Dried Blood Spot (DBS) Methodology Study for Biomarker Discovery in Lysosomal Storage Disease (LSD)

Abstract

Lysosomal storage diseases (LSDs) are a heterogeneous group of inherited metabolic diseases caused by mutations in genes encoding for proteins involved in the lysosomal degradation of macromolecules. They occur in approximately 1 in 5000 live births and pose a lifelong risk. Therefore, to achieve the maximum benefit from LSDs therapies, a fast and early diagnosis of the disease is required. In this framework, biomarker discovery is a significant factor in disease diagnosis and in predicting its outcomes. On the other hand, the dried blood spot (DBS) based metabolomics platform can open up new pathways for studying non-directional hypothesis approaches to biomarker discovery. This study aims to increase the efficiency of the developed methods for biomarker development in the context of rare diseases, with an improved impact on the reliability of the detected compounds. Thereby, we conducted two independent experiments and integrated them into the screening of the human blood metabolome: (1) comparison of EDTA blood and filter cards in terms of their suitability for metabolomics studies; (2) optimization of the extraction method: a side-by-side comparison of a series of buffers to the best utility to the disease of interest. The findings were compared to previous studies across parameters such as metabolite coverage, sample type suitability, and stability. The results indicate that measurements of metabolites are susceptible to differences in pre-analytical conditions and extraction solvents. This proposed approach can increase the positive rate of the future development of biomarkers. Altogether, the procedure can be easily adapted and applied to other studies, where the limited number of samples is a common barrier.

Keywords: biomarkers; dried blood spot (DBS); lysosomal storage diseases (LSD); mass spectrometry; metabolomics.

Figure 1

Investigation of the DBS stability. (a) The effect of six years of storage on the stability of DBS cards. Per year, blood was drawn from the same individuals (six controls), dripped onto DBS cards within two hours of collection, and stored at −20 °C. As seen in the figure, the optimal variation (5%) was present in few metabolites, while most of them had variations of up to 50%. (b) Box plots show an example of a random compound (7.30_439.3416 m/z) in DBS samples from the six control subjects collected from 2014 to 2019 and its variation in abundance throughout the years. Box = 25th and 75th percentiles; bars = min and max values.

Figure 2

Inter-day influence (three days) of the storage conditions on the human blood metabolome. When DBS samples stored at room temperature were compared to DBS samples maintained at −20 °C, majority of the compounds showed a CV ranging from 20% to 50%.

Figure 3

Overview of sample selection, storage and DBS preparation. (a) Type of samples received from physician per patient. The samples received from each patient were in the form of either DBS card, or whole EDTA blood, or DBS and EDTA. Stability study 2 comprised only patients with DBS and EDTA samples combined. (b) Three different categories of samples were used in stability study 2.

Figure 4

The heatmaps show differences in the abundance of identified compounds across various types of samples. Only the 50 most significant compounds were selected. The colors indicate the abundance of the metabolites: brown indicating the higher level and blue the lower level. The darker the square, the more significant the difference. The data were log-transformed and auto-scaled. The individual samples (columns) and compounds (rows) are separated using hierarchical clustering (Ward’s algorithm).

Figure 5

A comparison of the total number of compounds extracted from each of the four extraction solvents.

Figure 6

Dotted lines represent the number of features detected and their abundances across the four different extraction solvents. Each dot represents a solvent type.

Figure 7

Heatmap showing the reproducibility of the extraction procedure.

Figure 8

Boxplots showing the distribution of a random metabolite (4.40_421.2811 m/z) across different extraction solvents. Control–asymptomatic LSD subjects (n = 12); affected–LSD symptomatic patients (n = 8). Box = 25th and 75th percentiles; bars = min and max values.

Figure 9

Boxplots showing the “behavior”:abundance distribution of a CLN6 (red color) specific biomarker candidate (4.42_384.3240n) in four cohorts (LSD n = 27, NCL n = 19, CLN6 n = 30, control n = 20) using methanol extraction. The circles represent the outliers. Boxes represent the 25th and 75th percentiles, while bars represent minimum and maximum values.