Arabidopsis NPH1: A Flavoprotein with the Properties of a Photoreceptor for Phototropism

See allHide authors and affiliations

Science  27 Nov 1998:
Vol. 282, Issue 5394, pp. 1698-1701
DOI: 10.1126/science.282.5394.1698


The NPH1 gene of Arabidopsis thaliana encodes a 120-kilodalton serine-threonine protein kinase hypothesized to function as a photoreceptor for phototropism. When expressed in insect cells, the NPH1 protein is phosphorylated in response to blue light irradiation. The biochemical and photochemical properties of the photosensitive protein reflect those of the native protein in microsomal membranes. Recombinant NPH1 noncovalently binds flavin mononucleotide, a likely chromophore for light-dependent autophosphorylation. The fluorescence excitation spectrum of the recombinant protein is similar to the action spectrum for phototropism, consistent with the conclusion that NPH1 is an autophosphorylating flavoprotein photoreceptor mediating phototropic responses in higher plants.

Plants rely heavily on the surrounding light environment to regulate normal growth and development. Over the past two decades, considerable progress has been made in characterizing the phytochrome family of photoreceptors that monitor the red and far-red regions of the electromagnetic spectrum (1). However, only recently have advances been made that increase our understanding of ultraviolet-A (UV-A)–blue light perception in plants (2).

Cryptochromes are UV-A–blue light photoreceptors with homology to microbial DNA photolyases (2). Like photolyases, the cryptochromes contain dual light-harvesting chromophores—flavin adenine dinucleotide (FAD) and either a deazaflavin (3) or a pterin (4)—but exhibit no DNA repair activity (4, 5). The two cryptochrome genes ofArabidopsis, CRY1 and CRY2, encode homologous proteins (3, 6) that appear to overlap in function to mediate the blue light regulation of seedling development (7). CRY2 also plays a major role in floral induction (8).

Arabidopsis mutants deficient in phototropism, designated nph1 (nonphototropic hypocotyl 1), were previously shown to lack the blue light–dependent phosphorylation of a 120-kD protein associated with the plasma membrane (9). Because mutants at the NPH1locus lack all known phototropic responses in Arabidopsis, it has been hypothesized that NPH1 encodes a photoreceptor for phototropism (9). The NPH1 gene was recently isolated and found to encode a serine-threonine protein kinase (10). The NH2-terminal region of the NPH1 protein contains two copies of a motif, designated the LOV domain, present in a number of proteins from organisms including archaea, eubacteria, and eukaryotes. These include NIFL (11) and Aer (12), both of which are reported to bind FAD. The LOV domain has therefore been proposed to reflect a flavin-binding site, regulating kinase activity in response to blue light–induced redox changes (10).

To characterize NPH1 in the absence of other plant proteins, we expressed the 120-kD phosphoprotein in insect cells transfected with recombinant baculovirus containing the NPH1 coding sequence (13). Although most of the recombinant NPH1 expressed (designated BacNPH1) was insoluble, a small amount of the protein was found to be localized to the soluble fraction (Fig. 1A). The soluble protein produced was recognized by specific polyclonal NPH1 antisera (Fig. 1B). The high specificity of the antibody is demonstrated by the lack of NPH1 protein in membrane fractions isolated from the null mutant nph1-5(10). The heterologously produced protein is slightly higher in molecular weight (125 kD) than the native protein because of the presence of additional peptide sequences derived from the baculovirus expression vector (13). Ultracentrifugation and protein immunoblot analysis revealed that soluble BacNPH1 is not membrane associated. In contrast, NPH1 is associated with the plasma membrane upon isolation from Arabidopsis and several other plant species (14). Thus, although the nature of its association with the plant plasma membrane remains to be determined, this process does not appear to be operative in insect cells.

Figure 1

(A) Expression of NPH1 in insect cells. A Coomassie blue–stained SDS-polyacrylamide gel (12.5%) is shown for the total protein of insect cells expressing NPH1 (BacNPH1) or biotin carboxylase (control). Also shown are the proteins of pelleted (insoluble) and soluble fractions isolated from insect cells expressing NPH1. Cells were lysed by sonication and separated into pelleted and soluble fractions by centrifugation at 16,000g for 10 min. Molecular weight markers (MW) are shown on the left. (B) Immunoblot analysis of NPH1 protein. Growth of Columbia wild-type (WT) and nph1-5mutant seedlings and preparation of microsomal membranes were as described (9). Membrane protein (20 μg) and soluble protein (5 μg) prepared from insect cells expressing NPH1 (BacNPH1) were resolved on an SDS-polyacrylamide gel (7.5%), and the immunoblot was probed with anti-NPH1 (30). (C) Autoradiogram showing the blue light–dependent phosphorylation of BacNPH1. Membrane preparations from etiolated wild-type (WT) or nph1-5 mutant seedlings and soluble protein extracts isolated from insect cells expressing NPH1 (BacNPH1) or biotin carboxylase (control) were used for in vitro phosphorylation analysis, as described (9). All manipulations were carried out under dim red light. Samples were given a mock irradiation (D) or irradiated with blue light (L) at a total fluence of 3300 μmol m–2.

Photophysiological, genetic, and biochemical evidence suggests that NPH1 is a putative photoreceptor (9, 15) that undergoes blue light–dependent autophosphorylation (14). We therefore investigated whether NPH1 expressed in insect cells could be phosphorylated in response to blue light irradiation. Insect cells expressing NPH1 were grown in complete darkness and harvested under dim red light. Soluble protein samples were isolated and used for in vitro phosphorylation analysis. Autoradiography revealed that BacNPH1 is highly phosphorylated after a brief irradiation with blue light (Fig. 1C). No light-activated phosphorylation was detectable in soluble fractions prepared from control cells expressing biotin carboxylase. These results indicate that NPH1 is a photosensitive, autophosphorylating protein kinase. The blue light–induced phosphorylation of BacNPH1 was observed with six independently transfected cultures of insect cells. Furthermore, the relative increase in phosphorylation induced by blue light is comparable to that observed in membrane fractions isolated from etiolatedArabidopsis seedlings (Fig. 1C). Remarkably, both the fluence-response requirements and the phosphorylation kinetics of BacNPH1 strongly resemble those of the native protein in microsomal membranes (Fig. 2, A and B). In contrast, after a saturating light pulse, recovery of light sensitivity in darkness for native NPH1 is more rapid than that observed for the recombinant protein (Fig. 2C). Dark attenuation of the in vitro blue light–mediated phosphorylation has also been reported for pea (16), maize (17), and oat (18) and is believed to restore the light-sensitive phosphorylation system to its initial ground state. An explanation for the observed difference could be that someArabidopsis protein or factor involved in desensitizing NPH1 activation in the absence of light is simply lacking from insect cells.

Figure 2

(A) Fluence response of NPH1 phosphorylation in wild-type membranes (○) and soluble protein extracts prepared from insect cells expressing NPH1 (▪). (B) Kinetics of NPH1 phosphorylation in wild-type membranes (○) and soluble protein extracts prepared from insect cells expressing NPH1 (▪). (C) Effect of dark incubation on NPH1 phosphorylation in wild-type membranes (○) and soluble protein extracts prepared from insect cells expressing NPH1 (▪). Samples were irradiated and incubated on ice for the times indicated before the addition of radiolabeled adenosine triphosphate. In each case, all values are relative to dark controls and represent the average of three independent experiments. Standard errors are shown. The extent of phosphorylation was quantified with a PhosphorImager (Molecular Dynamics).

The above results support the earlier hypothesis (9) that NPH1 is a photoreceptor mediating blue light–dependent autophosphorylation. We therefore investigated whether BacNPH1 binds a cofactor that could function as a light-harvesting chromophore. Several chromophore moieties have been proposed for blue light photoreceptors, including carotene (19), flavins (20), pterins (21), retinal (22), and zeaxanthin (23). Because the insoluble form of BacNPH1 represents the majority of extractable protein from insect cells (Fig. 1A), this fraction was initially used for the analysis of potential chromophores. The cofactor was found to be noncovalently bound, as it was released by heat or acid denaturation of BacNPH1. Spectral analysis of the released chromophore revealed the pigment to be fluorescent, with excitation and emission maxima resembling those of free flavins (Fig. 3, A and B). Furthermore, the absorption spectrum of the free chromophore exhibited the two prominent absorbance peaks characteristic of flavins (Fig. 3C). These spectral characteristics were also detectable in soluble protein extracts prepared from insect cells expressing BacNPH1 but were minimal in extracts from control cells expressing biotin carboxylase (Fig. 3, A to C). (A background level of flavin fluorescence was routinely observed in soluble fractions isolated from the control cells, likely from flavins released from endogenous flavoproteins present.) The flavin associated with BacNPH1 was identified as flavin mononucleotide (FMN) by thin-layer chromatography, according to its mobility relative to FAD, FMN, and riboflavin standards (Fig. 3D). These observations confirm previous biochemical evidence suggesting that the photodetection mechanism for the blue light–dependent phosphorylation reaction requires a flavin species (24).

Figure 3

(A to C) The fluorescence excitation spectrum (A), fluorescence emission spectrum (B), and absorption spectrum (C) of the chromophore released from insoluble protein extracts prepared from insect cells expressing NPH1 (solid lines). In each case, an equal amount of protein extract from insect cells expressing biotin carboxylase (dotted line) was used as a control (31). (D) Identification of the chromophore bound to BacNPH1 as FMN. The chromophore bound to BacNPH1 was released by boiling the pelleted fraction (1 mg) in 70% ethanol and used for thin-layer chromatography as described (32) using n-butanol–acetic acid–water (3:1:1 v/v) as solvent. Retardation factor (Rf ) values for the BacNPH1 chromophore (BacNPH1C) and other flavins are shown.

Phototropism is induced by green light in addition to UV-A–blue light in Arabidopsis (9, 25). It will be interesting to establish whether the redox properties of the FMN bound to NPH1 lead to a stable semiquinone, thereby generating additional sensitivity in the green region of the spectrum, as was found for the FAD chromophore bound to CRY1 (5). Given that the fluence-response requirements and phosphorylation kinetics for BacNPH1 correspond to those of the native protein in microsomal membranes (Fig. 2, A and B), it seems most likely that FMN is also the chromophore that mediates the photoactivation of NPH1 in vivo.

Action spectra for a number of processes initiated by UV-A–blue light, including phototropism, have been described (26) and are reported to resemble the absorption spectrum of a flavoprotein. The action spectrum for phototropism shows maximal activity between 400 and 500 nm and reveals a degree of fine structure with a major band at 450 nm and subsidiary shoulders at 430 and 470 nm (dashed line, Fig. 4B). An additional broad, less effective peak is typically observed at 380 nm. We therefore examined whether a similar degree of fine structure could be detected for the insoluble form of BacNPH1. Indeed, the uncorrected fluorescence excitation spectrum for BacNPH1 displays the characteristic fine structure observed in the action spectrum for phototropism (Fig. 4A). Such fluorescent peaks were also clearly visible in soluble protein extracts prepared from insect cells expressing NPH1, but they were undetectable in extracts from cells expressing biotin carboxylase (Fig. 4A). Subtraction of the background fluorescence observed in the control gave a corrected fluorescence excitation spectrum for BacNPH1 that resembles the action spectrum for phototropism (Fig. 4B). The turbid nature of the sample used for this analysis would be expected to reduce fluorescence in the UV-A region of the spectrum and to account for the less prominent peak at 380 nm. These findings are also consistent with the hypothesis thatNPH1 encodes the apoprotein of a blue light photoreceptor for phototropism.

Figure 4

(A) Fluorescence excitation spectra of insoluble protein extracts prepared from insect cells expressing NPH1 (solid line) and biotin carboxylase (dotted line). (B) The corrected fluorescence excitation spectrum (solid line) for BacNPH1 [the difference between the fluorescence spectra shown in (A)] plotted with the action spectrum for the ascending arm of alfalfa hypocotyl phototropism [dashed line; redrawn from (33)].

Recent genetic evidence has implicated an involvement of cryptochrome in phototropism (27). An Arabidopsismutant lacking CRY1 and functional CRY2 displayed an apparent lack of first-positive phototropic curvature in response to blue light irradiation. However, cry1cry2 double mutants retained a significant degree of second-positive curvature, indicating the presence of an independent photoreception system for phototropism. Null mutants lacking CRY1 and CRY2 protein exhibited first-positive blue light– induced phototropic curvature (28), suggesting that CRY1 and CRY2 are not the primary photoreceptors mediating phototropic curvature in Arabidopsis. Thecry1cry2 double mutants also retained normal in vitro blue light–dependent phosphorylation of NPH1 (28). Thus, light-induced phosphorylation of NPH1 does not appear to result from the action of cryptochrome. Instead, it is possible that cryptochrome, like phytochrome (29), functions to modulate the response output, leading to enhanced first- and second-positive phototropic curvatures. A more detailed photophysiological characterization of phototropic responses in cryptochrome- and phytochrome-deficient mutants will aid our understanding of the phototropic detection system, which appears to involve the interaction of both red–far-red and other blue light photoreceptors. The present results, in conjunction with the established role of NPH1 in phototropism (9, 10), lead us to propose that NPH1 is an autophosphorylating flavoprotein, unrelated to cryptochrome, that serves as a photoreceptor for phototropism in higher plants.

  • * To whom correspondence should be addressed. E-mail: briggs{at}


View Abstract

Navigate This Article