Involvement of Arabidopsis HAC family genes in pleiotropic developmental processes

Abstract

With intrinsic acetyltransferase activities, CREB binding protein (CBP)/p300 proteins mediate a variety of physiological events, such as proliferation, differentiation, and apoptosis, by regulating both histones and non-histone proteins. Arabidopsis CBP-type histone acetyltransferase family proteins, HACs, have been found to influence flower by regulating the expression of Flowering Locus C. We recently reported that HAC family genes involved in the ethylene signaling pathway. Thereafter, we systematically analyzed the morphological and developmental phenotypes of all the hac mutant combinations including plant size, root, flower, leaf defects, and late-flowering. Here we reinforce the ubiquitous regulation mechanism of HAC family genes, in which HAC1 plays a dominant role with the synergistic assist of HAC5 and HAC12, whereas HAC4 slightly alleviates the influence of HAC1 and HAC5.

Keywords: :

• histone acetyltransferase
• CBP/p300
• HAC
• ethylene
• signaling
• development

This article refers to:

In mammals, CBP and p300 are very versatile proteins with intrinsic acetyltransferase activities, which participate in almost all processes of cell biology and organism development.^1-3 On the one hand, CBP/p300 modifies histones by acetylating the lysines in their N-terminal tails, whereby relaxes the chromatin structure and make chromosomal DNA more accessible to the transcription complex.⁴ On the other hand, structural and functional studies have shown that CBP/p300 regulates many nonhistone proteins as well, including the important transcription factors, p53 and MEF2.^5-7 Both in vitro and in vivo data showed that CBP/p300 acetylates multiple lysine sites in the carboxyl terminus of the p53, which is indispensable for the p53-mediated transcription of p21.⁷ Upon DNA damage, the CBP/p300 mediated acetylation of the p53 is greatly enhanced, which led to the stabilization of p53.⁸

With a broad spectrum of CBP/p300s functions, the studies on the HAC family genes in plants just show up a tip of the iceberg. In Arabidopsis, there are 5 HAC family members, namely HAC1, HAC2, HAC4, HAC5, and HAC12. The HAT activity of HAC1 was confirmed, whereas HAC2 was proved to lack of the HAT activity. Therefore, our studies were generated in terms of HAC1, HAC4, HAC5, and HAC12. There were 2 reports demonstrated that the HAC family genes promote flower by regulating the expression of the FLC.^9,10 Concurrently with the 2 groups, we found that hac1-involved mutants were significantly late-flowering compared with the WT plant. We measured the flowering time of all the hac mutant combinations grown under the long day condition (16h light/8h dark). We used the rosette leaf number, first flower bud appear time and first flower open time as the standards. The results were consistent for the 3 sets of data, following the sequence as WT (11.1 ± 1 RL), non-hac1-involved mutants (about 12.5 RL), hac1 (14.9 ± 1.6 RL), hac1hac4 (14.8 ± 1.6 RL), hac1hac12 (17.6 ± 2 RL), hac1hac12hac4 (18.2 ± 1.8 RL), hac1hac5 (21.9 ± 2.1 RL), hac1hac4hac5 (24 ± 2.8 RL). The results showed that all the hac1-involved mutants flower obviously later than WT, among which hac1hac5 and hac1hac4hac5 flower very late compared with WT.

We recently reported that HAC family genes influenced the ethylene signaling pathway.¹¹ The hac1-involved mutants were hypersensitive to ethylene both in the dark and in the light. Among all the hac mutant combinations, hac1hac5 mutant showed the most dramatic hypersensitivity to ethylene and hac1hac4hac5 mutant was the second behind hac1hac5. We subsequently found that the transcriptional levels of the ethylene-responsive genes were obviously higher in the hac1hac5 mutant than in the WT plant. Furthermore, our results showed that the ethylene synthesis inhibitor cannot release the classical ethylene “triple responses” of hac1hac5 and hac1hac4hac5 mutants, which excluded the possibility that hac mutations affect the ethylene sensitivity by operating the ethylene synthesis pathway. All the results led to the conclusion that HAC family genes were involved in the ethylene signaling pathway.

One special thought that we brought out in our previous report and that is further reinforced here, is the regulatory mechanism of the HAC family members. Knowing various functions of CBP/p300 ranging from cell growth to apoptosis, it’s not unforeseen to discover a variety of HACs’ functions. In addition to late-flowering and ethylene-hypersensitive, hac1-involved mutants displayed multiple morphological and developmental defects, such as short primary root, small plant size, short filament, and petal.^9-11 Here we show that all the hac1-involved mutants displayed small, dark green, wrinkly, and indented rosette leaf phenotypes (Fig. 1). Taken together, for all the morphological and developmental phenotypes, hac1hac5 exhibited the most severe phenotypes; hac1hac4hac5 was ranked in second place, succeeded by hac1hac12 and other hac mutants. In the case of flowering time, hac1hac4hac5 flowered a bit later than hac1hac5. Considering that hac1hac5 was very weak and hac1hac4hac5 was much healthier than hac1hac5, the flowering time between these 2 mutants was not comparable. All the phenotype analysis data together suggest that HAC1, HAC12, and HAC5 function synergistically, with HAC1 being dominant and HAC5 being the closest coordinator of HAC1; HAC4 alleviates the effects of HAC1 and HAC5.

Figure 1.hac1-involved mutants display small, dark green, wrinkly, and indented leaves. The plants were grown in the long day photoperiod (16h light/8h dark) for 25 d. Among all the mutants, the hac1hac5 mutant exhibited the most severe phenotypes, followed by hac1hac4hac5, hac1hac4hac12, hac1hac12, hac1hac4, hac1, and other hac mutants.

An interesting outcome from this study was the hac1 mutation-induced dose dependent phenotypes, which was in line with the dose dependent regulatory mechanism of CBP/p300 in the embryo development.¹¹ Focusing on the protruding gynoecia and late-flowering phenotypes of the hac1 mutant, we found that the WT background transgenic plants harboring the pHAC1:HAC1 resembled the hac1 mutant; the hac1 mutant background transgenic plants harboring the 35s:HAC1 showed more severe phenotypes than hac1 mutant; the most severe phenotypes among all the HAC1 transgenic plants occurred in the WT background transgenic plants harboring the 35s:HAC1 (Table 1). These results indicated that fine tuning of the HAC1 protein levels was crucial for maintaining the homeostasis of the plant growth and development.

Table 1. Phenotype analysis of HAC1 transgenic plants

Transgene Types	Constructs	Number of transgenic lines	Phenotype Descriptions
pHAC1:HAC1 hac1	pHAC1:HA-HAC1	16	All display WT looking.
	pHAC1:flag-HAC1	16	9 lines display WT looking, 7 lines resemble hac1 mutant.
pHAC1:HAC1 WT	pHAC1:flag-HAC1	14	5 lines resemble hac1 mutant, 9 lines display WT looking.
35s:HAC1 hac1	35s:flag-HAC1	16	12 lines display enhanced hac1 mutant phenotypes, 4 lines resemble hac1 mutant.
35s:HAC1 WT	35s:flag-HAC1	18	All lines display enhanced hac1 mutant phenotypes, among which 4 lines flower very late.

Knowing the severe morphological and developmental defects of hac mutants, it’s important to study the HAC family genes’ functions comprehensively and thoroughly. In theory, the morphological defects such as the substantial reduction of leaf and plant size would disturb the crop yield. In particular, the defective male and female gametophytes in the hac1hac5 mutant resulted in sterility. Therefore, much deeper understanding of the molecular and cellular mechanisms is required to manipulate the HAC family genes. With regard to histone targets, ChIP-on-chip could be utilized to systematically search the targets. For non-histone protein targets, yeast-2-hybrid screening based on the protein-protein interaction will be an ideal method.

Abbreviations:
HAT	=	histone acetyltransferase
CBP	=	CREB-binding protein
HAC	=	Histone

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank the Arabidopsis Biological Resources Center at the Ohio State University and Bernd Weisshaar for providing SALK and GABI T-DNA insertion lines in the Col-0 background. This work was supported by the Ministry of Science and Technology of China 863 project, Transgenic Program of China (2009ZX08009–065B), and the National Natural Science Foundation of China (31071351).