Nasal Mucosa

The Iso-Seq method for sequencing full-length transcripts was developed by PacBio during the same time period as the genome assembly. We therefore used this technique to improve characterization of transcript isoforms expressed in cattle tissues using a diverse set of tissues collected from L1 Dominette 0 1449 upon euthanasia. The data were collected using an early version of the Iso-Seq library protocol [26 ] as suggested by PacBio. Briefly, RNA was extracted from each tissue using Trizol reagent as directed (Thermo Fisher). Then 2 μg of RNA were selected for PolyA tails and converted into complementary DNA (cDNA) using the SMARTer PCR cDNA Synthesis Kit (Clontech). The cDNA was amplified in bulk with 12–14 rounds of PCR in 8 separate reactions, then pooled and size-selected into 1–2, 2–3, and 3–6 kb fractions using the BluePippin instrument (Sage Science). Each size fraction was separately re-amplified in 8 additional reactions of 11 PCR cycles. The products for each size fraction amplification were pooled and purified using AMPure PB beads (Pacific Biosciences) as directed, and converted to SMRTbell libraries using the Template Prep Kit v1.0 (PacBio) as directed. Iso-Seq was conducted for 22 tissues including abomasum, aorta, atrium, cerebral cortex, duodenum, hypothalamus, jejunum, liver, longissimus dorsi muscle, lung, lymph node, mammary gland, medulla oblongata, omasum, reticulum, rumen, subcutaneous fat, temporal cortex, thalamus, uterine myometrium, and ventricle from the reference cow, as well as the testis of her sire. The size fractions were sequenced in either 4 (for the smaller 2 fractions) or 5 (for the largest fraction) SMRTcells on the RS II instrument. Isoforms were identified using the Cupcake ToFU pipeline [27 ] without using a reference genome.
Short-read–based RNA-seq data derived from tissues of Dominette were available in the GenBank database because her tissues have been a freely distributed resource for the research community. To complement and extend these data and to ensure that the tissues used for Iso-Seq were also represented by RNA-seq data for quantitative analysis and confirmation of isoforms observed in Iso-Seq, we generated additional data, avoiding overlap with existing public data. Specifically, the TruSeq stranded mRNA LT kit (Illumina, Inc.) was used as directed to create RNA-seq libraries, which were sequenced to ≥30 million reads for each tissue sample. The Dominette tissues that were sequenced in this study include abomasum, anterior pituitary, aorta, atrium, bone marrow, cerebellum, duodenum, frontal cortex, hypothalamus, KPH fat (internal organ fat taken from the covering on the kidney capsule), lung, lymph node, mammary gland (lactating), medulla oblongata, nasal mucosa, omasum, reticulum, rumen, subcutaneous fat, temporal cortex, thalamus, uterine myometrium, and ventricle. RNA-seq libraries were also sequenced from the testis of her sire. All public datasets, and the newly sequenced RNA-seq and Iso-Seq datasets, were used to annotate the assembly, to improve the representation of low-abundance and tissue-specific transcripts, and to properly annotate potential tissue-specific isoforms of each gene.

We designed a computational pipeline (Figure 1B, further details in Supplementary Methods), to accurately map and quantify usage of different poly(A) sites on a genome scale, profiled by the SAPAS method. In summary, we first filtered Illumina-sequenced SAPAS reads to discard the reads with unrecognizable linker sequence, and trimmed to remove the linker and the ‘T's that just followed the linker until a not-‘T’ was met. If the length of a trimmed read was <25 nt, we discarded the read too. We then aligned all qualified reads to the corresponding genome using Bowtie software, version 0.12.5 (26 ). For internal priming filtering, we used the uniquely mapped reads by detecting the downstream genomic sequence 1 to 20 of cleavage sites as previously (8 (link)), that is, the read was regarded as an internal priming candidate if this 20-nt region contained more than 12 ‘A's or one of the following patterns: 5′-AAAAAAAA-3′ and 5′-GAAAA+GAAA+G-3′, where ‘+’ means ‘or more’. We defined cleavage sites by iteratively clustering the reads, locating tailing ends within 24 nt from each other and which were also aligned to the same strand of a chromosome. Cleavage clusters with two or more normalized reads were taken as poly(A) sites, and we searched for the corresponding poly(A) signals within the upstream sequence 1 to 50 nt from each poly(A) site. Using the gene structure and annotation from bioinformatics sites such as Ensembl and UCSC (27 (link),28 (link)), we annotated the poly(A) sites and corresponding poly(A) signals.
Based on this pipeline, we processed SAPAS raw reads of samples from three model organisms, zebrafish (D. rerio), mouse (M. musculus) and human (H. sapiens), to generate poly(A) site datasets. These raw reads of samples (see details in Supplementary Notes), invole in zebrafish embryos in various development stages from 0 h post fertilization (hpf) to 5 day post fertilization (dpf), mouse thymic development from 15.5 days post fertilization (dpf) to 90 days post parturition (dpp), human normal 22 tissues (brain, lung, thyroid, spleen, stomach, kidney, cervix, heart, lymph node, placenta, uterus, bladder, breast, prostate, liver, pancreas, small intestine, thymus, adipose, skeletal Muscle, ovary and testicle), human breast cancer and normal cells, human carcinomatous and normal tissues of intestinum rectum, as well as the nasal polyps and nasal uncinate process mucosa of chronic rhinosinusitis patients with nasal polyps.

Total RNA was extracted from nasal mucosa samples using RNeasy Mini Kit (Qiagen, Valencia, CA, USA) with the addition of a DNAse treatment (Promega, Madison, WI, USA) and re-purification phase. Extracted RNA was further purified using the RNA Clean and Concentrator (ZYMO Research, Irvine, CA, USA) and quality assessed by RNA 6000 Pico Kit and 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). From those samples exhibiting high RNA integrity, 650 ng of RNA was submitted to the UCSF Shared Microarray Core Facilities for microarray experiments employing the Whole Human Genome 4X44K array platform (#G4112F, Agilent Technologies). Microarray data has been deposited in the Gene Expression Omnibus [GEO:GSE46171].

Mice were vaccinated by TA route with 1 or 10 μg of S1 antigen with or without SF‐10 twice or thrice every 2 weeks. TA vaccination was performed by administering 30 μL of S1‐SF‐10 or saline containing S1 antigen into the nasal cavity of mice using a pipette; the dosage is the amount sufficient covering a mucous membrane from the nasal cavity to the lower respiratory tract (Figure S1).
As a positive control group, mice were immunized intramuscularly (into thigh muscles) with 10 μg S1 in 50 μL of saline or 10 μg S1 mixed with an adjuvant AddaS03™ (InvivoGen, San Diego, CA) using a 1 mL plastic syringe. S1‐AddaS03™ solution was prepared according to the instructions provided by the manufacturer. At 2 weeks after the last immunization, serum and bronchoalveolar lavage fluid (BALF) samples were obtained as described in detail previously.¹⁵