Maternal allergic sensitization affects the T cell modulatory capacity of milk derived extracellular vesicles

Background: Human milk extracellular vesicles (EVs) affect various cell types in the gastrointestinal tract, including T cells, and play a role in the development of the newborn's immune system by delivering specific molecular cargo to target cells. Although maternal allergic sensitization alters the composition of milk, it is unknown whether this impacts the function of milk EVs. Therefore, we analyzed the T cell modulatory capacity and compared the protein and miRNA cargo of EVs from milk of allergic and non-allergic mothers. Methods: EVs were isolated from human milk from allergic and non-allergic donors by differential centrifugation, density gradient floatation and size exclusion chromatography. Functional modulation of primary human CD4+ T cells by EVs was assessed in vitro. Proteomic analysis and small RNA sequencing was performed on milk EVs to evaluate protein and miRNA abundance and to identify cellular targets of this EV cargo in relevant T cell signaling pathways. Results: T cell proliferation, activation and cytokine production were suppressed in the presence of milk EVs. Remarkably, milk EVs from allergic mothers modulated T cell activation to a lesser extent than EVs from non-allergic mothers. Integrative multi-omics analysis identified EV cargo of which the cellular targets could be linked to T cell activation-associated processes. Conclusions: Milk EVs from non-allergic mothers are stronger inhibitors of T cell activation compared to milk EVs from allergic mothers. This altered functionality might be linked to small changes in modulation of certain T cell signaling pathways. Overall design: Collection of human milk and was done as previously described [Zonneveld, JEV. 2021; PMID: 33732416]. Milk EVs were isolated as previously described [van Herwijnen, MCP. 2016; PMID: 27601599 ]using differential centrifugation and sucose density gradient separation. Milk EVs were isolated from 3.5-4 ml 10,000g milk. EV-enriched fractions from sucrose gradient isolation were diluted with PBS and pelleted at 192,000 g. EV pellets were resuspended in 700 µl Qiazol and frozen at -80°C until sRNA isolation for sequencing. Small RNA (< 200 nt) was isolated using the miRNeasy micro kit (Qiagen) according to manufacturer's instructions. Input quantity of sRNA for library preparation was based on equal volume of purified milk EVs and ranged between 10-150 ng of sRNA per sample. The Illumina® Truseq small RNA Sample Prep Kit was used to process the samples according to the kit-specific guidelines. Small RNA sequencing was done on the Illumina NextSeq 500 platform with 75 bp single-end sequencing. Samples were sequenced in 1 flow cell with 4 lanes. Samples with less than 7 million reads were sequenced in a second flow cell and data from 2 flow cells were combined. For each individual sample, 4 lanes were combined in a single fastq file. Subsequently, files for those samples which were sequenced in 2 different flow cells were combined. In total 40 fastq files, 20 from non-allergic donors and 20 from allergic donors were obtained. FastQC (version 0.11.5) was used to perform quality control checks on sequencing data. Remaining TruSeq small RNA adapter sequences were checked and clipped from all reads at the 3' end using cutadapt (version 2.8). Remaining adapters were removed from the 5' end of the reads using cutadapt and the low quality bases were trimmed from all sequence reads using sickle (version 1.33). all sequence reads shorter than 15 bases were discarded. The processed reads were aligned to human reference genome GRCh38 from Ensembl and the number of aligned reads per annotated sRNA was calculated. These processes were performed by Manatee algorithm (version 1.2) applying bowtie to align reads (version 1.0.1), allowing for alignments to a maximum of 50 loci and reporting 3 of them, with equal alignment scores, and a maximum of 3 mismatches. Manatee counts output was rounded and used as raw expression data. The small RNA annotation track was obtained from miRBase (version 22.1), GtRNAdb (version 2.0, January 2016) and Ensembl 99 in order to analyse the miRNAs.