The mitochondrial connection

Abstract

Arg catabolism to cytoplasmic urea and glutamate is initiated by two mitochondrial enzymes, arginase and ornithine aminotransferase. Mutation of either enzyme leads to Arg sensitivity, and at least in the former, an arginine-induced seedling morphology similar to exogenous auxin treatment. We reported that single mutants lacking either of two arginase isozymes exhibited more NO accumulation and efflux, and increased responses to auxin (measured by DR5 reporter expression and auxin-induced lateral roots). We discuss evidence for stimulation of NO by arginine, either directly, or via polyamines derived from arginine. We favor the “direct” route because mitochondria are sites of NO ‘hot spots,’ and the location of arginine-degrading enzymes and the NO-associated protein1. The polyamine “branch” invokes more than one cell compartment, at least two intermediates (polyamines and H₂O₂) between Arg and NO, and is not consistent with enhanced lateral root formation in arginine decarboxylase mutants. Genetic tools are at our disposal to test the two possible routes of arginine-derived NO.

Arg Degradation and Arg-Derived NO

We suggested that increased mitochondrial NO resulted from blocking the mitochondrial Arg degradation pathway leading to localized increases in Arg or an Arg derivative.¹ We focused on mutants with reduced activities of ARGAH which catalyzes conversion of Arg to urea and Orn. Funck et al.,² examined mutants disrupted in the single structural gene for the next enzyme in the Arg degradation pathway, mitochondrial δOAT. The oat mutant could not use either Arg or Orn as N sources. And, oat showed sensitivity to Arg, as do the argah mutants, even in the presence of glutamine N source.

Our case for Arg involvement in NO generation was that argah1 and argah2 showed increased NO accumulation and efflux and enhanced propensity for LRF, especially in the presence of naphthalene acetic acid. Auxin induction of LRF was blocked by the NO scavenger cPTIO. In agreement, the auxin-responsive promoter, DR5, was more active in the argah mutants. Finally, argah1 and argah2 seedlings exhibited an auxin-like phenotype in the presence of 3 mM Arg. They resembled seedlings exposed to the natural auxin, IAA (Fig. 1), or an IAA precursor.³

Figure 1.

Root and seedling morphology in response to Arg and auxin. wt plants grown on 0.1 µM NAA exhibit enhanced lateral root formation,¹ a phenotype essentially identical to spe1 and spe2 point mutants with reduced levels of ADC.²³ Exposure to higher levels of natural auxin IAA results in dwarf seedlings with elongated petioles and curled cotyledons in addition to thickened and branched roots,³ morphology similar to that of argah1 and argah2 grown in the presence of 3 mM Arg.¹

The Mitochondrial Connection in Arg Conversion to NO

A plant ortholog of the animal NOS is controversial. NOS catalyzes the oxidation of Arg to NO and Cit. A putative Arabidopsis NOS (NOS1)⁴ was subsequently renamed NOA1 (NO-Associated1)⁵ since it exhibited no structural or enzymatic similarity to animal NOS. However, the noa1 mutant produces less NO and lacks other NO-mediated responses such as ABA-induced stomatal closure4 and bacterial lipopolysaccharide elicitation of a series of defense genes in the “innate immunity” response.⁶ NOA1 is mitochondrial, and the authors identified mitochondria as an NO source, one reduced in noa1.⁷

NOA1, ARGAH^1,8 and δOAT² are mitochondrial. It is worthy to assess the interplay among these proteins, including the basic amino acid mitochondrial inner membrane carriers, BAC1,⁹ and BAC2,¹⁰ in the conversion of Arg to NO (via ‘NOS’) or to Glu (via ARGAH/δOAT²). Given that blocking Arg degradation leads to more NO, the crucial question is whether this NO comes directly from Arg. Both Flores et al.,¹ and Funck et al.,² reported that Arg levels were lowered, contrary to expectation. However, in the presence of 5 mM Arg in the growth medium, Arg pools were elevated 6 to 7-fold over wt in oat1.² No mitochondrial Arg determinations were done in either mutant.

The Case for Plant NOS

Corpas et al.,¹¹ assayed an Arg-dependent NO-producing activity in extracts of pea seedlings. NO was detected by chemiluminescence of its adduct with ozone, and NO production was inhibited by L-NAME, an effective inhibitor of animal NOS. Corpas et al.,¹² cited 12 reports in various plants, of Arg-dependent production of Cit or NO, the latter detected by ozone chemi-luminescence or by spin trap EPR. While no plant NOS structural gene has been reported, it took sophisticated genome searches for conserved catalytic motifs to identify other genes, such as that of the putative NO target, guanylyl cyclase.¹³

Given binding sites in animal NOS for Arg, Heme, BH₄, CaM, FMN, FAD and NADPH¹⁴ it would not surprise were plants found to use separate proteins in a functional complex to produce NO. Fatty acyl synthase is one example of a ‘super-peptide’ in animals¹⁵ versus a protein complex in plants.¹⁶ However, to our knowledge, there is no current biochemical proof of the direct conversion of Arg to NO, such as the conversion of [guanido-¹⁵N]Arg to ¹⁵NO, or of [guanido-¹⁴C]Arg to [¹⁴C]Cit.

A Weaker Case for PA Involvement in Arg-Induced NO Generation

A causal effect of Arg on NO production could be mediated via Arg-derived PAs. Short of a direct oxidation of PAs to NO, the model calls for PA-mediated increases in H₂O₂ that, in turn, stimulates NO production via NR. Our plants were grown in the presence of 9 mM nitrate. Its reduced product, nitrite, can be further reduced to NO by cytoplasmic NR, via a reported membrane- bound nitrite-NO reductase and by electronic transport in the mitochondrion and chloroplast (reviewed by Neill et al¹⁷). According to this scenario, increased Arg pools lead to increased production of PAs via ADC—the only path to PAs since Arabidopsis has no ODC.¹⁸ PAs, in turn, generate H₂O₂ via localized action of PAOs that generate H₂O₂.^19,20 In Arabidopsis guard cells ABA causes increased H₂O₂, responsible for NO generation, specifically, via the NIA1 NR isozyme, by an as yet unknown mechanism.²¹ In contrast, Zeier et al.,²² proposed that during the hypersensitive response, NO increases H₂O₂ levels by inhibiting H₂O₂ scavenging enzymes. Moreover, ADC-deficient mutants spe1 and spe2 show enhanced LRF²³ (Fig. 1), further militating against Arg acting via the “PA scheme” (Arg → PA → H₂O₂ → NO) (Model, Fig. 2).

Figure 2.

Model for metabolic fates of Arg. The catabolic route, initiated in mitochondria, is via ARGAH and dOAT, resulting in export of P5C that is converted to cytosolic glutamate.² The only route for polyamine production in Arabidopsis is via ADC,¹⁸ which is not localized to the mitochondrion (SUBA database²⁶). There is no reported in vitro activity for direct oxidation of polyamines to NO. Polyamine oxidases (DAO and PAO) generate H₂O₂, either in the apoplast or peroxisome.²⁰,²⁵ The relationship between H₂O₂ and NO is complex. In guard cells H₂O₂ stimulates production of NO via a cytoplasmic NR (dotted arrow)²¹ and NO can conceivably inhibit heme-containing catalases, thereby increasing H₂O₂ levels (hence, the double-arrow between H₂O₂ and NO). The complexity of any pathway to NO production from Arg via polyamines leads us to favor a direct Arg to NO conversion. Increased LRF in low-ADC mutants, spe1 and spe2,²³ is also inconsistent with polyamine generation of NO.

Although Tun et al.,²⁴ showed PA stimulation of NO production without lag in Arabidopsis, explaining Arg-induced increase in mitochondrial NO by the “PA scheme” needs to accommodate the non-mitochondrial locations of PAO's (cell wall²⁰ and peroxisome²⁵) and of ADC.²⁶ In situ localization of root H₂O₂ indicates that it is involved in growth restriction and root hair formation.²⁷ Thus, given the complexity of ROS signaling, the explanation for the auxin-like phenotype observed in WT exposed to 3 mM H₂O₂,²⁸ needs further experimental data.

One Model, Two Routes to NO

Of the two routes of Arg to NO we opt for direct conversion via a NOS-like activity (Model, Fig. 2). Blocking Arg conversion to Glu in argah mutants (and, we predict, in oat) leads to more available Arg for NOS. That lowering ADC activity (spe1 and spe2) stimulates LRF goes against a PA role in NO production. Analyses of the oat and argah mutants will need to include PA levels in response to Arg, and the potential reversal of auxin-like effects on seedling development by genetic and pharmacological inhibition of NO accumulation and action.

Abbreviations

ABA

abscisic acid

ADC

arginine decarboxylase

Arg

arginine

ARGAH

arginase (Arg amidohydrolase)

BH4

tetrahydrobiopterin

CaM

calmodulin

cPTIO

carboxy-2-phenyl-4,4,5,5-tetramethylimidazolinone-3-oxide-1-oxyl

DAO

diamine oxidase

EPR

Electron paramagnetic resonance

IAA

indole acetic acid

NAA

naphthalene acetic acid

L-NAME

N^G-nitro-L-arginine methyl ester

LRF

lateral root formation

NO

nitric oxide

NOS

nitric oxide synthase

NR

nitrate reductase

δOAT

ornithine-d-aminotransferase

ODC

ornithine decarboxylase

Orn

ornithine

P5C

pyrroline 5-carboxylate

PA

polyamine

PAO

polyamine oxidase

wt

wild type

Addendum to: Flores T, Todd CD, Tovar-Mendez A, Dhanoa PK, Correa-Aragunde N, Hoyos ME, Brownfield DM, Mullen RT, Lamattina L, Polacco JC. Arginase-negative mutants exhibit increased NO signaling in root development. Plant Physiol. 2008;147:1937–1946. doi: 10.1104/pp.108.121459.