**3. Results**

#### *3.1. CBD-Induced Inflammatory Response Pathway Changes in E2-Deficient and -Sufficient Female Mice*

A whole transcriptomic RNA-Seq analysis of colon tissues (*n* = 4/group) was performed to investigate differential gene expression due to OVX surgery or CBD treatment. The PCA plot showed that samples within surgery and treatment groups clustered together (Figure S1).

Comparing VEH- and CBD-treated OVX groups, there was a total of 2585 differentially expressed genes (DEGs, q < 0.05) of which 1334 genes were upregulated and 1255 were downregulated (Table 1). A comparison of VEH- and CBD-treated SS groups revealed 14,508 DEGs (q < 0.05) where 964 genes were upregulated and 13,544 were downregulated (Table 1). There were 3162 DEGs (q < 0.05), 1552 increased and 1610 decreased, due to a loss of ovarian E2 (SS+VEH vs. OVX+VEH).

**Table 1.** Differentially Expressed Genes in Total Transcriptome.


The number of differentially expressed genes (DEGs) generated from EdgeR analysis of total transcriptome based on surgery or treatment comparison. Significance based on false discovery rate (FDR) correction (q < 0.05). OVX: ovariectomized; SS: sham surgery; VEH: vehicle treatment; CBD: cannabidiol treatment.

Gene ontology (GO) enrichment analysis was performed with ShinyGO for OVX+VEH vs. OVX+CBD, SS*+*VEH vs. SS+CBD, and SS+VEH vs. OVX+VEH using the gene ontology biological process (GOBP) database. For each of these comparisons, 1000 significantly altered GOBP pathways were identified after FDR correction and the top 20 pathways are shown in Figure S2. In a previous study, relative to VEH-treatment, CBD was found to reduce the expression of inflammatory mediators in the colon (*Il1b*, *Il6*, *Tnf*) and ileum (*Il1b*, *Il6*) in SS and/or OVX mice [26]. To find pathways related to inflammation the GOBP pathways were searched using the key word "inflammatory" and the inflammatory response pathway was found for OVX*+*VEH vs. OVX*+*CBD (pathway ranked 323), SS*+*VEH

vs. SS*+*CBD (pathway ranked 219), and SS+VEH vs. OVX+VEH (pathway ranked 729). Inflammatory pathways were not detected when DEGs were mapped using KEGG.

There were 114 DEGs for the OVX+VEH vs. OVX+CBD comparison, 111 DEGs for the SS+VEH vs. SS+CBD comparison, and 121 DEGs for the SS+VEH vs. OVX+VEH comparison (Figure 1A–D). The annotations of DEGs are provided in Supplementary File S1. There were 39 DEGs uniquely altered due to the CBD treatment of OVX mice and 40 DEGs uniquely altered in the CBD-treated SS group (Figure 1A). For the OVX+VEH vs. OVX+CBD and SS+VEH vs. SS+CBD comparisons, 24 DEGs were in common and all but one (*Epha2*) were changed by CBD in the same direction indicating that the changes were independent of E2 status (Figure 1A,B and Supplementary File S1). There were 34 DEGs due to OVX alone. The remaining overlapping DEGs (40, 11, and 36) were due to either OVX or CBD treatment (Figure 1A). Consistent with prior colon tissue qPCR analysis [26], *Tnf* was significantly decreased in CBD-treated OVX mice compared to VEH-treated OVX mice (Figure 1B,D and Supplementary File S1). *Tnf* was increased in the VEH-treated OVX group compared to the VEH-treated SS group but was not detected as a DEG when comparing the VEH- and CBD-treated SS groups (Figure 1C and Supplementary File S1).

**Figure 1.** *Cont*.

**Figure 1.** Differentially expressed inflammatory response genes. (**A**) Venn diagram showing the DEGs for each of the indicated comparisons as well as DEGs that are in common between comparisons. DEGs held in common do not necessarily indicate the same direction of change. DEGs for (**B**) OVX+VEH vs. OVX+CBD, (**C**) SS+VEH vs. SS+CBD, and (**D**) SS+VEH vs. OVX+VEH generated based on Euclidean clustering. Z-score scale indicates downregulated genes from 0 to −2.5 (orange to yellow shades) and upregulated genes from 0 to 2.5 (red to purple/navy shades).

For the OVX+VEH vs. OVX+CBD comparison, CBD treatment resulted in the downregulation of 78 of the 114 differentially expressed inflammatory pathway genes (Figure 1B). In contrast, for the SS+VEH vs. SS+CBD comparison, CBD treatment resulted in the upregulation of 63 of the 111 differentially expressed inflammatory pathway genes (Figure 1C). A similar heatmap pattern was observed for the SS+VEH vs. OVX+VEH comparison where 71 of 121 differentially expressed inflammatory pathway genes were increased due to ovarian E2 deficiency (Figure 1D).

#### *3.2. CBD-Induced Bile Secretion Pathway Changes in E2-Deficient and -Sufficient Female Mice*

A prior study found that, compared to VEH-treated OVX mice, CBD-treated OVX mice had alterations to serum and ileal content BA profiles [26]. ShinyGO enrichment analysis followed by GOBP pathway enrichment did not uncover any hits using the keyword "bile". Shiny GO enrichment analysis using the KEGG pathway database revealed that CBD significantly altered 134 pathways in OVX mice and 209 pathways in SS mice, while 92 pathways were altered due to ovariectomy. For each of these comparisons, the top 20 KEGG

pathways were ranked based on FDR correction (Figure S3). The "Bile secretion" pathway was detected in the top 20 pathways of DEGs for OVX+VEH vs. OVX+CBD and SS+VEH vs. SS+CBD but not for SS+VEH vs. OVX+VEH (Figure S3).

Relative to VEH-treatment, the CBD-treated OVX group had 27 DEGs involved in bile secretion and the CBD-treated SS group had 29 DEGs (Figure 2A and Supplementary File S2). The OVX+VEH vs. OVX+CBD and SS+VEH vs. SS+CBD comparisons had 17 DEGs (*Abcb1a, Abcc3, Abcg2, Abcg5, Adcy3, Adcy5, Nceh1, Nr1h4, Rxra, Sct, Slc10a2, Slc4a2, Slc51b, Ugt1a1, Ugt1a6a, Ugta7c, Ugt2b5*) in common, indicating that these changes to the bile secretion pathway were CBD-induced and unrelated to surgery (Figure 2A and Supplementary File S2). For OVX+VEH vs. OVX+CBD, all 17 DEGs were upregulated due to CBD (Figure 2A,B and Supplementary File S2). For the SS+VEH vs. SS+CBD comparison, 14 DEGs were upregulated and 3 were downregulated by CBD (Figure 2A,C and Supplementary File S2).

**Figure 2.** Differentially expressed bile secretion genes. (**A**) Venn diagram showing DEGs for indicated comparisons as well as DEGs that are in common between comparisons. DEGs held in common do not necessarily indicate the same direction of change. DEGs for (**B**) OVX+VEH vs. OVX+CBD and (**C**) SS+VEH vs. SS+CBD were generated based on Euclidean clustering. The Z-score scale indicates downregulated genes from 0 to −2.5 (orange to yellow shades) and upregulated genes from 0 to 2.5 (red to purple/navy shades).

#### *3.3. Colon Content and Hepatic BA Profiles*

Due to the upregulation of bile secretion pathway genes in the colon tissue, BAs in colon content were quantified. Hepatic BAs were profiled to investigate potential effects of CBD on hepatic BA production. The hepatic markers of inflammation were also investigated. The concentrations of total BAs (TBAs), primary BAs (PBAs), secondary BAs (SBAs), and conjugated BAs were similar between groups regardless of surgery or CBD treatment (Table S3). CBD did not alter the concentrations of individual BAs in colon content and liver tissue (Table S3). CBD did not induce differences in the hepatic expression of *Tnf*, *Nos2*, *Il1b*, or *Il6* (Figure S4). Compared to SS+VEH group, the OVX+VEH group showed less a hepatic expression of *Il6* (Figure S4).

#### *3.4. CBD Suppressed Inflammation in Ileal Organoids*

Compared to vehicle treatment, the combined TNF*a* and LPS (TL) treatment of ileal organoids induced the gene expression of inflammatory markers *Nos2* and *Tnf* (Figure 3). Organoids treated with CBD concentrations of 100 or 250 μM suppressed the TL-induced expression of *Tnf* and *Nos2* where the latter showed a dose-dependent effect. Organoids treated with 500 μM CBD also appeared to decrease the TL-induced expression of *Tnf* and *Nos2*; however, this reduction may be due to the lower viability of the organoids with the 500 μM CBD dose (Figure S5). The other treatments resulted in organoid viability which was similar to NT (Figure S5). CBD treatments alone did not alter the expression of *Nos2* and *Tnf* (Figure 3). The mRNA levels of Il1b and Il6 were also assessed by qPCR but were not detected.

**Figure 3.** *Cont*.

**Figure 3.** CBD decreased expression of inflammatory markers in intestinal organoids. Mature ileal organoids were treated with Tnfα + LPS (TL) to induce inflammation or treated with increasing concentrations of CBD in the absence or presence of TL. Negative controls consisted of no treatment (NT). On day 4, after organoids were passaged, organoids were incubated with treatments (*n* = 6 wells per treatment) for 24 h; each well contained approximately 100 mature organoids. Organoids were harvested and RNA was extracted for qPCR and the relative expression of target genes were determined using the 2e-Δct method. Outliers were not detected after ROUT test. Significant differences between treatments were detected using one-way ANOVA followed by a Tukey post hoc test. Different letters indicated significant difference between treatments.
