*3.3. Apically and Basolaterally Released EVs Show Distinct RNA Cargo Composition*

Small RNAs were isolated from EVs retrieved from the apical and basolateral compartments of airway epithelial air−liquid interface cultures from healthy and asthmatic subjects (each n = 3), and were further analyzed by small RNA sequencing. All libraries exhibited a minimum of 7.5 million uniquely mapped reads and were thus comparable in efficiency. RNA composition was determined by counting the percentages of reads mapped to different species of RNA, such as rRNAs, long-non-coding RNAs (lncRNAs), miRNAs, mRNAs, and piRNAs. Our analysis showed a significant difference in the composition of RNA subtypes between apical and basolateral EV populations. Apical EVs contained comparable proportions of miRNAs (37.3%) and lncRNAs (33.3%) as te most prominent fractions, while basolateral EVs contained miRNAs only at a percentage of 2.3% and were rather dominated by a high percentage of lncRNAs (50.5%; Supplementary Figure S1).

Further focusing on miRNAs composition, apical and basolateral vesicles could be clearly separated from each other as two distinct populations in a principal component analysis. Moreover, in each of these populations, two clearly different clusters representing either the healthy or the asthmatic condition were clearly distinguishable (Figure 4). When looking at differences in miRNAs composition between apical and basolateral EVs in all 12 samples, we found 236 significantly differentially expressed miRNAs between the two subgroups, of which 151 miRNAs were more and 85 miRNAs less abundant in the apical compared to the basolateral EVs (Figure 5A,B). More frequent miRNAs in the apical population were assigned to different miRNA families (groups of miRNAs with a high sequence similarity deriving from distinct genomic loci) than those found at higher levels in the basolateral EVs (Figure 5C,D). In apically secreted EVs, all family members of the miR-30 (6/6) and the miR-941 (5/5) family were present, pointing to a significant association of these miRNAs to processes specifically important to the apical environment. Additionally, 10 out of 12 miRNAs from the let-7, 6 out of 8 of the miR-10, and 5 out of 8 of the miR-17 families were present. On the basolateral side, the most represented miRNA family was the miR-320 family with 7 out of 8 members, followed by the miR-181 (4/6), the miR-550 (3/5), the let-7 (3/12), and the miR-154 (3/19) families.

We then investigated whether these differentially distributed EV miRNAs could be linked to specific biological effects by evaluating the KEGG pathways and biological functions predicted to be affected by them, according to the two databases, DIANAmiRPath v3.0 and TAM 2.0. As shown in Figure 6, the significantly associated KEGG pathways of the preferentially apically secreted miRNAs included, among others, the mTOR and MAPK signaling pathways. Interestingly, miRNAs on the basolateral side were associated with processes of T and B cell receptor signaling, along with others (Figure 6A). Thus, associated KEGG pathways deviated in apically and basolaterally secreted EVs, suggesting different downstream functions for EVs depending on the site of action that are linked to diverse potential biological functions, as shown in Figure 6B. Significantly enriched target regulatory proteins can be found in Figure 6C. The results showed very distinct differences in miRNA composition as well as in downstream targeted proteins and pathways of vesicular RNA, depending on their cell side of secretion.

**Figure 4.** PCA plot depicting the clustering of 12 EV samples according to their miRNA cargo depending on disease condition and cellular side of EV release. PCA—principal component analysis; EV—extracellular vesicle.

**Figure 5.** Differential expression analysis of vesicular miRNAs in apical versus basolateral compartments of bronchial epithelial cell cultures. (**A**,**B**) Volcano plot and heat map showing the differentially expressed miRNAs within EVs secreted to the apical versus basolateral compartment at *p*adj < 0.05 and FC ≥ 2, and (**C**,**D**) donut charts showing the distribution of the mainly represented miRNA families in both compartments. Digits in brackets depict the number of enriched miRNA family members out of the total number of miRNAs belonging to the respective family, \* *p* < 0.05, \*\* *p* < 0.01, miRNA—microRNA; EV—extracellular vesicle; FC—fold change.

**Figure 6.** Functional analysis of miRNAs differentially expressed in EVs released by bronchial epithelial cells to the apical versus basolateral compartments. (**A**) KEGG pathway and (**B**) biological functions analyses of differentially expressed miRNAs in both compartments and (**C**) target proteins potentially regulated by the differentially expressed miRNAs. EV—extracellular vesicle.
