followed by incubation with DBCO-Cy5 to detect cell-surface azido groups via click chemistry. Compared to PBS group, AAM-treated cells showed higher but only ~1.5-fold Cy5 fluorescence intensity (Figure 2B) and 4% Cy5-positive population (Figure S2C), indicating the low metabolic labeling efficiency of AAM in DC2.4 cells. All GNE inhibitors (I1, I2, and I3) and RENBP inhibitor (Ac₅GlutolNAc) resulted in increased Cy5 fluorescence intensity of DC2.4 cells in a concentration-dependent manner (Figures 2B, 2C, and S2C), demonstrating the feasibility to enhance the labeling efficiency of AAM in DC2.4 cells by inhibiting GNE or RENBP. The 2D scatterplots and histograms also confirm the enhancement of AAM labeling efficiency in the presence of GNE or RENBP inhibitors (Figures S2D and S2E). However, GNE inhibitors led to significant cytotoxicity toward DC2.4 cells (Figure 2D). In contrast, Ac₅GlutolNAc showed minimal cytotoxicity against DC2.4 cells even at a concentration of 400 μM (Figure 2D). Overall, 200–400 μM Ac₅GlutolNAc managed to significantly enhance AAM labeling efficiency without compromising the viability of DC2.4 cells, and was thus used for subsequent studies (Figures 2B and 2D). Confocal imaging also confirmed that Ac₅GlutolNAc could improve AAM labeling efficiency of DC2.4 cells (Figures 2E and 2F).

Figure 2. RENBP and GNE inhibition can enhance AAM labeling efficiency of DC2.4 cells, and RENBP inhibition induces minimal cytotoxicity.

(A) Chemical structures of three GNE inhibitors (I1, I2, and I3) and one RENBP inhibitor (Ac₅GlutolNAc).

(B–D) DC2.4 cells were incubated with AAM in the presence or absence of varied concentrations of I1, I2, I3, and Ac5GlutolNAc, respectively, for 24 h, followed by staining with DBCO-Cy5 for 30 min to detect cell-surface azido groups. (B) Mean Cy5 fluorescence intensity of DC2.4 cells for different groups.

(C) Mean Cy5 fluorescence intensity ratio of DC2.4 cells treated with AAM + inhibitors versus cells treated with AAM alone.

(D) Cell viability of DC2.4 cells for different treatment groups.

(E) Confocal images of DC2.4 cells treated with 25 μM AAM alone in combination with 200 μM Ac₅GlutolNAc for 24 h. Cells were stained with DBCO-Cy5, Cell Mask Green (membrane), and DAPI (nucleus).

(F) Quantification of Cy5 fluorescence intensity in (E).

All the numerical data are presented as mean ± SD (one-way ANOVA with post hoc Fisher’s LSD test was used; 0.01 < *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001).

RENBP inhibition amplifies AAM labeling efficiency in different APCs

After confirming that RENBP inhibition amplifies AAM labeling efficiency without reducing cell viability in DC2.4 cells, we proceeded to test the effect of Ac₅GlutolNAc on AAM labeling efficiency of different types of murine cells, including EG7-OVA, NIT-1, 4T1, DC2,4, RAW264.7, and A20. Cells were incubated with AAM alone or in the presence of 100 μM Ac₅GlutolNAc for 24 h, followed by the detection of cell-surface azido groups using DBCO-Cy5. All cells treated with AAM showed a higher Cy5 fluorescence intensity than cells treated with PBS, as expected (Figure 3A). However, for EG7-OVA, NIT-1, and 4T1 (all non-APCs), the co-incubation with Ac₅GlutolNAc did not increase the azido labeling efficiency (Figures 3A and 3B). In contrast, A20, RAW264.7, and DC2.4 cells, three types of APCs, showed enhanced azido labeling efficiency in the presence of Ac₅GlutolNAc (Figures 3A and 3B), demonstrating the ability of RENBP inhibitors to selectively enhance the metabolic labeling efficiency of AAM in APCs.

Figure 3. RENBP inhibition enhances AAM labeling efficiency for different types of antigen-presenting cells.

(A and B) Different types of cells (EG7-OVA, NIT-1, 4T1, DC2.4, RAW264.7, and A20) were incubated with PBS, 10 μM AAM, or 10 μM AAM +100 μM Ac₅GlutolNAc for 24 h, followed by DBCO-Cy5 staining to detect cell-surface azido groups. (A) Mean Cy5 fluorescence intensity of cells with different treatments. (B) Ratio of mean Cy5 fluorescence intensity in the AAM + Ac₅GlutolNAc group versus AAM alone group.

(C and D) DC2.4 cells or RAW264.7 cells were incubated with AAM in the presence or absence of varied concentration of Ac₅GlutolNAc for 12 h (red), 24 h (blue), and 48 h (purple), respectively, followed by DBCO-Cy5 staining to detect cell-surface azido groups. Cells treated with PBS (black) were used as the control.

(C) Mean Cy5 fluorescence intensity of DC2.4 cells with different treatments.

(D) Mean Cy5 fluorescence intensity of RAW264.7 cells with different treatments.

(E–G) Western blot analysis of RAW264.7 cell lysates. RAW264.7 cells were treated with AAM (25 μM) or both Ac₅GlutolNAc (200 μM) and AAM (25 μM) for 24 h, incubated with DBCO-Cy5 for 1 h to label surface azido groups, and lysed. The lysates were incubated with DBCO-biotin for 1 h to tag intracellular azido-tagged molecules. (E) Detection of protein bands via Cy5 fluorescence imaging, ECL imaging, and Ponceau S staining. (F) Quantification of protein band (25–250 kDa) signals from (E) using ImageJ. (G) Normalized Cy5 and ECL signals relative to Ponceau S signal from (F).

(H and I) DC2.4 or RAW264.7 cells were incubated with AAM in the presence or absence of Ac₅GlutolNAc and GlcNAc for 24 h, followed by DBCO-Cy5 staining to detect cell-surface azido groups. (H) Mean Cy5 fluorescence intensity of cells with different treatments. (I) Ratio of mean Cy5 fluorescence intensity to the AAM only group for different groups.

All the numerical data are presented as mean ± SD (one-way ANOVA with post hoc Fisher’s LSD test was used; 0.01 < *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001).

We further studied the labeling efficiency of AAM (25 μM) in the presence or absence of Ac₅GlutolNAc in DC2.4 cells with varied incubation times and Ac₅GlutolNAc concentrations. Ac₅GlutolNAc treatment consistently increased the labeling efficiency of AAM in DC2.4 cells in a concentration-dependent manner (Figures 3C and S3A). Increasing the incubation time of AAM from 12 h to 24 or 48 h did not result in an increased Cy5 fluorescence intensity (Figures 3C and S3A). Nevertheless, at all incubation times, Ac₅GlutolNAc managed to significantly improve the labeling efficiency of AAM (Figures 3C and S3A). We also studied whether Ac₅GlutolNAc can improve the labeling efficiency of AAM in RAW264.7 macrophages, another representative type of APCs. AAM-treated RAW264.7 cells showed a higher Cy5 fluorescence intensity than PBS-treated RAW264.7 cells, as expected (Figures 3D and S3B). Compared to AAM alone that induced 8.2% of Cy5⁺ RAW264.7 cells, AAM plus Ac₅GlutolNAc (200 μM) resulted in 47.0% of Cy5⁺ RAW264.7 cells after 12-h incubation (Figures 3D and S3B), showing dramatically improved azido labeling efficiency. With 24-h or 48-h incubation, the percentages of Cy5⁺ RAW264.7 cells decreased compared to 12-h incubation, but Ac5GlutolNAc treatment consistently improved the metabolic labeling efficiency of AAM in RAW264.7 cells (Figures 3D and S3B). These experiments demonstrated that RENBP inhibitors can significantly increase AAM labeling efficiency in a concentration-dependent manner for both DC2.4 and RAW264.7 APCs.

To better understand the effect of RENBP inhibition on AAM labeling, we incubated RAW264.7 cells with AAM + Ac₅GlutolNAc or AAM alone for 24 h, stained cells with DBCO-Cy5 (for cell surface azido-proteins), lysed cells, stained cell lysates with DBCO-Biotin (for intracellular azido-proteins), and performed western blot. As a result, the amounts of intracellular azido-tagged proteins slightly decreased in the AAM + Ac₅GlutolNAc group in comparison with AAM only group, in contrast to the increased level of cell-surface azido-tagged proteins (Figures 3E–3G). This result indicates that RENBP can convert ManNAz (hydrolyzed product of AAM) to GlcNAz for O-glycan labeling and its inhibition drives the metabolic incorporation of ManNAz into sialic acid biosynthesis that ends up being expressed on the surface. There was also a possibility that RENBP inhibition reduces the conversion of natural GlcNAc to ManNAc, allowing more unnatural ManNAz to enter sialic acid biosynthesis. To test this, we studied whether the addition of GlcNAc would affect AAM labeling efficiency in DC2.4 and RAW264.7 cells. Cells were incubated with AAM in the presence or absence of Ac₅GlutolNAc, GlcNAc, or both for 24 h. In the absence of Ac₅GlutolNAc, a high concentration of GlcNAc (1–5 mM) only slightly decreased the labeling efficiency of AAM in DC2.4 cells (Figures 3H and 3I). However, GlcNAc failed to abrogate the AAM labeling enhancement effect of Ac₅GlutolNAc in all tests (Figures 3H and 3I). These results indicate that the effect of RENBP inhibition lies more on the reduced catabolism of ManNAz instead of the decreased flux of ManNAz into sialic acid biosynthesis. To further confirm the presence of sialic acids on the cell surface, we treated DC2.4 cells with 500 U/mL neuraminidase, followed by DBCO-Cy5 staining and flow cytometry analysis. As a result, neuraminidase treatment decreased the levels of cell-surface azido groups in both AAM and AAM + Ac₅GlutolNAc groups, confirming the incorporation of azido groups into cell-surface sialylated glycans (Figure S3C). Additionally, we have utilized DMB derivatization followed by high-performance liquid chromatography (HPLC) analysis to quantify Neu5Ac expression levels in DC2.4 cells. Both Neu5Ac and Neu5Gc were detected, with the majority being Neu5Ac (Figures S3D–S3G). Compared with AAM only group, Ac₅GlutolNAc + AAM led to an increase of Neu5Ac levels (Figure S3H), consistent with our hypothesis that Ac₅GlutolNAc inhibits the flux from ManNAc toward GlcNAc and facilitates subsequent sialic acid biosynthesis.

RENBP inhibition specifically amplifies AAM labeling efficiency in primary dendritic cells

We next studied whether RENBP inhibition by Ac₅GlutolNAc can improve the metabolic labeling efficiency of AAM in bone marrow-derived dendritic cells (BMDCs). BMDCs were incubated with different concentrations of AAM alone or in the presence of 200 μM Ac₅GlutolNAc for 24 h, followed by the detection of cell-surface azido groups using DBCO-Cy5. As a result, Ac₅GlutolNAc was able to enhance the azido density on the surface of BMDCs at all tested AAM concentrations (Figures 4A–4C), without reducing the viability of BMDCs (Figure S4A). It is noteworthy that Ac₅GlutolNAc treatment allows for the use of a much lower concentration of AAM to achieve a similar azido labeling efficiency. For example, 100 μM AAM alone resulted in 20% Cy5⁺ BMDCs, while Ac₅GlutolNAc plus 25 μM AAM induced a similar labeling efficiency (~20% Cy5⁺ BMDCs). To determine if Ac₅GlutolNAc-induced enhancement in metabolic labeling efficiency was specific to ManNAz metabolism, we also tested its impact on the labeling efficiency of tetraacetyl-N-azidoacetylgalactosamine (AAG), a GalNAz analog. BMDCs were incubated with AAG alone or in the presence of 200 μM Ac₅GlutolNAc for 24 h. As a result, the addition of Ac₅GlutolNAc showed no effect on the labeling efficiency of AAG in BMDCs (Figures 4D, S4B, and S4C) and the viability of BMDCs (Figure S4D), confirming that Ac₅GlutolNAc interferes with the sialic acid synthetic pathways but not the GalNAc glycan synthesis pathways.

Figure 4. RENBP inhibition amplifies the labeling efficacy of AAM instead of AAG in primary BMDCs.

BMDCs were incubated with varying concentrations of (A–C) AAM or (D) AAG in the presence or absence of 200 μM Ac₅GlutolNAc for 24 h, followed by staining with DBCO-Cy5 to detect cell-surface azido groups.

(A) Percentages of Cy5⁺ BMDCs at varied AAG concentrations in the presence or absence of 200 μM Ac₅GlutolNAc.

(B) Mean Cy5 fluorescence intensity of BMDCs at varied AAM concentrations in the presence or absence of 200 μM Ac₅GlutolNAc.

(C) Cy5 fluorescence intensity ratios of Ac₅GlutolNAc-treated BMDCs versus control BMDCs across different AAM concentrations.

(D) Mean Cy5 fluorescence intensity of BMDCs at varied AAG concentrations in the presence or absence of 200 μM Ac₅GlutolNAc.

All the numerical data are presented as mean ± SD (unpaired t tests were used; 0.01 < *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001).

RENBP inhibition amplifies AAM labeling efficiency of APCs in vivo

We next investigated whether Ac₅GlutolNAc treatment can improve AAM labeling efficiency of APCs in vivo. C57BL/6 mice (8 weeks, n = 4) were intraperitoneally injected with PBS, AAM, or AAM + Ac₅GlutolNAc twice a day for three days. Blood was then extracted at different times and different types of immune cells, including CD45⁺CD3⁺CD19⁻ T cells, CD45⁺ CD3⁻CD19⁺ B cells, FSC^HISSC^HICD45⁺CD3⁻CD19⁻ granulocytes, and other white blood cells (WBCs) (FSC^lowSSC^lowCD45⁺CD3⁻CD19⁻) were analyzed using flow cytometry after staining with DBCO-Cy5 (Figures 5A and S5A). At 12 h or 2.5 days, granulocytes in AAM + Ac₅GlutolNAc group showed a significantly higher Cy5 fluorescence intensity than granulocytes in AAM alone group (Figures 5B and 5C). At 4.5 or 7.5 days, granulocytes in AAM + Ac₅GlutolNAc group and AAM alone group showed negligible differences in Cy5 fluorescence intensity (Figures 5B and 5C). Strikingly, B cells in AAM + Ac₅GlutolNAc group showed a significantly higher Cy5 fluorescence intensity than B cells in AAM alone group even after 7.5 days (Figures 5D and 5E). For T cells, AAM + Ac₅GlutolNAc treatment managed to maintain a higher azido density than AAM alone for up to 2.5 days (Figures 5F and 5G). For other WBCs, AAM + Ac₅GlutolNAc group showed a higher Cy5 fluorescence intensity than cells in AAM alone group at 12 h or 2.5 days (Figures 5H and 5I). However, at 4.5 or 7.5 days, negligible differences in Cy5 fluorescence intensity were detected (Figures 5H and 5I). Overall, Ac₅GlutolNAc treatment induced the most profound effect in metabolic labeling of circulating B cells even after 7.5 days post the injections of AAM and Ac₅GlutolNAc (Figure 5J). In contrast, granulocytes, T cells, and other WBCs showed rapid decay in the density of cell-surface azido groups over time (Figure 5J). These experiments demonstrated that RENBP inhibition via Ac₅GlutolNAc treatment can also increase the labeling efficiency of APCs such as B cells in vivo. It is noteworthy that the injection of AAM alone or in combination with Ac₅GlutolNAc did not result in changes in the populations of granulocytes (Figure S5B), B cells (Figure S5C), T cells (Figure S5D), and other WBCs (Figure S5E) in the bloodstream. AAM and Ac₅GlutolNAc treatments also did not induce any noticeable changes in the mouse body weight or overall counts of WBCs (Figures S5F and S5G), indicating their potential benign safety profile.

Figure 5. RENBP inhibition amplifies AAM-mediated metabolic labeling of B cells and other immune cells in vivo.

(A) Time frame of in vivo metabolic labeling study. C57BL/6 mice were intraperitoneally injected with PBS, AAM, or AAM + Ac₅GlutolNAc twice a day for three days. Blood was drawn at different times, isolated for WBCs, stained with DBCO-Cy5 and surface markers of different immune cells, and analyzed via flow cytometry.

(B) Mean Cy5 fluorescence intensity of FSC^HISSC^HICD45⁺CD3⁻CD19⁻ granulocytes at different times.

(C) Cy5 fluorescence intensity ratio (versus PBS group) for granulocytes in the AAM group or AAM + Ac₅GlutolNAc group at different times.

(D) Mean Cy5 fluorescence intensity of CD45⁺CD3⁻CD19⁺ B cells at different times.

(E) Cy5 fluorescence intensity ratio (versus PBS group) for B cells in the AAM group or AAM + Ac₅GlutolNAc group at different times.

(F) Mean Cy5 fluorescence intensity of CD45⁺CD3⁺CD19⁻ T cells at different times.

(G) Cy5 fluorescence intensity ratio (versus PBS group) for T cells in the AAM group or AAM + Ac₅GlutolNAc group at different times.

(H) Mean Cy5 fluorescence intensity of other WBCs (FSC^lowSSC^lowCD45⁺CD3⁻CD19⁻) at different times.

(I) Cy5 fluorescence intensity ratio (versus PBS group) for other WBCs in the AAM group or AAM + Ac₅GlutolNAc group at different times.

(J) Cy5 fluorescence intensity ratio (AAM + Ac₅GlutolNAc group/AAM group) for granulocytes, B cells, T cells, and other WBCs at different times.

All the numerical data are presented as mean ± SD (one-way ANOVA with post hoc Fisher’s LSD test was used for (B, D, F, H, and J) and unpaired t test was used for (C, E, G, and I); 0.01 < *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001).

DISCUSSION

Improving the metabolic glycan labeling efficiency of APCs could open up new opportunities for developing robust vaccines and immunotherapies.^25,42 Our study investigates two enzymes involved in ManNAz analog metabolism, i.e., RENBP and GNE, and demonstrates that the inhibition of RENBP significantly enhances the labeling efficacy of AAM in APCs while maintaining a higher cell viability than the inhibition of GNE. The increased labeling efficiency observed with RENBP inhibition could partially be attributed to the reduced catabolism of AAM/ManNAz, potentially resulting in elevated intracellular concentrations of ManNAz for glycan synthesis.^38,39 In this study, we simultaneously added AAM and Ac₅GlutolNAc for the metabolic labeling of cells. However, it is possible that the different cell uptake pathways and rates of AAM and Ac₅GlutolNAc could limit the achievable enhancement in AAM labeling efficiency, which could be further improved via the design of AAM-Ac₅GlutolNAc conjugates.

Ac₅GlutolNAc can enhance the labeling efficiency of AAM toward various types of APCs (dendritic cells, macrophages, and B cells) instead of non-APCs (cancer cells, beta cells, and T cells) in vitro. This enhancement in AAM labeling efficiency aligns with the upregulated RENBP expression levels in APCs. We also show that Ac₅GlutolNAc selectively enhances the metabolic labeling efficiency of AAM over AAG, demonstrating the specificity of RENBP toward the catabolism of ManNAc analogs. We envision similar strategies can be applied to other types of monosaccharide analogs, i.e., the inhibition of enzymes specific to the catabolism of GalNAc or GlcNAc analogs could improve the labeling efficiency of corresponding monosaccharide analogs.

Achieving long-term labeling of specific cell populations such as B cells in vivo provides opportunities for immune cell-targeted immunomodulation and immunotherapy development.^43–45 Our study provides a facile approach to metabolically tagging circulating B cells with azido groups via the co-administration of AAM and Ac₅GlutolNAc. Ac₅GlutolNAc is able to increase the in vivo azido labeling efficiency of B cells, and prolong the retention of azido tags on the surface of circulating B cells. This is particularly valuable for studies requiring extended tracking and modulation of B cells. We envision the azido-labeled B cells would enable covalent conjugation of DBCO-bearing immunomodulatory agents or any molecules of interest. For example, in vivo metabolic labeling of B cells with azido groups enables subsequent conjugation of DBCO-functionalized antigens and adjuvants for enhanced humoral responses. Additionally, this labeling and targeting approach could be applied to the treatment of B cell lymphoma by enabling the conjugation of DBCO-drugs to B cells.

To conclude, we report a facile approach to significantly improving the AAM-mediated metabolic glycan labeling efficiency of APCs, including dendritic cells, macrophages, and B cells, via the inhibition of RENBP. Ac₅GlutolNAc, an RENBP inhibitor, is able to selectively enhance the labeling efficiency of AAM in APCs over non-APCs in a concentration-dependent manner. We further show that the co-administration of AAM and Ac₅GlutolNAc can result in durable azido labeling of B cells in vivo. Our study yields a facile approach to enhancing the metabolic glycan labeling efficiency of APCs in vitro and in vivo for the development of APC-targeted immunotherapies and immunomodulation technologies, and provides new insights for the intervention of metabolic pathways in APCs.

Limitations of the study

Optimizing in vivo labeling protocols to minimize undesirable immune responses and off-target side effects is critical for the clinical translation of the metabolic glycan labeling technology. The pharmacokinetics, biodistribution, and long-term safety profiles of Ac₅GlutolNAc remain to be fully characterized. While we explored the co-administration of AAM and Ac₅GlutolNAc to enhance AAM labeling in vivo, the administration timing and doses of the two compounds should be further optimized to achieve efficient labeling of APCs. Future work should also investigate the development of biocompatible delivery systems, such as nanoparticles, hydrogels, or cell-targeted carriers that enable controlled and sustained release of RENBP inhibitors, thereby reducing systemic exposure and off-target toxicity. Additionally, combinatorial strategies that integrate RENBP inhibition with other metabolic interventions, such as agents that modulate the unnatural sugar uptake rate or other glycosylation pathways, may further improve the specificity and efficiency of APC labeling. Together, these approaches could help overcome current limitations in unnatural sugar uptake, distribution, and intracellular processing, advancing the in vivo translation of this platform for targeted modulation of APCs and systemic immune responses.

RESOURCE AVAILABILITY

Lead contact

Requests for further information and resources should be directed to and will be fulfilled by the lead contact, Hua Wang (huawang3@illinois.edu).

Materials availability

Cell lines and chemical compounds used and generated in this study will be provided in reasonable amounts by the lead contact on request, but we may require payment and/or a completed materials transfer agreement for use.

Data and code availability

• All data reported in this paper will be shared by the lead contact upon request.

• This paper does not report original code.

• Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

STAR★METHODS

EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS

Animals

Female C57BL/6 mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). Feed and water were available ad libitum. Artificial light was provided in a 12 h/12 h cycle. All procedures involving animals were done in compliance with the National Institutes of Health and Institutional guidelines with approval from the Institutional Animal Care and Use Committee at the University of Illinois at Urbana-Champaign.

Cell lines

The RAW264.7 (male), A549 (male), A20 (female), EG7-OVA (male), 4T1 (female), NIT-1 (male), and Jurkat (male) cell lines were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA), and the DC2.4 cell line was purchased from Sigma-Aldrich. The sex of DC2.4 cells was not specified by the supplier. Cell lines were tested for mycoplasma and verified using short-tandem repeat (STR) profiling. RAW264.7, A549, A20, EG7-OVA, 4T1, NIT-1, and Jurkat cells were cultured in media recommended by ATCC for each respective line at 37°C in a humidified incubator with 5% CO₂. DC2.4 cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/mL penicillin-streptomycin.

General culture conditions for BMDCs

Bone marrow cells were isolated from the femurs and tibiae of female C57BL/6 mice and seeded at 2 × 10⁶ cells per 100 mm non-tissue culture-treated petri dish in 10 mL R10 medium (RPMI-1640 with 10% FBS, 1% penicillin/streptomycin, and 50 μM 2-mercaptoethanol) supplemented with 20 ng/mL GM-CSF. On day 3, 10 mL of R10 medium containing 20 ng/mL GM-CSF was added. On days 6 and 8, half of the medium was replaced with R10 medium containing 20 ng/mL GM-CSF. Cells were harvested on days 8–10 for downstream assays.

METHOD DETAILS

Instruments

Analytical high-performance liquid chromatography (HPLC) was performed on a Nexera 40 Series HPLC System (Shimadzu) equipped with an Eclipse XDB-C18 column (5 μm, 4.6 mm ID × 100 mm), unless otherwise specified in the method. Preparative HPLC was performed on a CombiFlash^®Rf system (Teledyne ISCO, Lincoln, NE, USA) equipped with a RediSep^®Rf HP C18 column (Teledyne ISCO, 30 g, Lincoln, NE, USA). Lyophilization was conducted in a Labconco FreeZone lyophilizer (Kansas City, MO, USA). Nuclear Magnetic Resonance (NMR) spectra were recorded on a Varian U500 (500 MHz) or VXR500 (500 MHz), or a Bruker Carver B500 (500 MHz) spectrometer. Electrospray ionization (ESI) mass spectra were obtained from a Waters ZMD Quadrupole Instrument (Waters, Milford, MA, USA). Confocal laser scanning microscopy (CLSM) images were taken by using a Zeiss LSM 700 or LSM 880 Confocal Microscope (Carl Zeiss, Thornwood, NY, USA). Western blotting protein bands were imaged on an ImageQuant LAS 4000 gel imaging system (GE, Pittsburgh, PA, USA). Flow cytometry analysis was conducted on an Attune NxT flow cytometer and analyzed on FlowJo V10.8.1.

RENBP and GNE mRNA expression analysis

Single-cell mRNA expression data and cell type classifications were obtained from The Human Protein Atlas (for human datasets) and the EMBL-EBI Single Cell Expression Atlas (for mouse datasets). Cell types with high mRNA expression levels (nTPM > 5) were selected for visualization in the bar graph with R. For the pie chart, the proportion of mRNA expression for each cell group type was calculated relative to the total mRNA expression across all cell types. The pie chart was created using R.

Synthesis of Ac₅GlutolNAc

N-Acetyl-D-glucosaminitol (GlutolNAc, 1.0 e.q.) was dissolved in anhydrous DMF, followed by the addition of pyridine (1.1 e.q.) and acetic anhydride (1.1 e.q.). Then a small amount of DMAP was added to the solution and reacted at room temperature overnight. The DMF was removed by rotary evaporation and the crude product was purified by preparative HPLC using a linear gradient from 30% methanol in 0.1% trifluoroacetic acid (TFA)/water to 100% methanol over 10 minutes at a flow rate of 30 mL/min. Elution was monitored at 230 nm, and fractions containing Ac₅GlutolNAc were collected. After confirming the purity of Ac5GlutolNAc with analytical HPLC, the solvent was removed by rotary evaporation and lyophilization to afford a colorless oil-like liquid product. 1H NMR of Ac₅GlutolNAc (CD₃COCD₃, 500 MHz): δ 5.44 – 5.35 (m, 2H), 5.11 (td, J = 6.0, 4.0 Hz, 1H), 4.58 – 4.52 (m, 1H), 4.27 (dd, J = 12.2, 4.0 Hz, 1H), 4.11 (ddd, J = 11.3, 9.0, 6.0 Hz, 2H), 3.93 (dd, J = 11.3, 6.1 Hz, 1H), 2.08 (d, J = 0.6 Hz, 3H), 2.01 (s, 6H), 1.99 (d, J = 0.6 Hz, 3H), 1.97 (d, J = 0.6 Hz, 3H), 1.95 (d, J = 0.7 Hz, 3H). High resolution ESI MS (m/z): Calculated [M+H]⁺ mass: 434.1662; found: 434.1664.

Synthesis of AAM and AAG

Mannosamine hydrochloride or galactosamine hydrochloride (1.0 e.q.) was dissolved in anhydrous methanol and cooled in the ice bath, followed by the addition of sodium methoxide in methanol (1.0 e.q.). Chloroacetic anhydride (1.05 e.q.) and triethylamine (1.0 e.q.) were then added. The reaction mixture was stirred at room temperature overnight. Upon removal of the solvent, the crude product was re-dissolved in water. Sodium azide (4.0 e.q.) was then added and the reaction mixture was stirred at 60°C overnight. After removing the solvent, the crude product was re-dissolved in pyridine, followed by the addition of acetic anhydride and 4-dimethylaminopyridine. After 24 h, the solvent was removed and the crude product was purified via silica column chromatography using ethyl acetate and hexane as the eluents. 1H NMR of AAM (CDCl3, 500 MHz): δ (ppm) 6.66&6.60 (d, J= 9.0 Hz, 1H, C(O)NHCH), 6.04&6.04 (d, 1H, J=1.9 Hz, NHCHCHO), 5.32–5.35&5.04–5.07 (dd, J=10.2, 4.2 Hz, 1H, CH2CHCHCH), 5.22&5.16 (t, J=9.9 Hz, 1H, CH2CHCHCH), 4.60–4.63&4.71–4.74 (m, 1H, NHCHCHO), 4.10–4.27 (m, 2H, CH2CHCHCH), 4.07 (m, 2H, C(O)CH2N3), 3.80–4.04 (m, 1H, CH2CHCHCH), 2.00–2.18 (s, 12H, CH3C(O)). 13C NMR (CDCl3, 500 MHz): δ (ppm) 170.7, 170.4, 170.3, 169.8, 168.6, 168.3, 167.5, 166.9, 91.5, 90.5, 73.6, 71.7, 70.5, 69.1, 65.3, 65.1, 62.0, 61.9, 52.8, 52.6, 49.9, 49.5, 21.1, 21.0, 21.0, 20.9, 20.9, 20.9, 20.8. ESI MS (m/z): calculated for C16H22N4O10Na [M+Na]+ 453.1234, found 453.1230. 1H NMR of AAG (CDCl3, 500 MHz): δ (ppm) 6.29 (d, J = 9.2 Hz, 1H, NHCHCHO), 6.25 (d, J = 3.7 Hz, 1H, C(O)NHCH), 5.50 – 5.44 (m, 1H, NHCHCHO), 5.27 (dd, J = 11.4, 3.2 Hz, 1H, CH2CHCHCH), 4.72 (ddd, J = 12.3, 9.1, 3.8 Hz, 1H, CH2CHCHCH), 4.28 (t, J = 6.8 Hz, 1H, CH2CHCHCH), 4.11 (qd, J = 11.2, 6.6 Hz, 2H, CH2CHCHCH), 3.97 (s, 2H, C(O)CH2N3), 2.21 (s, 3H, CH3C(O)), 2.19 (s, 3H, CH3C (O)), 2.05 (d, J = 2.3 Hz, 6H, CH3C(O)). ESI MS (m/z): calculated for C16H22N4O10Na [M+Na]+ 453.1234, found 453.1229.

In vitro labeling of DC2.4 cells

DC2.4 cells were seeded in 96-well plates in DMEM medium at 50,000 cells per well. AAM (25 μM) and Ac₅GlutolNAc or different GNE inhibitors of different concentrations were added to cells. After 24 h, DC2.4 cells were lifted by trypsin and washed twice with FACS buffer (1% FBS in PBS, 500 g, 5 min), followed by incubation with DBCO-Cy5 (0.01 mg/mL in FACS buffer) at room temperature for 30 min. Then cells were washed three times and stained with fixable viability dye for 10 min at room temperature. DC2.4 cells were then fixed with 0.4% PFA and stored at 4°C for flow cytometry.

In vitro labeling of different cell lines in the presence of Ac₅GlutolNAc

Cells were seeded into 96-well plates at 10 k cells per well. Then 10 μM AAM or 10 μM AAM+100 μM Ac₅GlutolNAc were added to each well and further incubated for 24 h. Cells were then harvested either by pipetting or trypsinization and washed with FCAS buffer, followed by incubation with DBCO-Cy5 (0.01 mg/mL in FACS buffer) at room temperature for 30 min. Cells were washed three times, fixed with 0.4% PFA, and stored at 4°C for flow cytometry.

Confocal imaging of DC2.4 cells

DC2.4 cells were seeded in 6-well plates with coverslips at a density of 100,000 cells per well. AAM (25 μM) or AAM (25 μM) + Ac₅GlutolNAc (200μM) were added and incubated with cells for 24 h. After the incubation, cells were washed with FACS buffer (1% FBS in PBS) three times, followed by incubation with DBCO-Cy5 (0.01 mg/mL in FACS buffer) at room temperature for 30 min. Then cells were washed three times and stained with cell mask green and DAPI according to the manual protocol. After washing, cells were fixed with 4% PFA for 10 min at room temperature. Then the coverslips bearing the cells were transferred to microscope slides with the addition of ProLong^™ Gold, and stored at 4°C prior to imaging.

Ac₅GlutolNAc inhibition study for DC2.4 and Raw264.7 cells

DC2.4 or Raw264.7 cells were seeded in 96-well plates in DMEM medium at 50,000 per well density. AAM (25 μM) and Ac₅GlutolNAc (0, 20, 40, 100, or 200 μM) were added to cells at different times. At the endpoint, cells were lifted by trypsin and washed twice with FACS buffer (1% FBS in PBS, 500 g, 5 min), followed by incubation with DBCO-Cy5 (0.01 mg/mL in FACS buffer) at room temperature for 30 min. Then cells were washed three times and fixed with 0.4% PFA before running flow cytometry.

Neuraminidase treatment of DC2.4 cells

DC2.4 cells were cultured in the presence of PBS, 25 μM AAM, or 25 μM AAM + 200 μM Ac₅GlutolNAc for 24 h. Cells were then detached using 0.25% trypsin and washed with DPBS. Following detachment, cells were stained with fixable Live/Dead eFluor^™ 780 dye and washed with PBS. Cells were then fixed with 1% paraformaldehyde (PFA) in PBS at 4°C for 1 h. After fixation, cells were washed three times with DPBS and incubated with 500 U/mL neuraminidase at 37°C for 1 h. Following neuraminidase treatment, cells were washed three times with DPBS and incubated with 10 μg/mL DBCO-Cy5 in FACS buffer for 1 h. Finally, cells were washed three times with FACS buffer prior to flow cytometry analysis.

HPLC analysis of sialic acids

To analyze cell surface sialic acids, 1,2-diamino-4,5-methylenedioxybenzene dihydrochloride (DMB) derivatization was performed. Briefly, 2 million cells were collected and washed twice with DPBS. The cell pellet was resuspended in 50 μL of 2 M acetic acid and incubated at 80°C for 1.5 h to hydrolyze and release sialic acids. After cooling, the sample was centrifuged at 21,000 × g at 4°C for 10 min, and the supernatant was transferred to a new tube. An equal volume of DMB labeling solution (14 mM DMB, 36 mM sodium hydrosulfite, 1.5 M β-mercaptoethanol, and 0.8 M acetic acid) was added to the supernatant. The mixture was incubated at 50°C for 3 h in the dark to allow DMB derivatization. The reaction was quenched by dilution with water. 5 μL of the quenched sample was injected into a Kinetex 2.6 μm Biphenyl 100 Å LC column (100 × 2.1 mm). Separation was performed using a linear gradient from 20% to 80% methanol over 7 minutes at 40°C, and fluorescence detection was carried out with excitation at 373 nm and emission at 448 nm.

In vitro labeling of BMDCs

BMDCs were differentiated from bone marrow cells following a previously reported protocol. Briefly, signal bone marrow cell suspension was extracted for C57BL/6 mice and were cultured in the presence of 20 ng/mL GM-CSF for 7 days and then cultured in the presence of 10 ng/mL GM-CSF for following days. Day-6 BMDCs were seeded in 96-well plates with 50k cells per well. AAM or AAG of different concentrations and Ac₅GlutolNAc were added to each well and cultured for 24 h. The medium was removed and washed with FACS buffer, following by incubating with DBCO-Cy5 (0.01 mg/mL in FACS buffer) at room temperature for 30 min. Then cells were washed three times and fixed with 0.4% PFA before running flow cytometry.

In vivo labeling study

C57BL/6 mice (8 weeks, n=4, female mice were used for easiness of housing and care) were intraperitoneally injected with AAM (200 mg/kg) alone or with AAM (200 mg/kg) and Ac₅GlutolNAc (200 mg/kg) twice a day for 3 days. After the final injection, the blood was extracted from mice to FCAS buffer at selected time points and stained with DBCO-Cy5 (0.01 mg/mL) at room temperature for an hour. Cells were washed once and further incubated with antibody cocktail (FITC-conjugated anti-CD45 FITC, PE-conjugated anti-CD19, PE-Cy5.5-conjugated anti-CD3) at room temperature for 10 min in. Cells were then stained with fixable viability, washed twice with FACS buffer and fixed with 0.4% PFA before running flow cytometry. The RBCs were removed prior to fixation by adding RBC lysis buffer and incubate for 5 min at room temperature.

Flow cytometry analysis of RENBP in cells

The intracellular RENBP detection was achieved by using BD staining kit. Briefly, cells (A20, Jurkat, DC2.4, A549 or BMDCs) were harvested either by pipetting or trypsinization and washed with FCAS buffer once. Then cells were stained with fixable viability dye according to manual protocol. Cells were fixed and permeabilized by adding fixation/permeabilization buffer for 20 min at 4°C. Cells were washed twice and stained with PE-conjugated anti-REBNP for 30 min at 4°C. After washing, cells were analyzed via flow cytometry.

Western blot analysis of RENBP

Cells were harvested either by pipetting or trypsinization and washed with cold PBS once. Then the cell pellet was resuspended in cell lysis buffer (10mM Tris-HCl, 2mM EDTA, 1% SDS, pH=8.0) with protease inhibitor (Pierce^™ Protease Inhibitor Tablets, EDTA-free) and sonicated to fully release the protein. Then the solution was placed at 4°C for 1 h to fully solubilize the protein. Then the sample was centrifuged at 21,000 g for 30 min at 4°C and the supernatant was used for western blot analysis. Protein concentration was determined using the BCA assay. A total of 30 μg of protein per sample was resolved on a 4–20% precast SDS-PAGE gel at 150 V for 40 min. Proteins were then transferred onto a nitrocellulose (NC) membrane at 100 V for 2 h. Membranes were blocked with 5% BSA in PBST (PBS with 0.1% Tween-20) for 1 h at room temperature. Following blocking, membranes were incubated with primary antibodies against RENBP (1:1000), and GAPDH (1:20,000) or β-tubulin (1:10,000) for 1 h at room temperature. After three washes with PBST, membranes were incubated with HRP-conjugated goat anti-rabbit secondary antibody (1:10,000) for 1 h at room temperature. Enhanced chemiluminescence (ECL) imaging was performed using the ImageQuant 800 imaging system.

Western blot analysis of membrane and cytosolic azido-labeled proteins

RAW264.7 cells were cultured with AAM (25 μM) or a combination of Ac₅GlutolNAc (200 μM) and AAM (25 μM) for 24 h. Cells were then detached using trypsin, washed twice with FACS buffer (1% FBS in PBS, 500 g, 5 min), and incubated with DBCO-Cy5 (0.01 mg/mL in FACS buffer) at room temperature for 30 min to label cell-surface azides. After two additional washes with cold PBS, cell were resuspended in lysis buffer (10 mM Tris-HCl, 2 mM EDTA, 1% SDS, pH 8.0) supplemented with EDTA-free protease inhibitors (Pierce^™ Protease Inhibitor Tablets) and sonicated to fully release intracellular proteins. Lysates were cleared by centrifugation at 21,000 g for 30 min at 4°C, and the supernatant was reacted with 50 μM DBCO-PEG4-biotin for 2 h at 4°C to label intracellular azido-modified proteins. Samples were analyzed by SDS-PAGE and western blotting using streptavidin-HRP for detection.

QUANTIFICATION AND STATISTICAL ANALYSIS

Statistical analyses

Statistical analysis was performed using GraphPad Prism v6 and v8. Sample variance was tested using the F test. For samples with equal variance, the significance between the groups was analyzed by a two-tailed student’s t test. For samples with unequal variance, a two-tailed Welch’s t-test was performed. For multiple comparisons, a one-way analysis of variance (ANOVA) with post hoc Fisher’s LSD test was used. The results were deemed significant at 0.01 < *P ≤ 0.05, 0.001 < **P ≤ 0.01, highly significant at 0.0001 < ***P ≤ 0.001, and extremely significant at ****P ≤ 0.0001.

Statistics & reproducibility

The sample sizes were determined empirically. In general, in vitro studies involve n=3–6 and in vivo labeling studies involve n=4. No data were excluded from the analyses. Randomization was used for animal studies.

Supplementary Material

SUPPLEMENTAL INFORMATION

Supplemental information can be found online at https://doi.org/10.1016/j.chembiol.2025.07.001.

KEY RESOURCES TABLE.

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Rat anti-mouse CD45 FITC	Invitrogen	Catalog # 11-0451-82
Rat anti-mouse CD19 PE	Invitrogen	Catalog # 12-0193-82
Armenian hamster anti-mouse CD3e PE-Cy5.5	Invitrogen	Catalog # 35-0031-82
Rabbit anti-Human, Mouse Renin Binding Protein (RENBP) Polyclonal Antibody, PE-Linked	MyBioSource	Catalog # MBS2068052
β-Tubulin (9F3) Rabbit mAb	Cell Signaling Technology	Catalog # 2128
GAPDH (D16H11) XP® Rabbit mAb	Cell Signaling Technology	Catalog # 5174
Chemicals, peptides, and recombinant proteins
GNE Inhibitor (I1)	MCULE	P-520049529
GNE Inhibitor (I2)	MCULE	P-24026011
GNE Inhibitor (I3)	MCULE	P-856021082
RENBP inhibitor (Ac5GlutoINAc)	This study	NA
D-Mannosamine hydrochloride	Sigma-Aldrich	Catalog # M4670
N-Acetyl-D-glucosaminitol	Abcam	Catalog # ab275052
D-galactosamine hydrochloride	Sigma-Aldrich	Catalog # G0500
DBCO-Cy5	Sigma-Aldrich	Catalog # 777374
DAPI	Sigma-Aldrich	Catalog # D9542
Sodium azide	Sigma-Aldrich	Catalog # S2002
Chloroacetic anhydride	Sigma-Aldrich	Catalog # 215163
Acetic anhydride	Sigma-Aldrich	Catalog # 539996
CellMask™ Plasma Membrane Stains	Invitrogen	Catalog # C37608
LIVE/DEAD™ Fixable Near IR (780) Viability	Invitrogen	Catalog # L34992
Critical commercial assays
SuperSignal™ West Atto Ultimate Sensitivity Substrate	Thermo Scientific	Catalog # A38554
Ponceau S Staining Solution	Thermo Scientific	Catalog # 40000279
Experimental models: Cell lines
RAW264.7	ATCC	ATCC # TIB-71
DC2.4	Sigma Aldrich	Catalog # SCC142M
A549	ATCC	ATCC # CCL-185
A20	ATCC	ATCC # TIB-208
EG7-OVA	ATCC	ATCC # CRL-2113
4T1	ATCC	ATCC # CRL-2539
NIT-1	ATCC	ATCC # CRL-2055
Jukart	ATCC	ATCC# TIB-152
Experimental models: Organisms/strains
Female C57BL/6	The Jackson Laboratory	Strain # 000664
Software and algorithms
GraphPad v 10.1.0 (264)	GraphPad	https://www.graphpad.com/
Rstudio	N/A	https://posit.co/download/rstudio-desktop/
FlowJo software 10.8.1	Becton Dickinson & Company	https://flowjo.com/
ImageJ 1.53t	National Institutes of Health	https://imagej.nih.gov/ij/
Zen Blue 3.6	ZEISS	https://www.zeiss.com/microscopy/us/products/software/zeiss-zen.html
ChemDraw Ultra 14.0	PerkinElmer	N/A
Mnova NMR version 15.0.1	Mestrelab Research	https://mestrelab.com/download
Human single-cell RENBP expression	Human Protein Atlas	https://www.proteinatlas.org/humanproteome/single+cell/single+cell+type/data#datasets
Mouse single-cell RENBP expression	EMBL-EBI	https://www.ebi.ac.uk/gxa/genes/ensmusg00000031387?bs=%7B%22mus%20musculus%22%3A%5B%22CELL_TYPE%22%5D%7D&ds=%7B%22kingdom%22%3A%5B%22animals%22%5D%7D#baseline
Other
Nexera 40 Series HPLC System	Shimadzu	https://www.ssi.shimadzu.com/products/liquid-chromatography/hplcuhplc/nexera-series/index.html
Attune NxT flow cytometer	ThermoFisher Scientific	https://www.thermofisher.com/us/en/home/life-science/cell-analysis/flow-cytometry/flow-cytometers/attune-nxt-flow-cytometer/models/nxt.html
CombiFlash®Rf system	Teledyne ISCO	SM100624-RF+LUV1R

SIGNIFICANCE.

Antigen-presenting cells (APCs) are key mediators of adaptive immunity and the targets of modulation for various immunotherapies. Efficient strategies for targeted modulation of APCs with immunomodulatory agents are extensively explored for the development of robust APC-targeted immunotherapies for cancer and other diseases. Metabolic glycoengineering of unnatural sugars is one widely used tool to label cell membranes with clickable chemical tags for subsequent conjugation of molecules via click chemistries. However, the metabolic glycan labeling efficiency of APCs is often low. In this study, we identify GlcNAc 2-epimerase (RENBP) as a key regulator of unnatural sugar metabolisms in APCs and demonstrate that the pharmacological inhibition of RENBP significantly enhances the incorporation of azido-mannosamine analogs into cell-surface glycans on APCs. This boost in metabolic labeling efficiency applies to multiple APC subtypes, including dendritic cells, macrophages, and B cells, both in vitro and in vivo. These findings establish RENBP inhibition as a simple and effective strategy to improve the metabolic glycan labeling efficiency of APCs, enabling targeted modulation and tracking of APCs and thus the development of potent APC-targeted immunotherapies.

Highlights.

• Antigen-presenting cells upregulate GlcNAc 2-epimerase (RENBP)

• RENBP inhibition enhances AAM-mediated labeling efficiency in antigen-presenting cells

• RENBP inhibition selectively enhances labeling efficiency of AAM over AAG

• RENBP inhibitors can improve AAM-mediated labeling of B cells in vivo