**2. Results**

### *2.1. Determination of Total Alkaloid and Polysaccharide Contents in Di*ff*erent Species and Di*ff*erent Tissues*

Polysaccharides and alkaloids are the main medicinal components of *Dendrobium*. Therefore, we determined the polysaccharide and total alkaloid in three different tissues of three *Dendrobium* species (Figure 1). The biennial *Dendrobium o*ffi*cinale* has the highest content in the comparison of different tissues and species. For polysaccharide content, the stem is the main enrichment tissue, while for total alkaloid, the leaf is the main enrichment tissue. In the stem, the polysaccharide content of *Dendrobium* varied from 23.34 to 37.41%. In the leaf, the total alkaloid content of *Dendrobium* varied from 0.0291 to 0.0421%.

**Figure 1.** Polysaccharide and total alkaloid contents variation in different tissues of different species. (**a**) Polysaccharide content. (**b**) Total alkaloid content. There are three replicates of each sample. 2-Dh: two-year-old *Dendrobium huoshanese*, 2-Do: two-year-old *Dendrobium o*ffi*cinale*, 2-Dm: two-year-old *Dendrobium moniliforme*.

### *2.2. Di*ff*erential Expression Analysis*

In this study, we calculated the FPKM of each sample to regulate the expression and investigate the gene expression differences among different tissues of different *Dendrobium* species. We compared the three data sets from different comparison groups using a Venn diagram (Figure 2a–c). In addition, we compared the number of DEGs between different species in the same tissue, and many genes were differentially expressed only in one or two comparisons. Then the edgeR was used to test the differential expression of the repeated count data. We used "FDR < 0.05 & | log2FC | ≥1" as the criterion for significant differences in gene expression. When log2FC > 1, DEG is considered to be up-regulated. In contrast, for log2FC < −1, it is considered a downward adjustment. We did up and down-regulated analysis of DEGs for three different tissues and three different *Dendrobium* species, and the results are shown in Figure 3a–i.

According to the transcriptome data, all DEGs were used for hierarchical cluster analysis of transcription abundance in three different tissues. The heatmap of DEGs between different tissues and different *Dendrobium* species show similar transcriptome profiles for Dh\_R, Dh\_L, Dh\_S, Do\_R, Do\_L, Do\_S, Dm\_R, Dm\_L, and Dm\_S (Figure 4a). A total of 35,159 DEGs were identified and analyzed using criteria of log10 (FPKM+1) and *p* < 0.05. The trend of the specific expression level is shown in the number under the color bar at the top left. On the left is the gene cluster tree. The closer the degree of separation between the two genes, the closer their expression is. In order to reflect the main trends and tissue-specific expression of different *Dendrobium* species, all DEGs were clustered into ten expression profiles (Figure 4b) using the K-means method and hierarchical clustering with similar regulation model and log2 (foldchange). DEGs belonging to cluster 3 were more highly expressed in leaves than in other tissues. Except for subcluster 1 and subcluster 3, all the other subclusters have just one gene trend.

**Figure 2.** Venn diagram of differentially expressed genes (DEGs) in different comparisons. All DEGs are grouped into three comparison groups represented by three circles. The overlapping portions of the different circles represent the number of DEGs common to these comparison groups. (**a**) Venn diagram of three kinds of *Dendrobium* leaves. (**b**) Venn diagram of three kinds of *Dendrobium* roots. (**c**) Venn diagram of three kinds of *Dendrobium* stems.

**Figure 3.** Volcano plots of the DEGs in different comparisons. Red dots indicate significant up-regulation of genes, and blue dots indicate significant down-regulation of genes. Black dots represent non-DEGs. (**a**) Dm\_L vs. Dh\_L volcano; (**b**) Dm\_L vs. Do\_L volcano; (**c**) Do\_L vs. Dh\_L volcano; (**d**) Dm\_R vs. Dh\_R volcano; (**e**) Dm\_R vs. Do\_R volcano; (**f**) Do\_R vs. Dh\_R volcano; (**g**) Dm\_S vs. Dh\_S volcano; (**h**) Dm\_S vs. Do\_S volcano; (**i**) Do\_S vs. Dh\_S volcano.

**Figure 4.** Expression profiles of all DEGs. (**a**) The heatmap of DEGs and the FPKM distribution of all unigenes obtained by hierarchical cluster analysis. Each column in the figure represents a sample, and each row represents a gene. The colors in the graph indicate the magnitude of gene expression (log10 (FPKM + 1)) in the sample. Red indicates that the gene is highly expressed in the sample, and the blue indicates that the gene expression is low. (**b**) K-means clustering analysis of gene expression profiles. The blue line represents the expression model. The gray lines are the expression profiles of each DEGs. The x-axis represents different tissues of different Dendrobium plants. The y-axis represents log2 (ratio).

#### *2.3. GO and KEGG of DEGs Annotation and Enrichment Analysis*

GO is a database established by the Gene Ontology Consortium. Its purpose is to standardize biological terms about genes and gene products in different databases, to define and describe gene and protein functions. Using the GO database, genes can be classified according to biological process, cellular component, and molecular function. The GO annotation statistics were performed on the differentially expressed genes in pairs, and one of the samples was used as a control. The obtained results can be used to plot the GO annotation bar graph of up and down DEGs (Figure 5). Figure 5 only shows the comparison of the stems of different *Dendrobium* species, and the other two tissues are shown in the Supplementary Figure S1.

**Figure 5.** Gene Ontology (GO) annotations of up and down regulated DEGs. The bottom x-axis indicates the number of DEGs annotated to a GO term, the upper x-axis indicates the proportion of DEGs annotated to a GO terms to the total number of all GO annotated DEGs; and the y-axis represents each detailed classification of GO. (**a**) Dm\_S vs Dh\_S GO annotation; (**b**) Dm\_S vs. Do\_S GO annotation; (**c**) Do\_S vs. Dh\_S GO annotation.

KEGG focuses on biochemical pathways, especially genes involved in protein, carbohydrate, and energy metabolism. In the KEGG enrichment analysis, we mapped all these genes to a reference pathway in the KEGG database to determine the biological pathways in which these genes may be involved (Figure 6). In terms of the KEGG pathways, "Dm\_S vs. Dh\_S" comparisons are involved in 235 pathways (Figure 6a). In "Dm\_S vs. Do\_S" comparisons, 223 pathways were involved (Figure 6b). Finally, in "Do\_S vs. Dh\_S" comparisons, 237 pathways were involved (Figure 6c). Among these pathways, "flavonoid biosynthesis" was enriched in the Dm\_S vs. Dh\_S comparison. "Lipopolysaccharide biosynthesis" related to polysaccharide was the most enriched pathways in the Do\_S vs. Dh\_S comparison. Notably, we observed terpenoid biosynthesis in the secondary metabolites also enriched in the Do\_S vs. Dh\_S comparison. The KEGG pathways of different *Dendrobium* species of the other two tissues are shown in the Supplementary Figure.

**Figure 6.** Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment of DEGs. The x-axis represents the pathway name, and the y-axis represents the enrichment ratio (sample number/background number). (**a**) Dm\_S vs. Dh\_S; (**b**) Dm\_S vs. Do\_S; (**c**) Do\_S vs. Dh\_S. All pathways in the figure with asterisks indicate significant KEGG enrichment, with three asterisks indicating *p*-value < 0.001, two asterisks indicating *p*-value < 0.01, one asterisks indicating *p*-value < 0.05.
