The stem is an important organ in supporting plant body, transporting nutrients and communicating signals for plant growing. However, studies on the regulation of stem development in tomato are rather limited. Authors demonstrated that SlHB8 negatively regulated tomato stem development. SlHB8 belongs to homeo domain-leucine zipper Class III gene family transcription factors and expressed in all the organs examined including root, stem, leaves, flower, and fruit. During tomato stems development, SlHB displayed stable high expression level. Loss of function of SlHB8 induced stem diameter and xylem width, while overexpression of SlHB8 displayed opposite trend. Besides, inducing the expression level of SlHB8 resulted in lower lignin content as well as the expression level of lignin biosynthesis pathway genes both in tomato stem and leaves. In addition, lots of disease resistance genes were found differentially expressed in the SlHB8 transgenic plants indicating a possible role of SlHB8 in the biotic resistance pathway. Overall, SlHB8 acts as a negative regulator in stem development and lignin biosynthesis and has a potential role in the abiotic and biotic resistance pathway.
Stems are the central part of the plant, connected with the leaves up and the roots down, and transport important substances for long-distance cell-to-cell communication. Besides, the stem is involved in carbon storage and remobilization of plants, influencing the control of plant’s carbon metabolism 
. Therefore, understanding the regulation mechanism of stem differentiation is instrumental. The stem development is moderated by an elaborated regulation network which has been well elucidated in Arabidopsis
and woody species 
. The homeo domain-leucine zipper Class III gene family transcription factors (HD-Zip III) were regarded as one of the key factors during the stem development from stem primary establishment to lateral growth 
, five Class III HD-Zip transcription factors (REVOLUTA/IFL1
), and ATHB-8
) were isolated with four recognizable domains including a DNA binding homeodomain followed immediately by a leucine zipper motif (HD-Zip); a sterol/lipid binding (START) domain for binding small hydrophobic molecules such as steroid, phospholipids, or carotenoids; and a PAS (Per-ARNT-Sim) domain for protein-protein interaction 
. These five HD-Zip III transcription factors were reported to play roles in the regulation of primary and secondary vascular cell differentiation 
, meristem maintenance 
, leaf patterning 
and so on. Tortuous stems and leaves, dwarfism, and shortened internodes were found in these genes’ mutants 
. All these five members affect vascular development in Arabidopsis
by altering their expression levels in a dependent or redundantly way 
. Overexpression of ATHB-8
promotes vascular cell differentiation and xylem tissue production in the inflorescence stems of Arabidopsis 
, while REV together with PHB and PHV regulated the meristem development in lateral organs 
. REV, PHB and PHV were revealed to be an activator, while CAN and ATHB-8 were repressors for the formation of interfascicular cambium of the inflorescence stem 
. The expression of HD-Zip III genes was mediated by multiple molecular mechanisms. Such as the small Zip protein (ZPRs) and MiR165/166 
. It was reported that ZPR3 inhibited the HD-Zip III protein activity by interacting with HD-Zip III protein to form nonfunctional heterodimers 
. There were MiR165/166
target sites in the coding sequences of HD-Zip III genes and their expression levels were negatively regulated by MiR165/166 
The lignin content is always related to the stem development and genes affecting stem development also impact lignin biosynthesis 
. Previous studies have identified that members of homeodomain-leucine zipper gene family play important roles in stem tissue development as well as lignin regulation of plants 
. For example, knocking down of the POPCORONA
gene, one member of Class III HD-Zip transcription factor family in populous, results in abnormal lignification in pith cells 
, the polar HD-Zip III genes, were downstream targets of poplar IAA9-ARF5 module which regulated the secondary growth of poplar woody stems 
. In Arabidopsis
, members of the HD-Zip III gene family function differently, the interfascicular fiber of rev-6
mutant disappeared and lignin decreased, while loss of function of CNA
gene impacted vascular bundle development and increased lignin content 
. Ectopic expression of Zinnia HB12
regulated xylem parenchyma cells differentiation and up-regulated the expression of genes related to lignin monomer synthesis 
Lignin is one of the complex phenylpropanoid polymer, which is one of the main substances in secondary cell walls of plant vascular systems 
. lignin which widely existed in stem vascular system provided the strength that allows the stem to grow upright 
. Previous research has revealed that lignin is connected to plants’ response to stress 
. Lignin biosynthesis is affected by the abiotic stress such as drought stress 
, cold stress 
, salt stress 
, nutrient stress such as nitrogen deficiency 
, calcium deficiency 
, gases stress (CO2
and ozone) 
, and heavy metals stress 
. Inducing the lignin content or altering the lignin composition enhanced their resistant ability to these abiotic stresses. Such as: In grapevine, overexpression of VlbZIP30
enhances drought tolerance by activating the expression of lignin biosynthetic genes and increasing lignin deposition 
. Overexpression of PaSOD
improved plant’s tolerance to salt and cold stress by up-regulation of lignin induced by peroxide 
. And research on sweet potatoes has found that IbLEA14
overexpression plants exhibited increased drought and salt resistance due to the increase of lignin content caused by increased expression level of lignin biosynthesis gene 
. Over expression of two CBFs
changed the frost sensitivity of Eucalyptus by inducing lignin content and syringyl/guaiacyl (S/G) ratio as well as genes involved in the phenylpropanoid and lignin branch pathway 
. For the nitrogen fertilization affection on lignin is different with type and tissues examined. In pine (P. palustris
) seedlings, high-N fertilization reduced the lignin content in roots but had no effect on the lignin in aerial parts of the plant 
. In populous plants lignin content was increased by high-N due to elevated PAL activity 
. Apart from abiotic stress, lignin is involved in plant response to biotic stress. Lignin possesses antimicrobial properties that protect plants against pathogenic bacteria 
. Lignification is induced in response to attack by pathogen including bacteria, fungi and virus 
. In cotton, suppression of GhUMC1
reduced lignin biosynthesis genes due to decreased lignin content and further decreased the resistance of plants to Verticillium.
William has reported that AtMYB15
transcription factor acted in defense-induced lignification, having the capability of driving lignification, plants of myb15
mutant showed greater resistance to the bacterial pathogen Pseudomonas syringae 
. Moreover, lignin can be degraded to chemicals and fuels for industrial applications by many different species of microorganisms including fungi and bacteria, so lignin also protects the structural polysaccharides in plants, from microbial enzyme-mediated hydrolysis 
. Besides, lignin is important for the soil carbon cycling. Altering the lignin content in soil affects the bacterial community diversity index 
2. SlHB8 Displayed Stable and High Expression Level during Tomato Stem Development
Previous study showed that SlHB8
gene belongs to the HD-Zip III transcription factor family, as it contains the four conserved domains of HD, bZip, START and MEKHLA in the HD-Zip III transcription factor 
. Meanwhile, it expresses in all the tissues such as: root, stem, leaves, flower, mature green fruits, breaker fruits and red fruits and shows the highest expression level in stem tissue 
. To understand the possible function of the SlHB8
gene in tomato stem development, we checked its expression pattern in stems at different developmental stages by quantitative reverse transcription-polymerase chain reaction (qRT-PCR). The results showed that SlHB8
gene expressed in all the stages examined, including 20 D, 30 D, 45 D and 60 D stages stem tissues. Among these stages, the relative transcript level of SlHB8
gene maintained stable high in tomato stems at 20 D, 30 D, and 45 D stages but decreased a little in tomato stem at 60 D stage (Figure 1
Figure 1. Expression patterns of the SlHB8 gene in tomato stems. (A) Quantitative reverse transcription PCR analysis of the SlHB8 gene in different development stages of tomato stem. 20 D: 20 days after germination; Error bars mean ± standard error (SE) of three biological replicates. (B) RNA in situ hybridization of SlHB8 in stem tissues of SlHB8 overexpression tomato plant. Stems at 6th internodes of 2-month-old tomato plants cultivated in soil were cross-sectioned for hybridization with sense (upper) and antisense (lower) probes of SlHB8. The photos were taken under 10× (left) and 20× (right) microscopy. Black arrows indicate in situ hybridization signals for SlHB8 transcripts. Pi, pith; Ca, cambium; Ph, phloem; Xy, xylem. Bars: 101 um (left), 50 um (right).
3. SlHB8 Affects Tomato Stem Development through Mediating the Xylem Range
To identify its role in regulating stem development, we generated SlHB8 gene knockout mutant by using CRISPR/Cas9 technology and SlHB8 overexpressed transgenic tomato lines. Three kinds of SlHB8 loss of function mutants were verified by sequencing the sgRNA target site. Expression analysis by qRT-PCR showed that the relative transcript level of SlHB8 was strikingly upregulated in overexpression of SlHB8 lines (35sL1; 35sL2) but was specifically reduced in SlHB8-cr lines compared with wild type, respectively. Comparing to wild type plant, overexpression of the SlHB8 gene did not change plant height and internode length of stem, while loss of function of SlHB8 gene led to a 14 % reduction in plant height and the reduced plant height resulted from a 15 % reduction in internode length (Figure 2A). Increasing the relative transcription level of the SlHB8 gene or knock out of SlHB8 gene did not change the number of nodes in the plant. Phenotypic observation on stem diameters revealed that compared to wild type plant, overexpressing SlHB8 reduced stem diameter while loss of function of SlHB8 gene increased stem diameter (Figure 2C). To further understand the changed stem diameters in SlHB8 transgenic plant, we examined phenotypes of stem-associated cell types by carrying out the paraffin section analysis in WT and SlHB8 transgenic plants. There were apparent differences in the range of xylem in stems among different lines. These xylem cells were stained by toluidine blue. Quantitative measurement showed that compared with the wild type, overexpression of the SlHB8 gene reduced the xylem width of the tomato stem, while the xylem width enlarged in SlHB8-cr lines (Figure 2B,D). Overexpressing SlHB8 repressed the xylem development, with a 34 % decrease in the number of xylem cell layers, but SlHB8 gene knocking out increased the number of xylem cell layers by 12 %, compared with WT (Figure 2D,F). Furthermore, we measured the single cell size of xylem fibers, which had no obvious difference in all genotype plants (Figure 2E). The characteristics of pitch cells examination showed that overexpression of SlHB8 reduced the area of individual pitch cells in the stem and knocking out of SlHB8 gene did not result in significant differences in the size and number of pitch cells compared with WT. Interestingly, compared to the wild type, the size of xylem vessel cells did not change in the SlHB8-ox lines but decreased in SlHB8-cr mutants. To clarify whether the changed xylem width is related to the expression level of SlHB8, we determined the expression position of SlHB8 in the SlHB8-ox lines by using the RNA in situ hybridization on stems at the sixth internode of 2-month-old tomato (Figure 1B). The results revealed that strong expression signals of SlHB8 positive probes were observed in the area of pith, xylem, phloem and cambium regions compared with those of negative probes (Figure 1B), suggesting that SlHB8 gene was overexpressed in these tissues. Collectively, these data indicated that SlHB8 affects stem diameter by mediating the xylem range.
Figure 2. Phenotype analysis of SlHB8 overexpression and SlHB8 knock out lines. (A) Photos of adult plants of representative two-month-old SlHB8 overexpression and SlHB8 knock out lines. Bar: 5 cm; (C) Cross-sectioning and staining with toluidine blue of the 6th internode of 2-month-old wild-type, SlHB8 overexpression and SlHB8 knock out lines. Pi, pith; Ca, cambium; Ph, phloem; Xy, xylem. Bars: 200 um; (B) Measurement of stem diameter, (D) xylem width, (E) a single fiber cell size and (F) xylem cell layers in SlHB8 overexpression and SlHB8 knock out lines as well as WT plants. The calculation was performed on IMAGE J softer ware based on the images of toluidine blue-stained anatomical sections as described in the Materials and Methods section. In the bar chart, the gray barplots represent the wildtype line, the orange barplots represent the 35s-driven SlHB8 overexpression line, and the blue barplots represent the SlHB8 knockout line. Error bars mean ± standard error (SE) value. Stars indicate the statistical significance using Student’s t-test: * p-value < 0.05, ** p-value < 0.01, *** p-value < 0.001.
4. SlHB8 Affects Lignification in Tomato Stems and Leaves
As stem diameter is always positively related to the lignin biosynthesis, we examined the lignin content in stem tissues of SlHB8 transgenic plants by histochemical staining with hydrochloric acid-phloroglucinol which is used for lignin staining analysis. Staining results showed that compared with WT, the xylem of SlHB8-ox had a small lignin deposition area and significantly reduced staining brightness, indicating a decrease in lignin content, however, there was no significant difference in lignin deposition between SlHB8-cr and WT (Figure 3A). To confirm the level of lignification, the total lignin content in WT and SlHB8 transgenic plants was measured by the acetyl bromide (AcBr) method. Consistent with staining analysis result, the lignin content significantly decreased in stems and leaves of SlHB8-ox lines while not changed in stems and leaves of SlHB8-cr plants (Figure 3B,C).
Figure 3. SlHB8 affects lignification in tomato leaves and stems. (A) Free-hand sections of the 2-month-old stem were subjected to phloroglucinol-HCl staining. The red area represents lignin. Bars: 1.5 cm. (B) Acetyl bromide-soluble lignin assays were carried out on leaves (2-month-old tomato) of SlHB8 overexpression, SlHB8 knock out lines and WT plants. (C) The content of lignin in the stems (2-month-old tomato) of SlHB8 overexpression, SlHB8 knock out lines and WT plants was measured by acetyl bromide lignin assay. In the chart of B and C, the gray columns represent the wildtype line, the orange columns represent the 35s-driven SlHB8 overexpression (35sL2) line, and the blue columns represent the SlHB8 knockout (SlHB8-cr2) line Error bars mean ± standard error (SE) value for each line. Stars indicate the statistical significance using Student’s t-test: * p-value < 0.05.
5. Transcriptomic Analysis of WT, SlHB8-ox and SlHB8-cr Plants
To better understand the molecular mechanism of SlHB8 regulation of stem development, RNA-seq was carried out on stems of 2-month-old plant of WT, SlHB8-ox, and SlHB-cr mutant. Three biological replicates were included in each sample and finally generated 9 libraries. The high-quality clean reads of the library reached over 99 %. After filtering the rRNA, the library was uniquely mapped to the tomato genome (Solanum lycopersicum ITAG4.0). The mapped reads ranged between 97.04 % and 97.46 % and unique mapped reads ranged from 94.61 % to 95.32 %. The annotated gene numbers in the 9 libraries ranged from 22,265 to 22,885. A total of 627 novel transcripts were identified from the 9 libraries, each containing more than 570 novel genes. Principal component analysis (PCA) of the RNA-seq samples revealed highly repeatability between three replicates of each sample of the wild type, SlHB8-ox, and SlHB8-cr, and great differences among the stems of 2-month-old tomato in different lines (Figure 4A). The RNA-seq analysis showed a 3.8-fold difference in SlHB8 expression between wild-type and SlHB8-ox stems (p < 0.001, Student’s t-test), and the expression of the SlHB8 gene in the SlHB8-cr stems was 0.29 times than that in the, wild type stems (p < 0.001, Student’s t-test) closely corresponding to the results obtained by real-time quantitative PCR analysis.
Figure 4. Differentially expressed genes (DEGs) analysis in WT, SlHB8-ox, and SlHB8-cr plants. (A) Principal component analysis (PCA) of the three group samples (WT, red; SlHB8-cr, yellow; SlHB8-ox, blue); the x-axis represents the first principal component and the y-axis represents the second. (B) Histograms showing the DEGs number in WT vs. SlHB8-ox and WT vs. SlHB8-cr. (C) Top ten significantly enriched GO terms. (D) Heatmap of DEGs involved in the phenylpropanoid biosynthesis pathway. (E) Heatmap of DEGs belong to the MYB transcription factor. (F) Significantly enriched KEGG terms.
To identify candidate genes that are vital for stem development, we performed a comprehensive analysis of gene expression in stems at the 6th node of the 2-month-old tomato of WT, SlHB8-ox, and SlHB-cr mutant. Genes that satisfied the fold-change difference |log2 (fold-change)| > 1 and FDR < 0.05 were regarded as differentially expressed genes (DEGs). 1553 (656 up-regulated + 897 down-regulated) DEGs were detected in the comparison between WT and SlHB8-ox plants, and 1548 (586 up-regulated + 962 down-regulated) DEGs were found in the comparison between WT and SlHB8-cr plants (Figure 4B). A total of 2592 DEGs were found between WT and SlHB8 transgenic plants. To gain further insight into the putative functions of these DEGs between the wild type and SlHB8 transgenic lines, GO assignment and Kyoto Encyclopedia of Genes and Genomes (KEGG) database were used for the further analysis. Using q value ≤ 0.05 as the significant cut-off, the data revealed that these 2592 DEGs were significantly enriched in the GO terms related to disease resistance such as “response to endogenous stimulus,” “response to stimulus,” “response to fungus,” “response to external biotic stimulus” and “response to biotic stimulus”(Figure 4C) and 14 KEGG pathways were significantly enriched (Figure 4D) including pathways related to disease resistance and lignin biosynthesis such as “plant-pathogen interaction,” “MAPK (mitogen-activated protein kinase) signaling pathway-plant” and “phenylpropanoid biosynthesis”. As the lignin content was reduced in the SlHB8 overexpressing plant, we further analyzed the expression profile of DEGs related to lignin biosynthesis. The heatmaps revealed there were 19 DEGs differently expressed in the SlHB8 transgenic plant with 16 down-regulated in the SlHB8-ox lines which may account for the decreased lignin content (Figure 4F). 23 MYBs were found differently expressed in the SlHB8 transgenic plant including 13 down-regulated and 4 up-regulated in the SlHB8-ox lines (Figure 4E). All of these suggested that SlHB8 gene might regulate the synthesis of lignin.
Aims to narrow the range of SlHB8 regulated genes, genes with reversible expression profiles in SlHB8 overexpression and SlHB8 knock out lines were selected by overlapping the differentially expressed DEGs gene sets. The Venn diagram revealed that there were 116 DEGs with reversible expression pattern including 29 DEGs up-regulated in SlHB8-cr and down-regulated in SlHB8-ox and 87 DEGs down-regulated in SlHB8-ox and up-regulated in SlHB8-cr (Figure 5A,C). GO and KEGG functional analysis displayed these 116 DEGs were enriched in the GO terms of response to fungus, response to biotic stimulus, immune system process, and salicylic acid mediated signaling pathway which acts in the disease response pathway (Figure 5B); in the KEGG pathways of MAPK signaling pathway and plant-pathogen interaction (Figure 5D), indicating SlHB8’s role in the disease resistance. 47 out of 116 genes were related to disease resistance, among which 31 were down-regulated in SlHB8-ox and up-regulated in SlHB8-cr (Figure 5E).
Figure 5. Prediction and functional analysis of DEGs directly regulated by SlHB8. (A) Venn diagrams selecting 116 DEGs with reversible expression pattern potentially directly regulated by SlHB8. (C) Heat maps of predicted 116 DEGs directly regulated by SlHB8. (B) Significantly enriched GO terms based on the 116 DEGs; (D) Significantly enriched KEGG terms based on the 116 DEGs; (E) Heatmap of DEGs related to disease resistance.