Genetic Deletion of Trace-Amine Associated Receptor 9 (TAAR9) in Rats Leads to Decreased Blood Cholesterol Levels

In the last two decades, interest has grown significantly in the investigation of the role of trace amines and their receptors in mammalian physiology and pathology. Trace amine-associated receptor 9 (TAAR9) is one of the least studied members of this receptor family with unidentified endogenous ligands and an unknown role in the central nervous system and periphery. In this study, we generated two new TAAR9 knockout (TAAR9-KO) rat strains by CRISPR-Cas9 technology as in vivo models to evaluate the role of TAAR9 in mammalian physiology. In these mutant rats, we performed a comparative analysis of a number of hematological and biochemical parameters in the blood. Particularly, we carried out a complete blood count, erythrocyte osmotic fragility test, and screening of a panel of basic biochemical parameters. No significant alterations in any of the hematological and most biochemical parameters were found between mutant and WT rats. However, biochemical studies revealed a significant decrease in total and low-density lipoprotein cholesterol levels in the blood of both strains of TAAR9-KO rats. Such role of TAAR9 in cholesterol regulation not only brings a new understanding of mechanisms and biological pathways of lipid exchange but also provides a new potential drug target for disorders involving cholesterol-related pathology, such as atherosclerosis.


SUPPLEMENTARY MATERIALS
Supplementary Figure 1. Full-length agarose gel: reverse transcription-polymerase chain reaction (RT-PCR) with TAAR9 and HPRT (housekeeping gene) specific primers using RNA isolated from the main olfactory epithelium (MOE) of WT, KO delC , and KO insA rats confirmed a corresponding decrease in TAAR9 mRNA expression in the main olfactory epithelium (MOE) of both KO strains. Negative control is K-. Red is a source of cropped image. Supplementary Meta-analysis 1. TAAR9 expression in organs involved in cholesterol metabolism regulation (RNAseq datasets meta-analysis).

Supplementary
Cholesterol concentration in plasma depends on exogenous and endogenous sources under complex regulation. Approximately half of the dietary cholesterol entering the intestines is absorbed [1]. The liver plays the main role in cholesterol metabolism. In this organ, cholesterol contributes to de novo synthesis and storage. High-density lipoprotein formation, cholesterol-containing chylomicron remnants and low-density lipoprotein particles uptake from plasma also occur in the liver [2]. Cholesterol up taking in hepatocytes is also insulin-dependent [3]. The inverse association between insulin and LDL cholesterol plasma levels is described both in normoglycemic and diabetic subjects [4]. The liver is also the principal site of cholesterol excretion, converting cholesterol to bile acids and removing free cholesterol as neutral sterols via biliary excretion [5]. Circulating levels of leptin showed a significant positive correlation with LDL-cholesterol and a negative correlation with HDL-cholesterol [6], so the cholesterol metabolism depends also on adipokines from adipose tissue. To estimate possible TAAR9 contribution in cholesterol metabolism regulation, its expression was evaluated in public transcriptomic data for intestine, liver, pancreatic islets, and adipose tissue.

RNAseq data processing
After exclusion of all non-relevant datasets, 19 RNAseq datasets were selected for future analysis. GSE IDs and titles data for selected microarray-generated and NGS-generated datasets are listed in Supplementary Table S3. Count or normalized count matrices for these datasets were downloaded from its supplementary data on GEO Accession Display. Data were TMM normalized and CPM transformed by the Bioconductor R package edgeR [2] if necessary. TAAR9 or Taar9 expression data were extracted from TMM normalized CPM data for human or rat datasets, respectively.

Results
The following human datasets met the inclusion criteria: intestine -1 dataset, liver -3, Langerhans islets -2, and adipose tissue -6. Additionally, included rat datasets were as follows: liver -5 and adipose tissue -2. Only tissue samples from healthy human or non-treated animals were reviewed. No TAAR9 expression was detected in intestinal and liver samples. TAAR9 expression wasn't detected in rat adipose tissue samples, but it was positive in 15 of 520 (3%) human adipose tissue specimens. In human pancreatic islets, TAAR9 expression was detected in 99% of specimens; only one sample in GSE165121 [10] dataset was TAAR9 expression negative.
Using the Expression Atlas thresholds [11] as a guide, the TMM normalized CPM values were below the cutoff in all adipose tissue specimens in analyzed datasets (i.e. CPM < 0.05). In all islet samples in GSE108072 and one sample in GSE165121 [10] datasets was expressed below Expression Atlas cutoff. So 28 islet specimens in GSE165121 [10] were TAAR9 expression positive at low levels according to Expression Atlas thresholds (i.e. between 0.5 to 10 CPM). TAAR9 expression wasn't detected in the human gastrointestinal tract, liver, pancreas, or adipose tissue in the Expression Atlas database [11]. In mice, the low expression below cutoff was detected in the liver only [11].
Using scRNA-seq data generated by the Tabula Muris we found expression of TAAR9 in mouse epithelial cells of the large intestine. Interestingly, most of the cells do not express TAAR9. Only 17 of 2019 cells (0.84%) are positive for TAAR9 expression. The median expression level 4.2 CPM in this cell group may be considered as low positive according to Expression Atlas thresholds (i.e. between 0.5 to 10 CPM).
Taking into account low TAARs expression in tissues and heterogeneous distribution of TAAR9 mRNA between the cells demonstrated in the Tabula Muris dataset, rare TAAR9-positive cases and its controversial values in fat tissue and Langerhans islets aren't negligible for TAAR9 expression estimation.