Regulation of flowering time in Arabidopsis



FIBR homepage


Flowering pathway

Gibberellin pathway



The Gibberellin transduction pathway


SPY (SPINDLY) (also: GAS1)


Cloned by Jacobsen et al. (1996), SPY is a single gene in the Arabidopsis genome. The sequence contains 18 exons and encodes a predicted protein of 914 amino-acids.

SPY is a low-abundance protein, present in the nucleus and the cytosol (Swain et al., 2002).

SPY expression is ubiquitous:

SPY is expressed in the whole plant. Its expression is higher in the apical meristem, flowers, seedlings, and root tips. The expression level is otherwise uniform in vegetative organs, but decreases with age (Jacobsen et al., 1998; Swain et al., 2002).

The expression of SPY is apparently not affected by phytohormones other than Gibberellins (NAA, BAP, ABA or cytokinins), nor by ambient temperature (22oC or 30oC) (Swain et al., 2002).

spy mutants:

The spy-1, spy-2 and spy-3 mutants were isolated after screening for the ability to germinate on the GA inhibitor Paclobutrazol (PAC) (Jacobsen and Olszewski, 1993). Additional mutants were isolated in screens for suppressors of the gai-1 and ga1-3 mutant phenotypes (Wilson and Somerville, 1995; Silverstone et al., 1997(1)).

In the WT background, the spy mutants are pale and erect, they are resistant to PAC, and flower early in short days and long days. Overall, they look like plants treated with an excess of GAs. Besides, the addition of GAs further enhances their phenotype (Jacobsen and Olszewski, 1993). All spy mutant alleles, except spy-4, are recessive and partially suppress the gai-1 (semi-dominant) phenotype. The spy mutants also partially suppress the ga1 (GA-deficient) phenotype.

However, the spy mutants also show additional phenotypes not usually associated with GA response, such as altered rosette size and internode length, indicating that SPY also affects other regulatory pathways besides flowering time (Jacobsen and Olszewski, 1993; Wilson and Somerville, 1995; Swain et al., 2001; Silverstone et al., 2001; Tseng et al., 2004).

  • spy-1 and spy-2: A single nucleotide substitution causes an in-frame deletion of 23 amino-acids in the N-terminal TPR domain, which results in altered RNA splicing and a shorter transcript (Jacobsen et al., 1996). However, spy-1 is actually a spy hy2 double mutant (Jacobsen et al., 1996), so it is not much used any more.

  • Spy-3 and spy-5: A single nucleotide substitution causes an amino-acid substitution in the C-terminal domain (Jacobsen and Olszewski, 1993; Wilson and Somerville, 1995). These two are weak alleles. spy-5 mutants  flower early in SD and LD, but the effect is more marked in SD (Wilson and Somerville, 1995; Cowling et al., 1998).

  • Spy-4 contains a T-DNA insertion in the upstream coding region of the gene. It is a strong allele that completely suppresses the gai-1 phenotype (Swain et al., 2001). It was interpreted as a null allele, because no transcripts were detected (Jacobsen et al., 1996). However, according to more recent results, the expression of the spy-4 allele does not seem altered compared to the WT. As spy-4 is still expressed and the mutant phenotype is semi-dominant. I6t is now considered to be a modification-of-function mutation, rather than a loss-of-function mutation (Swain et al., 2002).

  • The gas1-1 mutant allele (Carol et al., 1995) is a mutation affecting the SPY locus, and was later re-named spy-7. It is a weak allele, recessive, that partially suppresses that gai-1 phenotype (Peng et al. 1999(1)).

  • Other spy mutants, spy-8 to spy-17, were isolated in a screen for suppressors of the ga1-3 phenotype (fast-neutrons). All are recessive and early flowering (Silverstone et al., 1997(1)). These mutants partially complement the ga1-3 mutant phenotype.

Function and properties of the SPY protein:

SPY is a TPR protein that presents similarities with Ser/Thr O-linked N-acetylglycosamine transferases. First characterized in animals, these proteins are nuclear and cytosolar enzymes that modify proteins by post-translational modification (by glycosylation or by competing for phosphorylation sites). This may affect the protein activity, stability or localization (Jacobsen et al., 1996; Silverstone et al., 1998; Thornton et al., 1999; Robertson et al., 1998).

SPY contains ten Tetratricopeptide (TPR) motifs at the N-terminus (Jacobsen et al., 1996). In TPR proteins, the TPR domain is typically involved in protein/protein interaction. SPY also contains a C-terminal domain of unknown function. Both domains are necessary for SPY function. The TPR proteins may act as part of larger protein complexes, as well as interact with other TPR proteins (Jacobsen et al., 1996). Actually, in the plants, SPY is part of a 850 kDa complex (unpubl., see Tseng et al.,  2001).

  • The TPR domain is necessary for SPY function: The importance of the SPY TPR domain was investigated by Tseng et al. (2001): different transformed Arabidopsis lines expressing only the TPR domain under the SPY promoter flower early, and partially suppress the gai-1 gain-of-function mutant phenotype. However,  they still flower later than the spy-3 or spy-4 mutants. In these lines,  SPY expression is similar to the WT, which means that the TPR domain plays a role in the GA signal transduction (Tseng et al., 2001). As none of these lines has a phenotype as strong as the mutants (spy-4 and spy-3), perhaps the TPR domain alone is not sufficient to ensure an adequate interaction with the SPY target, or is perhaps less stable than the complete protein (Tseng et al., 2001).

  • The TPR domain is involved in the SPY-SPY interaction: The TPR domain of SPY is able to interact in vitro and in yeast 2-hybrid systems with another SPY protein (Tseng et al., 2001). Besides, the spy-2 loss-of-function mutant protein has an incomplete TPR domain, which results in a weaker SPY-SPY interaction (Tseng et al., 2001).

  • The TPR domain is also able to interact with GI: See here.

  • The SPY protein targets RGA and GAI: See here.

Plants over-expressing SPY also flower early:

Plants over-expressing SPY under the control of the 35S promoter have a phenotype similar to the spy loss-of-function mutants, and flower early in both short days and long days. However, their phenotype is weaker than the spy-4 mutant and is closer to the recessive spy mutants (Swain et al., 2001; Tseng et al., 2001).

Plants over-expressing SPY have a higher level of transcript, both in the WT and the spy-3 backgrounds (Swain et al., 2001).

Plants over-expressing only the TPR domain of SPY also flower early. Jacobsen et al. (1998) reported that over-expression of the TPR domain gives no phenotype. However, this observation was contradicted by the work by Tseng et al. (2001), where over-expression of the TPR accelerates flowering (although less than over-expression of the complete SPY sequence). In Jacobsen's experiments, the TPR construct lacked the 5'UTR fragment, and this may account for the lack of phenotype: Swain et al. (2001) showed that the 5'UTR is necessary for the proper expression of SPY, as it enhances the ability to rescue spy mutants.

Just as the spy mutants, plants over-expressing SPY can partially complement the ga1 and gai mutant phenotypes. This phenotype is consistent with an increased GA response, but seems difficult to explain. A possible explanation is that an excess of SPY protein hinders the correct formation of the protein complex (Swain et al., 2001).


SPY interaction with GI

SPY interaction with the GA response pathway

Interaction of SPY with RGA and GAI