HEPES in Cell Culture Alters the Multi-Omics Profile Exhibited by Gaucher Disease Fibroblasts

Abstract

Lysosomal function can be affected by components in cell culture. This in turn may influence cellular metabolism and, consequently, research and diagnostics outcomes. One such component is the commonly used pH buffer 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES). HEPES specifically impacts the trafficking of the lysosomal enzyme glucocerebrosidase, which is deficient in Gaucher disease (GD). Understanding how HEPES affects cellular models of GD is essential, since glucocerebrosidase is central to diagnostic testing and the investigation of GD pathophysiology. Therefore, we examined the broader effects of HEPES on cultured fibroblasts from individuals with GD and healthy controls. We cultured dermal fibroblasts of eight adults with GD and seven healthy age- and sex-matched controls. The cells were cultured in two culture media, Ham's F10 and DMEM, both with and without HEPES. We assessed glucocerebrosidase enzyme activity and sphingolipid concentrations using a quantitative UPLC-MS/MS method. Additionally, we conducted multi-omics analyses, consisting of lipidomics, metabolomics and proteomics, to explore the broader impact of HEPES in cell culture on fibroblasts. Glucocerebrosidase activity in cell lysates increased after HEPES exposure in both GD and control fibroblasts, to an extent that may influence diagnostic outcomes. In GD fibroblasts, substrate accumulation was absent and not altered by HEPES exposure. GD fibroblasts exhibited a multi-omics profile largely overlapping with healthy controls and lacking the typical pathological features associated with GD in other cell types, such as mitochondrial dysfunction, dysregulated autophagy, disruption of intracellular calcium homeostasis, ER stress and chronic oxidative stress. In addition, the multi-omics profile was altered by HEPES, however in a non-specific manner. In conclusion, HEPES influences fibroblasts in culture, both from healthy controls and from patients with GD. Furthermore, GD fibroblasts lack a specific disease-related profile. This renders cultured fibroblasts unsuitable for studying pathophysiological processes in GD. Culturing GD fibroblasts with HEPES may compromise the reliability of diagnostics.

Keywords: Gaucher disease; HEPES; cell culture techniques; multi‐omics.

Figure 1

Effect of HEPES and culture medium on GCase activity and substrate accumulation. (A) GCase activity in different culture conditions. Each dot represents the GCase activity in the sample of one fibroblast donor (n = 4 each for controls and GD). Lines connect results of the same fibroblast donor. p‐values were calculated using a paired t‐test and adjusted according to Bonferroni, the level of significance is denoted by an asterisk or “ns” (not significant). In the comparisons at the bottom of the figure the mean GCase activity is compared between conditions. Superscript “c” denotes the samples that were set to 100% as a comparator. In the comparison investigating the effect of HEPES, each sample that was cultured without HEPES was compared to the identical sample cultured with HEPES (e.g. GD in DMEM without HEPES compared to GD in DMEM with HEPES). For the investigation of the effect of the culture medium, each sample cultured in DMEM was compared to the identical sample cultured in Ham's F10 (e.g. GD in DMEM without HEPES compared to GD in Ham's F10 without HEPES). For the investigation of the effect of GD, each GD sample was compared to the control sample that was cultured in identical conditions (e.g. GD in DMEM without HEPES compared to control in DMEM without HEPES). (B) Substrate accumulation in different culture conditions. Each dot represents the mean abundance of GlcSph (or lyso‐GlcCer) in the sample of one fibroblast donor (n = 4 each for controls and GD), lines connect results of the same donor. The comparisons at the bottom of the figure are the same as in Figure 1A and compare the fold change of the mean abundance of GlcSph in different culture conditions. p‐values were calculated using a paired t‐test for all analyses except the effect of GD which was calculated using an unpaired t‐test, and all p‐values were adjusted according to Bonferroni. DMEM, Dulbecco's modified Eagle medium; GCase, glucocerebrosidase; GD, Gaucher disease; HEPES, 4‐(2‐hydroxyethyl)‐1‐piperazineethanesulfonic acid; Lyso‐GlcSph, glucosylsphingosine.

Figure 2

Multi‐omics analysis comparing GD fibroblasts cultured with and without HEPES regarding the lipidome (A–C), metabolome (D,E) and proteome (G,H). (A, D, G) Principal component analysis of lipidomic (A), metabolomic (D) and proteomic (G) profiles in GD compared to control fibroblasts in culture with and without HEPES. Each dot represents the average of all replicates of one fibroblast donor. PCA plots of the two culture conditions were not combined into a single graph because samples from each condition were run separately on the mass spectrometer, therefore GD versus control within a culture condition can be compared in absolute terms but between culture conditions only relative comparisons, as depicted, are valid. (B, E, H) Volcano plots and Venn diagrams depicting changed lipids (B), metabolites (E) and proteins (H) in GD compared to control fibroblasts in culture with and without HEPES. Each dot represents the log2 fold change and p‐value of the mean of all GD samples compared to the mean of all control samples. Analytes with a log2 FC > |1| and a p‐value < 0.05 are annotated. p‐values were not adjusted for multiple comparisons for visualization purposes. In the Venn diagrams the number of hits with an unadjusted p‐value < 0.05 are depicted for each culture condition split into “up” (log2 fold change > 0) or “down” (log2 fold change < 0) groups. The overlapping analytes are named. (C) Changes in lipid classes in GD compared to control fibroblasts in culture with and without HEPES. Each bar represents the mean log2 fold change of the combined lipids in this class for all GD samples compared to all control samples. No lipid class is significantly enriched (p‐values not depicted). (F) Ridge plot of the changes in (polar) metabolites depicted per pathway in GD compared to control fibroblasts in culture with and without HEPES. Each dot represents the average log2 fold change of a metabolite in GD compared to controls. Metabolites are grouped by the pathway with which they are associated (Supporting Information Table S1). The same metabolite may appear in multiple pathways. No pathway is statistically significantly enriched (p‐values not depicted). GD, Gaucher disease; HEPES, 4‐(2‐hydroxyethyl)‐1‐piperazineethanesulfonic acid; PC, principle component; PPP, pentose phosphate pathway; TCA, tricarboxylic acid cycle.