LATE ELONGATED HYPOCOTYL potentiates resistance conferred by CIRCADIAN CLOCK ASSOCIATED1 to aphid by co-regulating the expression of indole glucosinolate biosynthetic genes

ABSTRACT

CIRCADIAN CLOCK ASSOCIATED1 (CCA1) and LATE ELONGATED HYPOCOTYL (LHY) are core components of the circadian clock in Arabidopsis thaliana that impacts plant response to biotic stresses. Their clock-regulating functions are believed to be partially redundant, and mutation of either gene leads to shortened periods of the circadian cycle. Our recent study has demonstrated that CCA1 promotes plant resistance to the green peach aphid (Myzus persicae) through modulation of indole glucosinolate biosynthesis, but the role of LHY remains to be elucidated. Here we showed that, similar to cca1-11, single mutant lhy-21 became more susceptible to aphid infestation. Damage to the cca1-11 lhy-21 double mutant by aphids was most pronounced, indicating that the defensive roles of CCA1 and LHY were not entirely redundant. Also, the cyclic expression pattern of key indole glucosinolate biosynthetic genes was considerably disturbed in both single mutants and this was more severe in the double mutant. Apparently, both CCA1 and LHY were necessary for circadian-regulated indole glucosinolate biosynthesis. Taken together, LHY-CCA1 coordination in transcriptional regulation of indole glucosinolate biosynthetic genes most likely contributed to plant defensive capacity against aphids.

The circadian clock is an endogenous time-keeping machinery occurring in most living organisms. The clock exhibits a 24-h rhythm and provides temporal coordination of biological pathways with the external environment. CIRCADIAN CLOCK ASSOCIATED1 (CCA1) and LATE ELONGATED HYPOCOTYL (LHY) are among the core components of the central oscillator in Arabidopsis thaliana.¹ These two transcription factors have strong sequence homology within their Myb domains¹ and share a similar expression pattern, peaking at dawn and bottoming out around dusk.² Constitutive expression of CCA1 (CCA1-OX) or LHY (LHY-OX) results in arrhythmicity in clock-regulated genes, morphological changes and altered stress responses.^3,4 Both of the loss-of-function single mutants, cca1 and lhy, exhibit shortened rhythmic periods and the double mutant has an even shorter period than either single mutant.⁵ It is believed that CCA1 and LHY are partially redundant in the maintenance of the circadian rhythm.

Recent studies, including ours, have uncovered the important role of Arabidopsis CCA1 in plant response to various pathogens and pests. The arrhythmic CCA1-OX transgenic plant confers resistance to Hyaloperonospora arabidopsidis, but susceptibility to Pseudomonas syringae and Botrytis cinerea.^5–7 Contrasting responses to herbivorous insects have also been observed in CCA1-OX; while susceptible to the cabbage looper (Trichoplusia ni), it suppresses feeding activity of the green peach aphid (Myzus persicae) and inhibits its growth and reproduction.^8–10 We further discovered that CCA1 promotes plant resistance to aphids by transcriptional activation of key indole glucosinolate biosynthetic genes.¹⁰ By contrast, our understanding of the defensive role played by LHY against insects is rather limited, despite the fact that weakened plant defense has been observed in lhy single mutant upon aphid infestation, similar to that of cca1.¹⁰ The underlying resistance mechanism mediated by LHY and the potential CCA1-LHY interaction remain to be explored.

To begin to address these issues, we crossed single mutant lines cca1-11 (CS9378) and lhy-21 (CS9379) (from the Arabidopsis Biological Resource Center stock center, Columbus, OH) to generate the cca1-11 lhy-21 double mutant as well as the wild type. Subsequently, we performed aphid bioassays with these lines under constant light as described previously.¹⁰ Eight, age-synchronized (within 24 h) second-instar nymphs were transferred to 3-week-old plants. Several days later, these nymphs became adults and started to reproduce. The number of progeny on each plant was counted 7 days after infestation. Meanwhile, adults from seedlings of the same genotype were pooled and weighed in groups of 20. As shown in Figure 1, nymphs reproduced were comparable among all genotypes (Figure 1a). The same was true for the average adult body weight (Figure 1b). This is consistent with our previous report that clock mutants don’t have major impact on aphid performance. We also conducted plant damage assays: since 3 to 4-week-old seedlings of double mutant cca1-11 lhy-21 were dwarf (Figure 1c), assays began 2 weeks after planting, when the sizes of all genotypes were comparable. Considering the retarded growth of the double mutant, aphids applied to cca1-11 lhy-21 was half of those to the rest of genotypes. By 13 days post infestation, stunted growth and chlorosis became very apparent in all tested lines, with the order of severity as cca1-11 lhy-21 > cca1-11 ≈ lhy-21 > wild type (Figure 1c). Measurements on plant weight (Figure 1d) and chlorophyll content, an indicator of aphid feeding-induced chlorosis (Goggin FL. 2007) (Figure 1e) were in agreement with visual observation, suggesting that CCA1 and LHY are not completely functionally redundant. Interestingly, when older seedlings (i.e. 3-week-old rather than 2-week-old) were infested, lhy-21 did not show a level of chlorophyll significantly lower than the wild type,¹⁰ suggesting that LHY-mediated resistance to chlorosis was developmental stage-dependent. Taken together, CCA1 and LHY are both required for a full scale of defense displayed by wild type plants. They likely work coordinately to minimize damage inflicted by aphids.

Figure 1.

LHY potentiates Arabidopsis resistance to aphids conferred by CCA1. Three-week-old Ws, cca1-11, lhy-21, and cca1-11 lhy-21 plants were transferred to constant light (LL) for 24 h. Eight age-synchronized second instar nymphs were transferred to each plant and reared for 7 days. The number of progeny per plant (a) and average adult body weight (b) were documented. To evaluate plant damage symptom in response to aphids, 2-week-old Ws, cca1-11, lhy-21, and cca1-11 lhy-21 were infested by 10, 10, 10, and 5 adult aphids, respectively after 24 h LL. Shown are representative images of uninfested plants at day 0 and day 13 time points, as well as infested plants on day 13 (c). Infested plant weight (d) and chlorophyll content (e) are presented as percentages of uninfested controls. All experiments were repeated three times. Data (means ± SE) on reproduction (n = 24), insect weight (n = 9), chlorophyll content (n = 18) and plant weight (n = 18) were analyzed by one-way ANOVA and Duncan multiple range test. Means with different letters were significantly different (P < .05)

Indole glucosinolates are important anti-aphid secondary metabolites in Arabidopsis.^11,12 Cytochrome P450s, i.e. CYP79B2, CYP79B3 and CYP83B1 and a transcription factor MYB34 are key enzymes or regulator involved in biosynthesis of indole glucosinolates.^13,14 We previously showed circadian expression patterns of these genes in wild type plants.¹⁰ In CCA1-OX line however, their expression became arrhythmic and constitutively high, leading to increased production of indole glucosinolates. Conversely, their rhythmic expression was drastically altered in cca1, consistent with significantly reduced indole glucosinolate contents.¹⁰ Given the partial functional overlap between LHY and CCA1, we examined expression patterns of CYP79B2, CYP79B3, CYP83B1, and MYB34 in lhy-21 and cca1-11 lhy-21 in addition to the wild type and cca1-11. In this study, 3-week-old seedlings were transferred to constant light (LL) for 26 h and harvested every 4 h for the next 24 h period. Samples were then subjected to total RNA extraction and reverse transcription-quantitative PCR (RT-qPCR) analysis. Data were inspected by the Cosinor.Online analysis for circadian oscillation via model fitting.^15,16 As expected, all genes exhibited circadian rhythmic expression in the wild type (P < .05) (Figure 2). Such pattern was disrupted in all mutants, with the exception of CYP79B2 (P= .046) and CYP83B1 (P= .006) in cca1-11. The most severe distortion was seen in the cca1-11 lhy-21 double mutant (Figure 2). These data indicated that both CCA1 and LHY were involved in transcriptional regulation of indole glucosinolate biosynthesis. It should be noted that basal indole glucosinolate levels in the wild-type must be considerably low because of its insignificant effect on aphid performance relative to that of mutants (Figure 1). However, impaired plant defense response in mutants, especially in cca1-11 lhy-21, was presumably due at least partly to the disruption of expression of indole glucosinolate biosynthetic genes.

Figure 2.

Co-regulation of indole glucosinolate biosynthesis-related gene expression by CCA1 and LHY. Three-week-old Ws, cca1-11, lhy-21, and cca1-11 lhy-21 were transferred to LL for 26 h, then harvested every 4 h for the next 24 h. Samples were subjected to total RNA extraction and reverse transcription-qPCR analysis. Relative expression levels of CYP79B2, CYP79B3, CYP83B1, and MYB34 were plotted in a Log₂ scale. Arabidopsis Polyubiquitin 10 (UBQ10) was used as an internal control gene. Experiments were performed two times and shown is one representative. Asterisk-marked individual curves indicate the circadian gene expression pattern in certain genotypes (P < .05) by the Cosinor.Online analysis (https://cosinor.online/app/cosinor.php). Subjective day, light gray bar; subjective night, dark gray bar

Previous studies have shown that CCA1 and LHY form homo- and heterodimers and that different dimerization could affect regulation of clock functions and freezing tolerance.^1,17,18 Presumably, the same mechanism may explain the expression patterns of indole glucosinolate biosynthetic genes and corresponding defense responses: In wild type plants, coordination of homo- and heterodimers could lead to a proper circadian expression pattern, necessary to confer temporal defense against aphids. In single mutants, homodimers could confer some regulatory activity and maintain certain levels of rhythmic oscillations due to partial functional redundancy of the two proteins, but lack of the heterodimer could impact their DNA binding affinity and specificity. Loss of both transcription factors in the double mutant would result in dramatic reduction of indole glucosinolates and seriously compromised plant defense. In support of this notion, we analyzed promoter sequences of CYP79B2, CYP79B3, CYP83B1, and MYB34 genes by Arabidopsis Motif Scanner.¹⁹ CYP79B2, CYP79B3 and MYB34 contained putative CCA1/LHY-binding sites (Table 1), making direct transcriptional regulation by CCA1 and/or LHY possible. While some of the putative cis-elements were common for CCA1 and LHY, such as Evening Elements (EE) and CCA1-binding sites (CBS), G-box and NAC transcription factor-binding site (NAC55) in CYP79B3 and MYB34 promoters were CCA-specific, whereas abscisic acid-responsive element (ABRE) in MYB34 promoter was LHY-specific (Table 1).^20,21 Although CYP83B1 lacked CCA1/LHY-associated cis-elements, CCA1 and LHY apparently affected its rhythmic expression (Figure 2). Most likely, CCA1 and LHY regulated CYP83B1 transcripts through MYB34, a known activator of CYP83B1.¹³

Table 1.

Putative CCA1 and/or LHY-associated cis-elements in promoters of indole glucosinolate biosynthetic genes

Locus	Gene name	CCA1-binding site (CBS) (CCA1/LHY targeting)	Evening element (EE) (CCA1/LHY targeting)	G-box (CCA1 specific)	NAC transcription factor-binding site (NAC55) (CCA1 specific)	Abscisic acid-responsive element (ABRE) (LHY specific)
AT4G39950	CYP79B2	−818 − 811 (-)a	−192 − 184 (-)
AT2G22330	CYP79B3	−790 − 783 (-)
		−925 − 918 (-)			−62 – 57 (+)
AT4G31500	CYP83B1
AT5G60890	MYB34	−990 − 983 (-)		−215 − 210 (+)		−215 − 208 (+)

^a(+), positive strand; (-) negative strand.

Alternative to direct inactivation of indole glucosinolate biosynthesis, extreme susceptibility of the double mutant could result from more dramatically disrupted circadian rhythm than single mutants. Since CCA1 and LHY also possess distinct function in controlling rhythmicity,² clock-dependent defense could be further compromised when both genes were knocked out.

In conclusion, we demonstrated that LHY also played an important role in Arabidopsis defense. Resistance to aphid herbivory was achieved at least in part through its circadian co-regulation of indole glucosinolate biosynthetic genes with CCA1. Future studies on potential molecular interaction between CCA1 and LHY, i.e. CCA1 and/or LHY homo- and hetero-dimerization in control of indole glucosinolate synthetic gene expression will be of an interesting research project.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.