The Bridging Vein to Vein with Big Data session included the following presentations:
1. Shuoyan Ning: Transfusion-related immune modulation and the use of big data analytics
2. Jingcheng Zhao: Bloody big data - Haemovigilance using routinely collected healthcare data
3. Angelo D'Alessandro: Chemical individuality of the blood donor as gleaned by high-throughput metabolomics of over 13,000 end of storage red blood cell samples and multi-omics analyses of 643 recalled donors
4. Abdirahaman Musa Jibrail: Bridging the Gap Geospatial Analysis to Estimate Demand and Unmet Need of Blood Products in Rural Kenya
MODERATORS: Gustaf Edgren and Katja van der Hurk
After the presentation, there was a questions and answers session, which is also included in the recording.
Chemical individuality of the blood donor as gleaned by high-throughput metabolomics of over 13,000 end of storage red blood cell samples and multi-omics analyses of 643 recalled donors
A D'alessandro1, T Nemkov1, D Stephenson1, C Erickson1, A Key1, A Moore2, G Page2, M Stone3, X Deng3, S Kleinman4, M Busch3, P Norris3
1University of Colorado Anschutz Medical Campus, Aurora, 2RTI International, Atlanta, 3Vitalant Research Institute, San Francisco, United States, 4University of British Columbia, Victoria, Canada
Background: Garrod's principle of chemical individuality posits that “each person is biochemically unique due to inherited differences in enzymes”. Application of omics technologies to the field of transfusion medicine is shedding new light on this principle, to the extent that the metabolic phenotype of donated blood impacts the storage quality of blood products and, potentially, transfusion efficacy. Using metabolomics approaches, we introduced the concept of “metabolic age” of the stored red blood cell, which is impacted by (i) biological factors (e.g., donor sex, age, body mass index), (ii) genetic polymorphisms and (iii) non-genetic factors (e.g., dietary habits, nicotine, caffeine or alcohol exposure, exposure to drugs that are not grounds for blood donor deferral). However, technical limitations have constrained studies of stored red blood cell metabolism to small scale investigations.
Aims: To determine the metabolic signature of haemolytic propensity and intra-donor reproducibility of metabolic phenotypes across multiple donations in the largest cohort of blood donors investigated to date with omics approaches.
Methods: Here we leveraged a novel ultra-high throughput metabolomics method to investigate the metabolic phenotypes of end of storage packed red blood cell samples from 13,091 donor volunteers enrolled in four different blood centres across the United States as part of the Recipient Epidemiology and Donor Evaluation Study—REDS RBC Omics. A subset of these donors (n = 643) were identified as extreme haemolyzers, as they ranked either below the 5th or above the 95th percentile with respect to end of storage red cell haemolytic propensity (spontaneous or following oxidative or osmotic insults). These donors were asked to donate a second unit of packed red blood cells, which were stored for 10, 23 and 42, for a total of 1929 samples. These longitudinal samples from the recalled donor cohort underwent multi-omics characterization via metabolomics, proteomics and lipidomics. For all the index and recall donors, genomics data on 879,000 SNPs were screened through a precision transfusion medicine array.
Results: Combined analysis of the index and recalled cohort of over 13,000 and 643 donors, respectively, identified novel markers of red blood cell haemolytic propensity, while indicating that only hypoxanthine strongly (positively) correlated with storage and oxidative haemolysis, among the previously reported eight markers of the metabolic storage lesion. We also observed that kynurenine—a metabolite derived from tryptophan catabolism—was significantly positively correlated with osmotic fragility (q = 5.56 E-05 at storage day 42). Like osmotic fragility, end of storage kynurenine levels were found to be extremely reproducible for the same donor across multiple donations (n = 643; rho = 0.523; p = 1.27 E-46), suggesting a strong impact of donor biology on this metabolic pathway. Quantitative Trait Loci (mQTL) analyses were then performed to identify the genetic underpinnings of end of storage kynurenine levels on over 13,000 blood donors, analyses that identified polymorphisms in the tryptophan transporter (SLC7A5/LAT1), ATXN2 and IDO1 as contributors to inter-donor heterogeneity in kynurenine levels and osmotic fragility.
Summary/Conclusions: Donor metabolite levels are linked to haemolytic propensity and genetic factors. Further mQTL studies on the whole metabolome, and linkage of metabolomics data (including the exposome) to available recipient databases (which include haemoglobin increments in recipients of units from the same donors characterised here) promise to revolutionize our understanding of the genetic and non-genetic factors impacting red cell storage biology and transfusion efficacy.