GhHB12, a HD-ZIP I Transcription Factor, Negatively Regulates the Cotton Resistance to Verticillium dahliae

The homeodomain-leucine zipper (HD-ZIP) is a plant-specific transcription factor family that plays important roles in plant developmental processes in response to multiple stressors. We previously isolated a cotton HD-ZIP class I transcription factor gene, GhHB12, which is regulated by the circadian clock and photoperiodism. Furthermore, it regulates cotton architecture, phase transition, and photoperiod sensitivity. Here we report that GhHB12 was induced by methyl jasmonate (MeJA) and Verticillium dahliae infection. Additionally, stress-responsive elements were found in the GhHB12 promoter. Promoter fusion analysis showed that GhHB12 was predominantly expressed in primary roots and that it was induced by mechanical damage. Overexpression of GhHB12 increased susceptibility of the cotton plant to the fungal pathogens Botrytis cinerea and V. dahliae, which was coupled with suppression of the jasmonic acid (JA)-response genes GhJAZ2 and GhPR3. Our results suggest that GhHB12, a cotton stress-responsive HD-ZIP I transcription factor, negatively regulates cotton resistance to V. dahliae by suppressing JA-response genes.


Introduction
Plants face various environmental challenges, including invasion by microorganisms, during every stage of their life cycle. Consequently, plants have evolved intricate mechanisms to perceive external stimulation and they have developed a complex signal transduction network to regulate their adaptation to biotic and abiotic stresses. Phytohormones, such as abscisic acid (ABA), salicylic acid (SA), jasmonic acid (JA), and ethylene (ET), primarily regulate the protective responses of plants against various stressors through synergistic and antagonistic actions [1][2][3][4].
Transcription factors play crucial roles in the phytohormone crosstalk network during plant abiotic and biotic stress responses [5,6]. For example, AtMYC2/JIN1/RD22BP1, a basic helix-loop-helix (bHLH) transcription factor, is not only important in the JA signaling pathway, but also a positive regulator of the ABA-dependent drought response [7,8]. The transcription factor WRKY70, a node of convergence for SA-and JA-mediated signals for plant defense, has recently been shown to be a negative regulator during the beginning of ABA-controlled stomatal closure [9,10].
Homeodomain-leucine zipper (HD-ZIP) transcription factors, which are unique to the plant kingdom, have a leucine zipper motif immediately downstream of the homeodomain (HD) and are grouped into four classes [11]. Functional data that is available on a subset of the class I and II genes show that a number of them are involved in developmental reprogramming in response to Chr11-A (Gh_A11g0906) and Ghr11-D (Gh_D11g1052). Gh_A11g0906 was used to transform the cotton. Sequence analysis showed that Gh_A11g0906 and Gh_D11g1052 have only 6 single nucleotide polymorphism (SNP) at nucleotide level and 2 different amino acid residues at protein level. There are 23 SNP and 6 deletions between the promoters of Gh_A11g0906 and Gh_D11g1052 (see the Supplementary File).
Our previous study showed that the expression of GhHB12 (DW511497) was significantly increased by ABA and salt treatment [33]. MeJA, as an important phytohormone involved in the plant immune system, could quickly upregulate transcription of GhHB12 ( Figure 1A). However, GhHB12 transcription did not respond to SA ( Figure 1A). Moreover, GhHB12 was also upregulated by V. dahliae ( Figure 1B), a major fungal pathogen of cotton in China. These results indicated that GhHB12 was involved in the response not only to abiotic stress, but also to biotic stress, possibly by regulating phytohormone signaling networks. of two exons and one intron ( Figure S1A), which were located on the chromosome Chr11-A (Gh_A11g0906) and Ghr11-D (Gh_D11g1052). Gh_A11g0906 was used to transform the cotton. Sequence analysis showed that Gh_A11g0906 and Gh_D11g1052 have only 6 single nucleotide polymorphism (SNP) at nucleotide level and 2 different amino acid residues at protein level. There are 23 SNP and 6 deletions between the promoters of Gh_A11g0906 and Gh_D11g1052 (see the Supplementary file).
Our previous study showed that the expression of GhHB12 (DW511497) was significantly increased by ABA and salt treatment [33]. MeJA, as an important phytohormone involved in the plant immune system, could quickly upregulate transcription of GhHB12 ( Figure 1A). However, GhHB12 transcription did not respond to SA ( Figure 1A). Moreover, GhHB12 was also upregulated by V. dahliae ( Figure 1B), a major fungal pathogen of cotton in China. These results indicated that GhHB12 was involved in the response not only to abiotic stress, but also to biotic stress, possibly by regulating phytohormone signaling networks. To determine if GhHB12 upregulation by multiple stressors and phytohormones was controlled by its promoter, cotton was transformed with a construct bearing 905 base pairs (868 bp upstream and 37 bp downstream of the GhHB12 start codon site) fused to the reporter gene β-glucuronidase (GUS). Histochemical analysis showed that GUS was expressed in cotton seedling cotyledons, primary roots, tips of lateral roots, vascular tissues, and the floral organ (Figure 2A,B). These results agree with the RT-PCR results of GhHB12 expression in different cotton tissues and organs ( Figure S1B). In addition, wounding stress induces GUS expression in cotton hypocotyls and cotyledons ( Figure 2C). To determine if GhHB12 upregulation by multiple stressors and phytohormones was controlled by its promoter, cotton was transformed with a construct bearing 905 base pairs (868 bp upstream and 37 bp downstream of the GhHB12 start codon site) fused to the reporter gene β-glucuronidase (GUS). Histochemical analysis showed that GUS was expressed in cotton seedling cotyledons, primary roots, tips of lateral roots, vascular tissues, and the floral organ (Figure 2A,B). These results agree with the RT-PCR results of GhHB12 expression in different cotton tissues and organs ( Figure S1B). In addition, wounding stress induces GUS expression in cotton hypocotyls and cotyledons ( Figure 2C).
To test whether GhHB12 has any transcription activation properties, the full-length GhHB12 and different truncated GhHB12 sequences were fused to the DNA binding domain of GAL4 (GAL4-BD). The full-length GhHB12 (1-239 aa) fused to GAL4-BD produced strong transactivation activity, whereas the transactivation activity was abolished for both the N-terminal (N: 1-27 aa) and the HD and Zip (HD-LZ: 28-123 aa) domains of GhHB12 fused to GAL4-BD ( Figure 3). These results indicated that the predicted C-terminus (C: 124-239 aa) was required for transcriptional activation in yeast. In addition, the transactivation activity of the truncated GhHB12 without the N-terminal region (GhHB12-∆NT: 28-239 aa) was stronger than that of the full-length GhHB12 ( Figure 3). Thus, this indicated that the N-terminal region of GhHB12 was a transcriptional repression region. To test whether GhHB12 has any transcription activation properties, the full-length GhHB12 and different truncated GhHB12 sequences were fused to the DNA binding domain of GAL4 (GAL4-BD). The full-length GhHB12 (1-239 aa) fused to GAL4-BD produced strong transactivation activity, whereas the transactivation activity was abolished for both the N-terminal (N: 1-27 aa) and the HD and Zip (HD-LZ: 28-123 aa) domains of GhHB12 fused to GAL4-BD ( Figure 3). These results indicated that the predicted C-terminus (C: 124-239 aa) was required for transcriptional activation in yeast. In addition, the transactivation activity of the truncated GhHB12 without the N-terminal region (GhHB12-ΔNT: 28-239 aa) was stronger than that of the full-length GhHB12 ( Figure 3). Thus, this indicated that the N-terminal region of GhHB12 was a transcriptional repression region.  To test whether GhHB12 has any transcription activation properties, the full-length GhHB12 and different truncated GhHB12 sequences were fused to the DNA binding domain of GAL4 (GAL4-BD). The full-length GhHB12 (1-239 aa) fused to GAL4-BD produced strong transactivation activity, whereas the transactivation activity was abolished for both the N-terminal (N: 1-27 aa) and the HD and Zip (HD-LZ: 28-123 aa) domains of GhHB12 fused to GAL4-BD ( Figure 3). These results indicated that the predicted C-terminus (C: 124-239 aa) was required for transcriptional activation in yeast. In addition, the transactivation activity of the truncated GhHB12 without the N-terminal region (GhHB12-ΔNT: 28-239 aa) was stronger than that of the full-length GhHB12 ( Figure 3). Thus, this indicated that the N-terminal region of GhHB12 was a transcriptional repression region.

Reduced Fungal Pathogen Resistance of Transgenic Cotton Overexpressing GhHB12
To evaluate the function of GhHB12 during cotton resistance to fungal pathogens, overexpressed and RNA-interference (RNAi) constructs of GhHB12 were assembled and transferred into cotton through Agrobacterium-mediated transformation. The transgenic cotton plants were confirmed by Southern blotting, RT-PCR, and RT-qPCR [34]. The two GhHB12-overexpressing cotton lines (OE37 and OE42) and the two GhHB12-RNAi cotton lines (R17 and R19) were used to analyze resistance to the fungal pathogens B. cinerea and V. dahliae. Leaves from GhHB12 transgenic cotton lines and wild-type (WT) plants were inoculated with B. cinerea and the lesion sizes were measured three days after inoculation. Lesion sizes in the GhHB12-overexpressing lines were larger than those in wild-type plants, whereas they were smaller in the GhHB12-RNAi lines ( Figure 4A,B). To analyze V. dahliae resistance responses in wild-type and GhHB12 transgenic cotton plants, 2-week-old seedlings were inoculated with V. dahliae strain V991. The results showed that 15 days after inoculation, the GhHB12-overexpressing cotton lines presented more wilting and de-colored leaves than the wild-type plants, whereas GhHB12-RNAi cotton lines had fewer wilting and de-colored leaves ( Figure 5A). These findings were further confirmed by the brown areas at the longitudinal sections of the cotton stems near the cotyledons node after V. dahliae inoculation ( Figure 5B), the fungal recovery from the stem section of the inoculated plants ( Figure 5C), the rate of diseased plants ( Figure 5D), and the relative colonization of V. dahliae ( Figure 5E). promoter; N: N-terminal region; C: C-terminal region; HD-LZ: the HD and Zip domain; BD-GhHB12(1-239): Full-length GhHB12 (1-239 aa) fused to GAL4-BD; BD-GhHB12-ΔNT(28-239): GhHB12 without N-terminal region fused to GAL4-BD; BD-GhHB12-ΔCT(1-123): GhHB12 without C-terminal region fused to GAL4-BD; BD-GhHB12-M(28-123): GhHB12 without N-terminal and C-terminal region fused to GAL4-BD.

Reduced Fungal Pathogen Resistance of Transgenic Cotton Overexpressing GhHB12
To evaluate the function of GhHB12 during cotton resistance to fungal pathogens, overexpressed and RNA-interference (RNAi) constructs of GhHB12 were assembled and transferred into cotton through Agrobacterium-mediated transformation. The transgenic cotton plants were confirmed by Southern blotting, RT-PCR, and RT-qPCR [34]. The two GhHB12-overexpressing cotton lines (OE37 and OE42) and the two GhHB12-RNAi cotton lines (R17 and R19) were used to analyze resistance to the fungal pathogens B. cinerea and V. dahliae. Leaves from GhHB12 transgenic cotton lines and wild-type (WT) plants were inoculated with B. cinerea and the lesion sizes were measured three days after inoculation. Lesion sizes in the GhHB12-overexpressing lines were larger than those in wild-type plants, whereas they were smaller in the GhHB12-RNAi lines ( Figure 4A,B). To analyze V. dahliae resistance responses in wild-type and GhHB12 transgenic cotton plants, 2-week-old seedlings were inoculated with V. dahliae strain V991. The results showed that 15 days after inoculation, the GhHB12-overexpressing cotton lines presented more wilting and de-colored leaves than the wild-type plants, whereas GhHB12-RNAi cotton lines had fewer wilting and de-colored leaves ( Figure 5A). These findings were further confirmed by the brown areas at the longitudinal sections of the cotton stems near the cotyledons node after V. dahliae inoculation ( Figure  5B), the fungal recovery from the stem section of the inoculated plants ( Figure 5C), the rate of diseased plants ( Figure 5D), and the relative colonization of V. dahliae ( Figure 5E).  Previous studies have demonstrated that lignin plays an important role in plant responses to fungal pathogens [27]. The lignin content was estimated in wild-type and GhHB12 transgenic cotton lines. Inoculation of cotton seedlings with V. dahliae increased the total stem lignin content at 15 dpi; however, there was less lignin in the GhHB12-overexpression lines than in the wild-type. Meanwhile, slight or non-significant differences were observed between the GHB12-RNAi line and the wild-type ( Figure 5F). These results demonstrated that GhHB12 negatively regulated the resistance to B. cinerea and V. dahliae in cotton.  Previous studies have demonstrated that lignin plays an important role in plant responses to fungal pathogens [27]. The lignin content was estimated in wild-type and GhHB12 transgenic cotton lines. Inoculation of cotton seedlings with V. dahliae increased the total stem lignin content at 15 dpi; however, there was less lignin in the GhHB12-overexpression lines than in the wild-type. Meanwhile, slight or non-significant differences were observed between the GHB12-RNAi line and the wild-type ( Figure 5F). These results demonstrated that GhHB12 negatively regulated the resistance to B. cinerea and V. dahliae in cotton.

Changes in Expression of JA-Responsive Genes in GhHB12 Transgenic Cotton Plants upon V. dahliae Infection
JA and SA are important plant hormones that respond to fungal pathogens. To further validate the function of GhHB12 in JA and SA signaling responses during V. dahliae infection, the expression of JA-and SA-related defense genes in transgenic lines and wild-type plants was examined. Compared to the control treatment, the GhJAZ2, GhPR3, GhLOX2, GhERF1, and GhPDF1.2 transcripts were increased in all plants after 24 h inoculation with strain V991, but fewer GhJAZ2 and GhPR3 transcripts were found in the GhHB12-overexpressing lines than in the wild-type and GhHB12-RNAi lines ( Figure 6A). The GhWRKY70, GhPR1, and GhPR5 transcripts were induced by infection with strain V991 similar to the JA-responsive genes, but no obvious change was found between the GhHB12-overexpressing lines and the wild-type ( Figure 6B). Only the transcript level of GhPR1 was upregulated in the GhHB12-RNAi plants in relation to the wild-type and GhHB12-overexpressing lines ( Figure 6B). These results indicated that the overexpression of GhHB12 repressed some JA-responsive genes in cotton.

Changes in Expression of JA-Responsive Genes in GhHB12 Transgenic Cotton Plants upon V. dahliae Infection
JA and SA are important plant hormones that respond to fungal pathogens. To further validate the function of GhHB12 in JA and SA signaling responses during V. dahliae infection, the expression of JA-and SA-related defense genes in transgenic lines and wild-type plants was examined. Compared to the control treatment, the GhJAZ2, GhPR3, GhLOX2, GhERF1, and GhPDF1.2 transcripts were increased in all plants after 24 h inoculation with strain V991, but fewer GhJAZ2 and GhPR3 transcripts were found in the GhHB12-overexpressing lines than in the wild-type and GhHB12-RNAi lines ( Figure 6A). The GhWRKY70, GhPR1, and GhPR5 transcripts were induced by infection with strain V991 similar to the JA-responsive genes, but no obvious change was found between the GhHB12-overexpressing lines and the wild-type ( Figure 6B). Only the transcript level of GhPR1 was upregulated in the GhHB12-RNAi plants in relation to the wild-type and GhHB12-overexpressing lines ( Figure 6B). These results indicated that the overexpression of GhHB12 repressed some JA-responsive genes in cotton.

Discussion
HD-ZIP transcription factors are known to play important roles during plant development and stress responses. The HD-ZIP I proteins are generally involved in responses related to abiotic stress, abscisic acid (ABA), blue light, de-etiolation, and embryogenesis [11]. ATHB7 and ATHB12, induced by drought stress and ABA, act as negative regulators of plant tolerance to abiotic stress by repressing ABA signaling [11]. Although it has been reported that ATHB7 and ATHB12 are important factors in beet severe curly top virus (BSCTV)-induced symptom development [17], the molecular mechanism is not well documented. In this study, we demonstrated that a cotton HD-ZIP I transcription factor, GhHB12, negatively regulates GhJAZ2 and GhPR3, as well as the plant's resistance to V. dahliae. Our previous study showed that GhHB12 promoted bushy architecture and delayed flowering in cotton [34]. Therefore, we can edit the two homologues of GhHB12 in cotton using the CRISPR/Cas9 system to generate early-maturing and V. dahliae-tolerant cotton mutant lines.
Recent reports show that the JA signaling may play a positive role in the disease resistance of cotton [25,26,30,32,35] and in tomato [23] to V. dahlia. Furthermore, some reports have demonstrated that SA signaling also plays an important role in plant resistance to V. dahliae [28,36], even though SA is an antagonist of JA signaling during plant defense responses. In this study, we found that both the JA-and SA-response genes were induced in cotton roots after unimpaired root-dip inoculation with V. dahliae (Figure 6), which is consistent with the hemi-biotrophic lifestyle of V. dahliae. Though both the JA-and SA-response genes were induced by V. dahliae infection, only the expression of GhJAZ2 and GhPR3 genes in the GhHB12-overexpressing lines was consistently less than that in the

Discussion
HD-ZIP transcription factors are known to play important roles during plant development and stress responses. The HD-ZIP I proteins are generally involved in responses related to abiotic stress, abscisic acid (ABA), blue light, de-etiolation, and embryogenesis [11]. ATHB7 and ATHB12, induced by drought stress and ABA, act as negative regulators of plant tolerance to abiotic stress by repressing ABA signaling [11]. Although it has been reported that ATHB7 and ATHB12 are important factors in beet severe curly top virus (BSCTV)-induced symptom development [17], the molecular mechanism is not well documented. In this study, we demonstrated that a cotton HD-ZIP I transcription factor, GhHB12, negatively regulates GhJAZ2 and GhPR3, as well as the plant's resistance to V. dahliae. Our previous study showed that GhHB12 promoted bushy architecture and delayed flowering in cotton [34]. Therefore, we can edit the two homologues of GhHB12 in cotton using the CRISPR/Cas9 system to generate early-maturing and V. dahliae-tolerant cotton mutant lines.
Recent reports show that the JA signaling may play a positive role in the disease resistance of cotton [25,26,30,32,35] and in tomato [23] to V. dahlia. Furthermore, some reports have demonstrated that SA signaling also plays an important role in plant resistance to V. dahliae [28,36], even though SA is an antagonist of JA signaling during plant defense responses. In this study, we found that both the JA-and SA-response genes were induced in cotton roots after unimpaired root-dip inoculation with V. dahliae (Figure 6), which is consistent with the hemi-biotrophic lifestyle of V. dahliae. Though both the JA-and SA-response genes were induced by V. dahliae infection, only the expression of GhJAZ2 and GhPR3 genes in the GhHB12-overexpressing lines was consistently less than that in the wild-type and GhHB12-RNAi lines ( Figure 6). Our previous reports showed that GhJAZ2 and GhPR3 were induced by MeJA [30,32]. Accordingly, the expression of GhHB12 was induced by MeJA, V. dahliae infection, and mechanical damage, but it was not induced by SA (Figures 1 and 2C). Taken together, these results suggest that GhHB12 is involved in the responses to V. dahliae and JA signaling, but not SA signaling. There are few reports that show that HD-ZIP transcription factors are involved in plant defense responses. During biotic stress and wounding responses, HaHB4 is a positive regulator in JA biosynthesis, whereas it is a negative regulator in ET sensitivity and SA accumulation [22]. In this study, GhHB12 suppressed some JA-response genes, but did not activate SA signaling ( Figure 6). Additionally, the JA biosynthesis gene (GhLOX2) and the JA-response maker gene (GhPDF1.2) were not repressed by GhHB12 ( Figure 6A), suggesting that GhHB12 only represses some JA-response genes (GhJAZ2 and GhPR3) but not the whole JA signaling pathway in cotton.
PR3 is a chitinase that hydrolyzes chitin, which is a major component of fungal cell walls [37]. Furthermore, plant chitinases have antifungal activity in vitro and can act synergistically with β-1, 3-glucanase to inhibit fungal growth, which enhances disease resistance [38,39]. Some reports show that the expression of chitinases in cotton, tomato, and strawberry improves plant resistance to V. dahliae [40][41][42][43]. Additionally, the silencing of three cotton chitinases impaired resistance against V. dahliae [44]. Our results illustrated that GhHB12 repressed the expression of GhPR3 ( Figure 6A), thereby supporting cotton resistance against V. dahliae.
Previous studies have demonstrated that lignin plays an important role in plant responses to fungal and insect attacks [27]. In this study, we found that GhHB12 decreased the lignin content in cotton stems with and without V. dahliae infection ( Figure 5F). Gb/GhERF1-like genes directly promote the expression of GhHCT1 involved in lignin synthesis in cotton, increasing the resistance of cotton to V. dahliae [29]. There is no significant difference in the expression of GhERF1-like genes between the wild-type and the GhHB12 transgenic cotton plants ( Figure 6A), suggesting that GhHB12 decreases lignin synthesis independently of GhERF1-like genes. Further analysis of the expression of lignin synthesis genes will expand our understanding of GhHB12 functions during cotton lignin synthesis in the response to V. dahliae.
GhHB12 is a repressor during the regulation of GhPR3 and GhJAZ2. Its transactivation activity is strong (Figure 3), suggesting that GhHB12 may interact with some transcription repressors, subsequently suppressing some JA-response genes. Therefore, further characterization of protein interactions with GhHB12 will expand our understanding of the functions of GhHB12 in cotton defense responses.

Plant Materials, Growth Conditions, and Stress Treatments
The seedlings of upland cotton (Gossypium hirsutum L. cv. YZ1) were grown in Hoagland solution under 16 h light/8 h dark conditions for 2-3 weeks. The cotton seedling roots were infected with 1 × 10 7 spore·mL −1 suspension of V. dahliae spores by unimpaired root-dip inoculation for 12 h, and then transplanted into distilled water. Control plants were inoculated with distilled water rather than spores but otherwise treated in the same way as the experimental plants. Roots were harvested at different time points (0 h, 1 h, 3 h, 6 h, 12 h, 24 h, 48 h, 72 h) after 12 h inoculation for RNA extraction. For treatments with MeJA and SA, the concentrations were applied at 100 µM and 1 mM in Hoagland solution, respectively [32]. The roots were harvested at different time points (0 h, 1 h, 3 h, 6 h) after treatment for later RNA isolation.

Expression Analysis
Total RNA from both control and stressed cotton tissues were isolated using a modified guanidine thiocyanate method [45]. For RT-qPCR analyses, 3 µg of total RNA/sample were used for cDNA biosynthesis with Superscript III reverse transcriptase (Invitrogen, San Diego, CA, USA). RT-qPCR was performed with SYBR green on the 7500 Real Time PCR System (ABI, Foster City, CA, USA). The relative changes were calculated with 2 −∆Ct and the cotton GhUBQ7 gene (GeneBank: DQ116441) from G. hirsutum was amplified as the endogenous reference gene [46]. The relative transcript level was determined and normalized using the reference level and then averaged over the three technical replicates. Primers for the RT-qPCR (Supplemental Materials Table S1) were designed according to the cDNA sequences with Primer Premier 5 (http://www.premierbiosoft.com/crm/jsp/com/pbi/crm/ clientside/ProductList.jsp).

Promoter Analysis and Histochemical Assay of GUS Activity
The promoter of GhHB12 was cloned into the pGWB433 vector, which was then transferred into YZ1 by an Agrobacterium-mediated method [47]. Histochemical analysis of GUS was performed as described by Deng et al. [48]. Fresh tissue was collected from transgenic cotton plants and then immediately infiltrated with GUS staining solution at 37 • C for 6-12 h. The stained tissue was rinsed 3-5 times with 75% ethanol and then photographed under a microscope (Leica MZFLIII, Solms, Germany) or with a camera (Nikon, Tokyo, Japan).

Transcriptional Activation Activity of the GhHB12 Protein
For the transactivation assay, the full open reading frame (ORF) and the different truncations of GhHB12 were obtained by PCR using the primers listed in Supplemental Table S1. The PCR products were fused in frame to the GAL4 DNA-binding domain in pGBKT7 through a recombination reaction. The fusion vectors were used to transform the yeast strain Y2HGold with the reporter gene TRP1, according to the manufacturer's protocols. The transformed strains were streaked on plates of SD-Trp plus x-α-gal media.

Fungal Pathogen Cultivation and Inoculation
The V. dahliae strain V991 was incubated on PDA for one week and then inoculated into Czapek broth. The culture was then incubated on a shaker at 120 rpm at 25 • C for 3-4 days until the concentration of spores reached~10 8 spore·mL −1 . The liquid suspension was adjusted 1 × 10 7 spore·mL −1 with sterile distilled water for inoculation. The roots of the cotton seedlings grown under hydroponic conditions for three weeks were infected by unimpaired root-dip inoculation into the spore suspension (1 × 10 7 spore·mL −1 ) for 12 h, and then transplanted into distilled water. Control plants (mock) were treated with distilled water. The severity of the disease symptoms on each cotton seedling was scored using a 0-4 rating scale as described previously [27]. RT-qPCR of the fungal colonization was performed by comparing the DNA levels between the V. dahliae internal transcribed spacer (ITS) (as a measure of fungal biomass) and the cotton GhUBQ7 at 15 days post-inoculation in representative cotton stems sampled from above the cotyledons [49]. Stem sections sampled above the cotyledons were taken from plants after V. dahliae inoculation. They were then surface sterilized for 15 min in 70% ethanol, followed by 15 min in 10% hypochlorite and finally rinsed three times with sterile water. The sections were sliced and then transferred onto supplemented potato dextrose agar and incubated at 22 • C [50].
A Botrytis cinerea strain (stored at 4 • C) was transferred onto PDA medium and cultured for 5 days and then further incubated on fresh PDA medium for another 5 days at 25 • C. The disks of colonized agar were inoculated onto the excised leaves of 3-week-old plants at 25 • C, which were kept under a cover to maintain high humidity. Three days after inoculation, lesion sizes were measured.

Determination of Lignin Content
A representative sample of 100 mg of dry stem fine powder was extracted with 1.5 mL of 80% methanol overnight on a shaker 150 rpm at room temperature. The resulting residue was used for determination of lignin. The total lignin content was determined in triplicate from 100 mg of dry stem powder using the Klason method with modifications [51].

Conclusions
In this study, we identified a cotton homeodomain-leucine zipper transcription factor, GhHB12. This transcription factor is expressed in roots and vascular tissues, is induced by MeJA and V. dahliae infection, represses the expression of GhPR3 and GhJAZ2, and negatively regulates resistance to the fungal pathogens B. cinerea and V. dahliae in cotton.
Supplementary Materials: The following are available online at http://www.mdpi.com/1422-0067/19/12/3997/s1. Author Contributions: X.H. designed the overall study, performed and analyzed most of the experiments, and wrote the manuscript. T.W., W.Z., and Y.W. assisted with fungal pathogen cultivation and inoculation and experimental data obtainment. L.Z. revised the manuscript. All authors read and approved the final manuscript.