The Drosophila melanogaster flightless I protein and its homologues in higher eukaryotes (FliI) are conserved members of the gelsolin family of actin-binding proteins. Members of the gelsolin family generally contain three or six copies of a 125-amino-acid residue gelsolin-related repeating unit, and may contain additional domains including the C-terminal villin-related ‘headpiece’ or N-terminal extensions such as the leucine-rich repeat of the FliI protein. Numerous studies including work done with mouse knockouts for gelsolin, villin and CapG support a role for the family in cytoskeletal actin dynamics. In both fruitfly and mouse, the FliI protein is also essential for early development. Recent studies indicate that supervillin, gelsolin and FliI are involved in intracellular signalling via nuclear hormone receptors including the androgen, oestrogen and thyroid hormone receptors. This unexpected role in signalling has opened a new area in research on the gelsolin family and is providing important new insights into the mechanisms of gene regulation via nuclear receptors.

Introduction

The gelsolin family of actin-binding proteins includes many proteins from a diverse range of eukaryotic organisms. Gelsolin [1], the prototype of the family, was cloned in 1986. The proteins generally contain three or six copies of a repeating unit of approx. 125 amino acid residues [2], with evidence of an ancestral gene duplication event leading to the 6 unit proteins. The known human, Drosophila melanogaster and Caenorhabditis elegans members of the family are depicted in Figure 1. There appears to have been a significant expansion of the family in the vertebrate lineage. The function of these proteins has been thought to involve the severing, capping, nucleating or bundling of actin filaments. Gelsolin itself also appears to be involved in apoptosis, cell motility and tumour suppression [1]. The crystal structures of gelsolin and of its subdomains bound to actin monomer have been solved and have led to important insights into the calcium-activated actin filament binding and severing activities of gelsolin, as well as its barbed-end capping and polyphosphoinositide-regulated uncapping activities [2].

Human, D. melanogaster and C. elegans members of the gelsolin family

Figure 1
Human, D. melanogaster and C. elegans members of the gelsolin family

The names of the proteins are shown with the gene names in parentheses. The chromosomal positions of the genes are indicated. The gelsolin-related repeating units are represented by the blue (1–3) and green (4–6) ellipses. Additional types of domains present in some members are indicated.

Figure 1
Human, D. melanogaster and C. elegans members of the gelsolin family

The names of the proteins are shown with the gene names in parentheses. The chromosomal positions of the genes are indicated. The gelsolin-related repeating units are represented by the blue (1–3) and green (4–6) ellipses. Additional types of domains present in some members are indicated.

The flightless I gene and homologues

The D. melanogaster flightless I (fliI) gene encodes a member of the gelsolin family (Figure 1) [3,4]. The protein also contains an N-terminal leucine-rich-repeat domain (Figure 1). Homologues of fliI (the protein encoded by the D. melanogaster fliI gene and its homologues) are present in C. elegans, mouse and human [3]. FliI is the most highly conserved member of the gelsolin family at the amino acid sequence level with 52% identity and 69% similarity between the C. elegans and human protein sequences. In Drosophila, severe mutations in fliI are embryonic-lethal, with development arresting at gastrulation in the absence of maternal product. The first defect seen involves irregularities in the highly ordered process of cellularization of the syncytial blastoderm [5]. Weak mutations are known that allow development of viable fertile adults that are unable to fly due to severely disordered indirect flight muscles [3,4].

Mouse knockouts for gelsolin family members

For gelsolin itself, the homozygous knockout mouse is viable and fertile, but has defects in cell motility, as might be expected for a protein involved in cytoskeletal actin filament regulation [1,6]. For villin, the homozygous knockout is also viable and fertile and there are only subtle defects in actin organization. Previously, villin had been believed to play an essential role in the morphogenesis of intestinal microvilli but, surprisingly, there is no effect on the structure of these microvilli in villin knockout mice [7,8], although the animals are more susceptible to colonic epithelial injury [8]. The knockout for CapG is also viable and fertile with indications of a role for CapG in receptor-mediated membrane ruffling and macrophage phagocytosis [9]. Overall, these mouse knockouts for gelsolin, villin and CapG have supported the concept that these proteins play a significant role in actin dynamics in vivo and have led to more detailed analysis of the exact roles of these proteins.

It has been thought that redundancy among related gelsolin family members in mammals (Figure 1) could help explain the mild knockout phenotypes. While this has not been tested extensively, gelsolin, CapG and adseverin have strikingly different but complementary expression patterns, indicating that they may have distinct in vivo functions [10]. Furthermore, studies with CapG/gelsolin double-knockout mice have established that CapG and gelsolin have distinct, non-overlapping functions, at least in macrophages [9]. A mouse knockout for FliI has also been generated by gene targeting. In this case, the homozygous knockout is embryonic lethal around the time of implantation [11]. Thus the gene is essential for early embryonic development in both Drosophila and mammals [3,11]. It was also shown that a human transgene encoding FliI fully rescues the embryonic lethality in the mouse. Although these mice completely lack the mouse FliI protein, the presence of the human gene permits development of viable, fertile mice, indicating that the human gene is appropriately expressed in mice and that the human FliI protein is capable of substituting functionally for the mouse protein [11].

The gelsolin family and NR (nuclear hormone receptor) signalling

An important recent development is the finding that some members of the gelsolin family are involved in NR-mediated signalling. Supervillin is capable of interacting in an androgen-enhanced fashion with the AR (androgen receptor) [12], and assays in mammalian cells showed that supervillin can enhance AR-mediated expression of reporter genes and can co-operate with other AR co-regulators such as ARA55 and ARA70 in this process. Evidence was also presented that supervillin interacts with other NRs including the oestrogen receptor, GR (glucocorticoid receptor) and peroxisome-proliferator-activated receptor-γ, although significant supervillin-modulated transactivation of reporter gene expression was found only with AR and GR [12]. Gelsolin itself binds to the AR and enhances its activity in the presence of androgen [13]. The androgen response element-mediated expression of reporter genes was dependent on the presence of AR agonists and co-expression of gelsolin, with some effect also on GR-responsive reporters. Gelsolin co-localized with AR during agonist-mediated AR translocation into the nucleus, further supporting a functional interaction.

The mammalian FliI protein is directly involved as a co-activator in signalling via NRs, in this case, the oestrogen and thyroid hormone receptors [14]. FliI was shown to interact directly with NRs and the NR-co-activators CARM1 (co-activator-associated arginine methyltransferase 1) and GRIP1 (GR-interacting protein 1) (Figure 2). Specific ablation of FliI in cultured cells using siRNA (small interfering RNA) effectively inhibited oestrogen-dependent reporter gene expression, and endogenous FliI was shown to be part of a large oestrogen receptor-associated co-activator complex assembled on the promoter region of an oestrogen-inducible gene. The NR co-activator function of FliI was dependent on the expression of the p160 co-activator GRIP1. FliI also interacts with the actin-related protein BAF53 (Brg- or Brm-associated factor 53) [14] and with actin itself, both components of SWI–SNF (mating type switching–sucrose non-fermenting) co-activator complexes. FliI may thus help in linking p160 coactivator complexes with SWI–SNF complexes; both of these complexes remodel chromatin by different but complementary mechanisms (histone modification and ATP-dependent nucleosome remodelling respectively). FliI is also known to interact with FLAP1/LRRFIP2 (where FLAP1 stands for FliI-leucine-rich repeat-associated protein 1 and LRRFIP2 for leucine-rich repeat in FliI-interacting protein 2) and FLAP2/TRIP/LRRFIP1/GCF2 (where TRIP stands for TAR RNA-interacting protein and GCF2 for GC-binding factor 2) and with Ras (reviewed in [4]). FLAP2 has been shown previously to play a role as a transcription repressor of several genes (reviewed in [4]). Whether these additional interactions of FliI (Figure 2) relate to the NR co-activator activity of FliI remains to be determined.

Interactions of FliI, a gelsolin family member, with proteins of the p160 and SWI–SNF co-activator complexes [14] at the promoter of an NR-responsive gene

Figure 2
Interactions of FliI, a gelsolin family member, with proteins of the p160 and SWI–SNF co-activator complexes [14] at the promoter of an NR-responsive gene

•, NR response element. Solid arrows indicate enzymic activities. Broken arrows indicate additional interactions (see text).

Figure 2
Interactions of FliI, a gelsolin family member, with proteins of the p160 and SWI–SNF co-activator complexes [14] at the promoter of an NR-responsive gene

•, NR response element. Solid arrows indicate enzymic activities. Broken arrows indicate additional interactions (see text).

Conclusion

At least three members of the gelsolin family, supervillin, gelsolin and FliI, now appear to be involved in NR-mediated signalling. It seems a safe prediction that additional members of the family, such as CapG, which exhibits significant nuclear localization [15], may be found to be involved in NR-related signalling processes, in addition to their well-recognized function in the modulation of cytoskeletal actin dynamics. The discovery of a completely unexpected role for supervillin, gelsolin and FliI in NR signalling has opened a new chapter in research on the gelsolin family as well as providing important new insights into the mechanisms of gene regulation via NRs.

Genes: Regulation, Processing and Interference: A Focus Topic at BioScience2004, held at SECC Glasgow, U.K., 18–22 July 2004. Edited by I. McEwan (Aberdeen, U.K.), B. White (Glasgow, U.K.), S. Graham (Glasgow, U.K.), S. Roberts (Manchester, U.K.), A. Sharrocks (Manchester, U.K.), D. Black (Organon, U.K.), S. Newbury (Oxford, U.K.), J. Sayers (Sheffield, U.K.) and A. Lloyd (University College London, U.K.).

Abbreviations

     
  • AR

    androgen receptor

  •  
  • FliI

    the protein encoded by the Drosophila melanogaster flightless I gene and its homologues

  •  
  • FLAP1

    FliI-leucine-rich repeat-associated protein 1

  •  
  • GR

    glucocorticoid receptor

  •  
  • NR

    nuclear hormone receptor

  •  
  • SWI–SNF

    mating type switching–sucrose non-fermenting

References

References
1
Kwiatkowski
D.J.
Curr. Opin. Cell Biol.
1999
, vol. 
11
 (pg. 
103
-
108
)
2
McGough
A.M.
Staiger
C.J.
Min
J.K.
Simonetti
K.D.
FEBS Lett.
2003
, vol. 
552
 (pg. 
75
-
81
)
3
Campbell
H.D.
Schimansky
T.
Claudianos
C.
Ozsarac
N.
Kasprzak
A.B.
Cotsell
J.N.
Young
I.G.
de Couet
H.G.
Miklos
G.L.
Proc. Natl. Acad. Sci. U.S.A.
1993
, vol. 
90
 (pg. 
11386
-
11390
)
4
Campbell
H.D.
Young
I.G.
Matthaei
K.I.
Curr. Genomics
2000
, vol. 
1
 (pg. 
59
-
70
)
5
Straub
K.L.
Stella
M.C.
Leptin
M.
J. Cell Sci.
1996
, vol. 
109
 (pg. 
263
-
270
)
6
Witke
W.
Sharpe
A.H.
Hartwig
J.H.
Azuma
T.
Stossel
T.P.
Kwiatkowski
D.J.
Cell (Cambridge, Mass.)
1995
, vol. 
81
 (pg. 
41
-
51
)
7
Pinson
K.I.
Dunbar
L.
Samuelson
L.
Gumucio
D.L.
Dev. Dyn.
1998
, vol. 
211
 (pg. 
109
-
121
)
8
Ferrary
E.
Cohen-Tannoudji
M.
Pehau-Arnaudet
G.
Lapillonne
A.
Athman
R.
Ruiz
T.
Boulouha
L.
El Marjou
F.
Doye
A.
Fontaine
J.J.
, et al. 
J. Cell Biol.
1999
, vol. 
146
 (pg. 
819
-
830
)
9
Witke
W.
Li
W.
Kwiatkowski
D.J.
Southwick
F.S.
J. Cell Biol.
2001
, vol. 
154
 (pg. 
775
-
784
)
10
Arai
M.
Kwiatkowski
D.J.
Dev. Dyn.
1999
, vol. 
215
 (pg. 
297
-
307
)
11
Campbell
H.D.
Fountain
S.
McLennan
I.S.
Berven
L.A.
Crouch
M.F.
Davy
D.A.
Hooper
J.A.
Waterford
K.
Chen
K.S.
Lupski
J.R.
, et al. 
Mol. Cell. Biol.
2002
, vol. 
22
 (pg. 
3518
-
3526
)
12
Ting
H.J.
Yeh
S.
Nishimura
K.
Chang
C.
Proc. Natl. Acad. Sci. U.S.A.
2002
, vol. 
99
 (pg. 
661
-
666
)
13
Nishimura
K.
Ting
H.J.
Harada
Y.
Tokizane
T.
Nonomura
N.
Kang
H.Y.
Chang
H.C.
Yeh
S.
Miyamoto
H.
Shin
M.
, et al. 
Cancer Res.
2003
, vol. 
63
 (pg. 
4888
-
4894
)
14
Lee
Y.H.
Campbell
H.D.
Stallcup
M.R.
Mol. Cell. Biol.
2004
, vol. 
24
 (pg. 
2103
-
2117
)
15
Van Impe
K.
De Corte
V.
Eichinger
L.
Bruyneel
E.
Mareel
M.
Vandekerckhove
J.
Gettemans
J.
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
17945
-
17952
)