The role of receptor interactions in regulating ethylene signal transduction

Abstract

The phytohormone ethylene is perceived in Arabidopsis by a five-member receptor family. Earlier work has demonstrated that the basic functional unit for an ethylene receptor is a disulfide-linked homodimer. We recently reported in The Journal of Biological Chemistry that the ethylene-receptor ETR1 physically associates with other ethylene receptors through higher order interactions, suggesting the existence of receptor clusters. Here we consider the implications of such clusters upon the mechanism of ethylene signal transduction. In particular, we consider how such clustering provides a cooperative mechanism, akin to what has been found for the prokaryotic chemoreceptors, by which plant sensitivity to ethylene may be increased. In addition, we consider how the dominant ethylene insensitivity conferred by some receptor mutations, such as etr1-1, may also be propagated by interactions among members of the ethylene receptor family.

The plant hormone ethylene regulates growth and development, and is perceived by a five-member family of receptors (ETR1, ERS1, ETR2, ERS2 and EIN4) in Arabidopsis.¹ Genetic analysis indicates that ethylene receptors are functionally redundant and negatively regulate ethylene responses through interactions with the Raf-like kinase CTR1.^2–5 The functional unit of an ethylene receptor in a disulfide-linked homodimer, with each homodimer capable of binding one ethylene molecule.^6,7 However, several observations suggest that propagation of the ethylene signal through the receptors is likely to involve more than just ethylene-induced changes within individual receptor homodimers. First, Arabidopsis is amazingly sensitive to ethylene and can respond to ethylene concentrations as low as 0.2 nl/L,⁸ 300-fold lower than the K_d of the receptors for ethylene, which suggests that some mechanism exists for amplifying the input signal.^7,9 Second, ethylene-insensitive mutations in the binding sites of the receptors exhibit greater dominance than would be predicted solely from a lesion within one member of the receptor family.¹⁰

In our paper published in The Journal of Biological Chemistry,¹¹ we demonstrate that the Arabidopsis ethylene receptor ETR1 physically associates with other ethylene receptors through higher order interactions. Such physical interactions suggest that the receptors exist in plants as clusters, and that models for cooperative signaling previously applied to the histidine-kinaselinked chemoreceptors of bacteria may also be applicable to the evolutionarily related ethylene receptors of plants. In bacteria, the highly packed chemoreceptors are found in clusters at one or both poles of the cell.^12,13 Structural studies indicate that chemoreceptors can associate to form a ‘trimer of dimers’^14,15 and also support the possibility that domain swapping may occur to produce a large interconnected array of receptors. 16 Our studies indicate that ethylene receptors can interact through their cytosolic GAF domains, identifying one possible interface through which conformational changes could be propagated in an ethylene receptor cluster.

A higher-order cooperative mechanism among the ethylene receptors may explain the high sensitivity of plants to ethylene. In this model, the ethylene receptors amplify ethylene signaling by lateral signal output. Binding of ethylene to one receptor induces the conformation change of the receptor from a tense state (T) to a relaxed state (R). This conformational change is then propagated to other empty receptors in the cluster due to their physical associations with the receptor in the R state. As a result empty receptors also adopt the relaxed state (R′), resulting in amplification of the initial signal. It should be noted here that mutational evidence supports the unbound state of the receptors (T state) as being the lower energy conformation of the receptors.¹⁷ Thus, according to this model, part of the energy from ligand binding would be used to transmit conformational changes to the neighboring receptors.

An alternative model that may also explain the high sensitivity of ethylene responsiveness in plants, and one that is not necessarily incompatible with the previous model, is a conjugation model.¹⁸ Here it is hypothesized that, due to the physical proximity of the ethylene receptors, that ethylene released from one receptor then binds to another receptor rather than diffusing away. Through this conjugation mechanism, one ethylene molecule could amplify its signal by converting the conformations of multiple ethylene receptors from the ethylene-unbound state (T) to the ethylene-bound state (R). This model is based on several assumptions. One assumption is that a single ethylene molecule can bind ethylene receptors in the same cluster multiple times due to the dynamic binding of ethylene and ethylene receptor. A second assumption is that, after ethylene is released from one ethylene receptor, the recovery time for that receptor to resume the T state is longer than the time required for the released ethylene to bind to and convert another receptor from the T to the R state.

Models for cooperativity need to also explain the dominant ethylene insensitivity of various mutant receptors such as etr1-1, in which a missense mutation results in a receptor incapable of binding ethylene. Several studies indicate that the etr1-1 mutant receptor acts cooperatively to affect the signal output from other wild-type receptors (i.e., the presence of the etr1-1 receptor in its T state increases the likelihood of other receptors adopting the T state).^10,11 This observation can be most readily explained if the dominant ethylene-insensitive mutations result in a receptor that requires more energy to undergo the T to R transition than do the wild-type receptors. For example, the etr1-1 mutation may increase the stability of the T form (a T′ state). There is evidence to support this possibility. The etr1-1 missense mutation results in a receptor unable to chelate a copper cofactor necessary for ethylene binding,¹⁹ but the effects of this mutation on signaling are different from wild-type receptors that lack their copper cofactor. The etr1-1 mutant receptor appears locked in its T state, whereas wild-type receptors lacking the copper cofactor appear to be in the R state.²⁰ Thus etr1-1 is truly a gain-of-function mutation that alters the conformation of the receptor in ways not necessarily predicted from just the loss of the copper cofactor.

In conclusion, we have attempted here to provide models that can resolve an apparent contradiction in the cooperative signaling behavior exhibited by ethylene receptors. The high sensitivity of plants to ethylene suggest cooperative changes in which an R state can be propagated within a receptor cluster, but the dominance of the ethylene ethylene-insensitive mutant etr1-1 suggests that the T state can also be propagated within a receptor cluster. It should be born in mind, however, that ethylene signaling is mediated by multiple signaling components. The ethylene receptors regulate ethylene responses through interaction with and modulation of CTR1 kinase activity. Thus, the total kinase activity of CTR1 represents the signal output from the receptors. This situation is very similar to that of the bacterial chemoreceptors, which regulate the activity of an associated histidine kinase, and, as with the chemoreceptors, the stoichiometry of CTR1 interactions with the ethylene receptors and the means by which its kinase activity is regulated are important for the elucidation of the mechanism of ethylene signal transduction.

Addendum to: Gao Z, Binder BM, Chen Y-F, Chang J, Chang Y-H, et al. Plant cells oxidize hydroxylamines to NO. J Biol Chem. 2008;283:23081–23810. doi: 10.1074/jbc.M800641200.