Naphthylphthalamic acid associates with and inhibits PIN auxin transporters

Abstract

N-1-naphthylphthalamic acid (NPA) is a key inhibitor of directional (polar) transport of the hormone auxin in plants. For decades, it has been a pivotal tool in elucidating the unique polar auxin transport-based processes underlying plant growth and development. Its exact mode of action has long been sought after and is still being debated, with prevailing mechanistic schemes describing only indirect connections between NPA and the main transporters responsible for directional transport, namely PIN auxin exporters. Here we present data supporting a model in which NPA associates with PINs in a more direct manner than hitherto postulated. We show that NPA inhibits PIN activity in a heterologous oocyte system and that expression of NPA-sensitive PINs in plant, yeast, and oocyte membranes leads to specific saturable NPA binding. We thus propose that PINs are a bona fide NPA target. This offers a straightforward molecular basis for NPA inhibition of PIN-dependent auxin transport and a logical parsimonious explanation for the known physiological effects of NPA on plant growth, as well as an alternative hypothesis to interpret past and future results. We also introduce PIN dimerization and describe an effect of NPA on this, suggesting that NPA binding could be exploited to gain insights into structural aspects of PINs related to their transport mechanism.

Keywords: NPA; PIN; auxin transport; auxin transport inhibitor; naphthylphthalamic acid.

Fig. 1.

NPA inhibits PIN-mediated ³H-IAA efflux in oocytes. (A) In oocytes, coinjected NPA_in fully (10 µM) or partially (1 µM) inhibits ³H-IAA efflux mediated by PIN1 (+YFP:D6PK), whereas external 10 µM NPA_out added immediately to the pH-7.5 medium has no effect (ANOVA; n = 12; a–c, b–c, P < 0.001; a–b, P = 0.029). (B) In pH-5.5 medium (Left), external 10 µM NPA_out is effective at inhibiting ³H-IAA efflux mediated by PIN3 (+YFP:D6PK) after a 10-min preincubation whereas external NPA_out at pH 7.5 (Right) causes only partial inhibition after 2.5 h, compared to full inhibition by coinjected 10 µM NPA_in (ANOVA; n = 10; P < 0.0001 for all subsets). (C) Coinjected 10 µM NPA_in fully inhibits PIN3-mediated ³H-IAA efflux from oocytes independent of the choice of activating kinase, YFP:D6PK or PID (two-way ANOVA, n = 5 to 8; P < 0.001 for ±kinase as well as ±NPA subsets). (D) In [ɣ-³²P]-ATP in vitro kinase assays, 10 µM NPA does not inhibit D6PK autophosphorylation or D6PK phosphorylation of PIN1 (Upper) or PIN3 (Lower) hydrophilic loops (HL); A single asterisk indicates PIN degradation products. AR, autoradiogram; CB, Coomassie Blue gel.

Fig. 2.

NPA binds to leaf epidermal membranes overexpressing AtPINs. (A) ³H-NPA binds in a specific/saturable manner to N. benthamiana epidermal peel membranes expressing AtPIN1/2/3/6 but not to membranes from mock controls, with accumulation (open red symbols) in pellet (Upper) and depletion from SN (Lower) competed by 20 µM unlabeled NPA (filled black symbols). Addition of 20 µM IAA or BA or 3 mM ATP did not compete for binding. Comparisons by t test (PIN6, n = 3) or ANOVA (n = 3 or 4). For pellet: a–b, P = 0.03; c–d, P = 0.001; e–f, P = 0.008. For SN: a–b, P = 0.004; c–d, P = 0.003; e–f, f–g, P = 0.001; e–g, P = 0.04. (B) Relative abundance values of 3,687 endogenous N. benthamiana proteins from QMS analysis (gray cloud; log₁₀ scale; AU, arbitrary units) were compared between controls and PIN-expressing samples (Left, P19 vs. PIN1+P19; Right, GFP vs. PIN2+GFP). Highlighted N. benthamiana proteins are described in the text. Heterologous AtPIN1/PIN2, GFP, and P19 are shown as red dots; AtPIN1/PIN2 are plotted against the technical noise in their respective controls (Materials and Methods). GFP refers to GFP-KDEL (SI Appendix, Fig. S2F). (C) Relative abundance values of selected N. benthamiana proteins from QMS analysis compared across three nonbinding control (P19, GFP, KfPIN) and three overexpressing PIN (AtPIN1/2/3) samples (log₁₀ scale; AU, arbitrary units). Selected relative abundance values in B and C are in SI Appendix, Table S1.

Fig. 3.

NPA binds to oocyte and yeast membranes expressing PINs. (A) ³H-NPA binds in a specific/saturable manner to membranes from oocytes expressing PIN1 or PIN1+YFP:D6PK but not to membranes from control oocytes (injected with water instead of RNA), with an increase in pellet (red symbols, Upper) and depletion in SN (red symbols, Lower) compared to samples with 30 µM unlabeled NPA (black symbols) (t test, n = 4 to 5). All samples contained 30 µM BA. (B) ³H-NPA binds in a specific/saturable manner to membranes from yeast expressing PIN6:GFP but not to noninduced control membranes with an increase in pellet (red symbols, Upper) and depletion in SN (red symbols, Lower) compared to samples with 30 µM unlabeled NPA (black symbols) (t test, n = 3). All samples contained 15 µM BA. (C) Anti-PIN1 immunoblots using membranes from PIN1 + YFP:D6PK (D6) oocytes detect PIN1 traveling as a slower migrating species with a smear, indicative of YFP:D6PK-mediated phosphorylation, which is not seen when PIN1 is expressed alone. (D) YFP:D6PK (D6) partitioned into the soluble SN fraction (S) and was not detectable in the oocyte membrane pellet fraction (P) used for binding assays (T, total extract; antiGFP immunoblot). (E) In 35S:PIN1 suspension culture cells, 10 µM NPA treatment (1 to 3 h) did not reduce the endogenous PIN1 phosphorylation smear (antiPIN1 immunoblot). Dephosphorylated PIN1 is shown for comparison.

Fig. 4.

NPA inhibits oPDM cross-linking of PIN1 into dimers.(A) Endogenous PIN1 dimers seen in nonreducing SDS/PAGE (No DTE) are unaffected by in vivo NPA treatment of suspension cells (10 µM, 2 h; antiPIN1 immunoblot). (B) In membranes from suspension cells or oocytes, PIN1 is cross-linked into DTE-resistant dimers by oPDM in vitro, and this is inhibited by NPA (antiPIN1 immunoblot). (C) Monomers and cross-linked PIN1 dimers from oPDM treatments (± NPA) were quantified from immunoblots and dimer:monomer ratios were calculated. The ratio in oPDM+NPA was compared to oPDM+DMSO by a one-sample t test with the latter used as the test value of 100% as indicated by the dashed red line. NPA reduced the amount of oPDM-linked PIN1 dimers in both suspension cells (n = 8, 95% CI [26, 62]) and oocytes (n = 6, 95% CI [31, 75]).