Fig 1.
Overview of ManNAc analog metabolism via sialic acid metabolism and glycosylation (SAMG) gene activity.
‘High-flux’ ManNAc analogs (1,3,4-O-Bu3ManNAc, 1,3,4-O-Bu3ManNAz, 1,3,4-O-Bu3ManNAl analogs) passively diffuse across the plasma membrane after which the core natural or R-modified ManNAc (i.e., ManNAc, ManNAz, or ManNAl) is released via non-specific carboxylesterases (CES) that hydrolyze the butyrate groups. ManNAc(R) is converted to sialic acids by the kinase activity of GNE and subsequent activities of NANS, and NANP in the cytosol; in this study these metabolites constitute ‘Compartment 1’ and are measured in aggregate using the periodate resorcinol assay. Once synthesized and dephosphorylated, sialic acid enters the nucleus where it is converted to the corresponding nucleotide sugar (e.g., CMP-Neu5Ac, CMP-Sia5Az, or CMP-Sia5Al) by CMAS; these activated sialic acids are then transported into the Golgi apparatus by SLC35A1 and SLC35A3 where a subset of the 20 human sialyltransferases created sialoglycoconjugates (primarily, N- and O- linked glyocoproteins or gangliosides [i.e., sialic acid-modified glycosphingolipids]) and these compounds constitute ‘Compartment 2’ and are also measured in aggregate using the periodate resorcinol assay (as outlined in the Materials and Methods section).
Fig 2.
Comparison of SAMG mRNA levels in the MCF10A, T-47D, and MDA-MB-231 breast cell lines not treated with ManNAc analog.
(A) Pair-wise statistical analysis, * p<0.05. (B) Heat map based on qRT-PCR analysis. (C) Number of genes up- or down-regulated (or not changed) in the given comparisons across cell lines. (D) Background sialic acid levels in each cell by ‘Compartment’ (as defined in Fig 1).
Fig 3.
Transcript levels of SAMG genes and correlated sialic acid levels in ManNAc analog-treated cells.
(A) Heat maps of SAMG genes associated with Compartment 1 and Compartment 2 in each of the breast cell lines upon analog treatment as indicated. (B) Compartment 1 and 2 levels of sialic acid in analog-treated cells (100 mM for 24 h) with the same data plotted in (C) with an expanded y-axis for the MCF10A and MDA-MB-231 lines. Although not readily apparent in the heat map color scheme, several minor but statistically-significant differences in transcript levels were observed as indicated by the asterisks (* p < 0.05, ** p < 0.01, *** p < 0.001 based on the statistical analysis presented in the supporting data (Columns D-F in S2 File).
Fig 4.
Metabolic profiles of Compartment 1 in analog-supplemented cells.
(A) 3D surface plots were generated for each cell line in response to treatments with analogs; sialic acid production is shown as a function of both dose and time (AZ stands for azimuth values, El stands for elevation values, and the three plots provided for each data set are rotated to depict different vantage points). (B) The Reserve Capacity for each cell line was calculated based on the highest observed increase in Compartment 1 sialic acid levels in analog-supplemented cells compared to untreated controls.
Fig 5.
‘Compartment Ratios’ of ManNAc analog-treated cells.
The ratio of sialic acid in Compartment 1 (i.e., unconjugated sialometabolites) compared to Compartment 2 (i.e., predominantly glycoconjugate-bound sialic acids) was calculated for cells after 24 h of treatment with each analog at the indicated concentrations.
Fig 6.
Verification of metabolic incorporation of sialometabolites into Compartment 2 glycoconjugates.
(A) Relative levels of sialic acid species in each cell line in Compartment 1 and 2 after analog treatment (250 mM for 24 h) compared to untreated control cells are summarized to show the comparatively minor perturbation of Compartment 2 in all cases. (B) The presence of azido- or alkyne-modified sialic acids on the surfaces of cells was determined by labeling cells with the appropriate complementary fluorescent probe followed by FACS to quantify the overall abundance of each type of non-natural sialic acid after incorporation into sialoglyconjgates (i.e., O-linked or N-linked glycoproteins or gangliosides, as shown in Fig 1). The flow cytometry results showed very low background signal unless both the analog and labeling agent were present, indicating that robust incorporation of sialosides into sialoglycoconjugates did occur. *** indicates p < .001, error bars represent ± SEM.
Table 1.
Linear regression model with calculated parameters.