Previous Article | Next Article ![]()
Eukaryotic Cell, April 2002, p. 241-248, Vol. 1, No. 2
1535-9778/02/$04.00+0 DOI: 10.1128/EC.1.2.241-248.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Department of Botany,1 Department of Pharmaceutical and Biomedical Sciences, University of Georgia, Athens, Georgia 306022
Received 6 December 2001/ Accepted 15 January 2002
|
|
|---|
|
|
|---|
Not surprisingly, the process of polar growth requires an intact cytoskeleton and proper vesicle transport. Pharmacological disruption of F actin prevents the polar growth of Neurospora crassa and Saprolegnia ferax (22). The actin-associated motor protein myosin has a high level of tip localization, and loss of myosin function in A. nidulans leads to large apolar cells (32, 37). Mutations in sepA, which encodes an actin-associated protein, result in abnormally large, septumless A. nidulans hyphae (20). In N. crassa, kinesin and dynein mutants grow with abnormally short, thick germ tubes (40, 41). Normal vesicle assembly is also required for polar growth. The A. nidulans
-cop-related gene sodVIC, involved in coated vesicle assembly, is essential both to establish polarity and to maintain polarity (45).
Because polar growth is coordinated with nuclear division in filamentous fungi, it is also not surprising that cell division signaling pathways are involved in polarity. A cyclin-dependent kinase mutant (phoA) and a mitogen-activated protein kinase deletion mutant (mpkA) both have abnormal germination and polarity maintenance (10). Strains with mutations in the protein phosphatase gene pphA display abnormal hyphal growth and mitotic defects (28). Temperature-sensitive (ts) mutations in N. crassa protein kinase A result in swollen cells with abnormal polarity maintenance (8). Also, in N. crassa the Ser/Thr protein kinase encoded by cot1 is necessary for polar growth (18, 47). In Penicillium marneffei, the Cdc42 (Rho) homologue encoded by cflA is required for both polarity establishment and polarity maintenance (7).
The A. nidulans swo (swollen cell) mutants are characterized by either continued isotropic growth without establishment of polarity or the inability to maintain polar growth at restrictive temperatures (35). Through a series of temperature shift experiments and genetic crosses, it was determined that the events of polarity establishment and polarity maintenance are genetically separable. swoC and swoD were found to be involved in polarity establishment, while swoA was found to be required for polarity maintenance. swoF was found to be involved in both processes.
In this study, we describe the cloning, sequencing, and further characterization of swoF (35). We show that swoF encodes an N-myristoyl transferase (NMT). NMTs are highly conserved proteins that catalyze the transfer of myristate (C14:0) from myristoyl coenzyme A (myristoyl-CoA) to the N-terminal glycine of a subset of cellular proteins (4-6). This modification increases the affinity of the target protein for membranes and has been proposed to constitute a reversible mechanism that allows the target protein to switch between membrane-bound and cytoplasmic states (5). Known targets of NMTs include protein phosphatases, protein kinases, kinase substrates, and G protein
subunits (9).
|
|
|---|
Growth of germlings and microscopic observations. Conditions for growth and preparation of germlings for observations were as previously reported (35). Microscopic observations were made by using an Axioplan microscope (Zeiss, Thornwood, N.Y.), and digital images were acquired by using an Optronics (Goleta, Calif.) digital imaging system. Images were prepared by using Photoshop 5.5 (Adobe, Mountain View, Calif.).
Complementation and plasmid recovery. A genomic library was generously provided by Greg May (University of Texas M. D. Anderson Cancer Center, Houston). This library was constructed by ligating Sau3A fragments of genomic DNA into the BamHI site of pRG3AMA1 (36). Protoplasts of AXL19 were produced and transformation was conducted by using standard A. nidulans protocols (48). Transformants were selected by assaying for restoration to pyrG prototrophy. Complementation was judged by restoration of wild-type growth at restrictive temperature (42°C). The complementing plasmid replicated autonomously (e.g., extrachromosomally) due to the AMA1 sequence (1) contained in the library vector. The complementing plasmid was recovered by transformation of Escherichia coli XL1-Blue with DNA isolated from the complemented strain and selection on ampicillin-containing medium. The restriction patterns of three rescued plasmids were compared, and those showing common insert patterns were used to retransform the swoF mutant. Two of three recovered plasmids complemented swoF and showed identical restriction patterns. One of these, p19c2, was chosen for sequencing.
Sequencing. p19c2 was transposon tagged by using a GPS-1 kit (New England Biolabs, Beverly, Mass.) and was transformed into E. coli XL1-Blue. Colonies representing individual randomly tagged plasmids were arrayed in a 96-well format. Plasmid preparation was carried out with an R.E.A.L. 96-well kit (Promega, Madison, Wis.). Label was incorporated by using Big Dye 2.0 (Perkin-Elmer Applied Biosystems, Boston, Mass.). Sequencing was done with outward-facing primers designed on the basis of the transposon provided with the GPS-1 kit. Unincorporated dyes were removed by using a DyeEx 96 kit (Qiagen, Valencia, Calif.). Sequencing in the 96-well format was performed with an ABI Prism 3700 robotic sequencer (Applied Biosystems Inc., Foster City, Calif.). Subsequent analysis was done with the programs Phred version 0.000925c and Phrap version 0.990319 for assembly and quality determination and Consed version 11.0 for sequence viewing and recovery (all three Unix-based programs are available from http://depts.washington.edu/ventures/collabtr/direct/ppccombo.htm). All sequences contained at least fourfold redundancy, with a quality rating of at least 20.
Identification of the swoF gene. The complete sequence for complementing genomic DNA was compared to sequences in the National Center for Biotechnology Information database (www.ncbi.nlm.nih.gov) by using blastX. Multiple open reading frames were found within the genomic sequence. Transposon-tagged plasmids with strategically placed transposon insertions were chosen from the plasmid array to test for complementation. Transposon-tagged plasmids were transformed into AXL19. Transformants were replica plated at permissive and restrictive temperatures. The open reading frame which, when disrupted by transposon insertion, lost the ability to restore AXL19 to wild-type growth at 42°C was identified as swoF.
Sequencing of the swoF mutant allele. Genomic DNA from AXL19 was isolated by using standard methods (39). This genomic DNA was used as a template in a PCR with the Expand High Fidelity PCR system (Roche Diagnostics, Indianapolis, Ind.). The following primers were used for PCR amplification: swoF5 BamHI, 5-CGGGATCCCGATGTCAGACTCAAAAGACTC-3, and swoF3 SpeI, 5-GGACTAGTCCCTAGAGCATAACAACGCCCA-3. Each primer contains a 20-mer based on the wild-type swoF sequence and a 10-mer to create a restriction enzyme recognition site. PCR products were cloned into the pGEM-T vector system (Promega), transformed into E. coli XL1-Blue, and selected on ampicillin-containing medium with blue-white selection. Clones were verified by restriction analysis. Three clones were sequenced by using the strategy described above. Sequences of all three clones were compared to the wild-type sequence to determine the mutant lesion. All three had the same base change.
Protein alignment. Sequences for orthologues of NMT were obtained from GenBank (http://www.ncbi.nlm.nih.gov). Protein sequence alignment was carried out by using the program GeneDoc version 2.6.001 (www.psc.edu/biomed/genedoc) with default parameters and minimal manual adjustment to align sequences with secondary structural features.
Homology modeling.
The model for the predicted A. nidulans SwoF protein, SwoFp, was prepared by homology modeling with the program Swiss PDB Viewer version 3.7b2 (http://www.expasy.ch/spdbv/). The templates used were the atomic structures determined by X-ray crystallography of the Saccharomyces cerevisiae NMT complexed with myrisoyl-CoA and a peptide analogue (Protein Data Bank accession number 2NMT) and the Candida albicans NMT (PDB accession number 1NMT). The structure of the S. cerevisiae NMT was used as the principal model as it represents a conformationally closed complex representative of the substrate-bound structure. Loops were built by using the Swiss PDB Viewer; in some cases, program O was used (24) because of its ability to merge coordinates and slightly better modeling capabilities. Poor side-chain orientations were identified by their high energies and corrected by using the optimize function in Swiss PDB Viewer. After all loops were built and side-chain conformations were corrected, the model was energy minimized by using steepest descent and conjugate gradients until the change in energy between steps was <0.05 kJ/mol and no locally poor energies remained. The model has threading energies, conformational energies, and phi-psi plots similar to those of the original crystal structures. Thus, no attempt was made to correct geometric or energetic problem areas that appeared to carry over from the atomic structures. The nomenclature for ß strands and
helices follows that used by Bhatnagar and colleagues (3). The substrate myristoyl-CoA and the peptide analogue from 2NMT were docked by using the magic-fit option. No energy minimization was required.
Growth restoration with myristate. To ascertain if myristic acid medium amendment allowed polar growth of swoF cells at restrictive temperature, A773 and AXL19 were grown either in liquid or on solid minimal medium containing 500 µM myristate and 1% (wt/vol) Brij 58 (polyoxyetheylene 20 cetyl ether) (Sigma, St. Louis, Mo.). The presence of 1% Brij 58 helps to solubilize myristate (14).
Nucleotide sequence accession number. The full genomic sequence, as well as intron locations, and the predicted protein sequence determined here have been deposited in GenBank under accession no. AY057437.
|
|
|---|
![]() View larger version (70K): [in a new window] |
FIG. 1. Wild-type, swoF, and complemented cells grown in complete medium for 16 h at restrictive temperature (42°C). (a) Wild-type strain A773. (b and d) Two morphologies of swoF strain AXL19. (c) AXL19 transformed with p19c2 containing wild-type swoF. Bar, 10 µm.
|
The genomic insert in p19c2 is 7,385 bp long (Fig. 2a). It encodes predicted proteins with high similarity to NMTs (see below), the C terminus of a tetrahydrofolylpolyglutamate synthase, and the N terminus of a ß-fructofuranosidase. The swoF mutant was transformed with plasmids containing transposon insertion disruptions in the tetrahydrofolylpolyglutamate synthase region (F10 and B11), the NMT region (E8 and B7), or the large region without a predicted open reading frame (C11). Only an insertion in the NMT region disrupted the ability of p19c2 to complement swoF temperature sensitivity (Fig. 3). Both tested transposon insertions in the NMT homologue failed to complement swoF (Fig. 3d and e); therefore, the complementing gene encodes an NMT. The A. nidulans NMT gene, swoF, is 1,639 bp long and contains three introns of 55, 56, and 49 bp, making the cDNA 1,479 bp long (Fig. 2b). Intron placement was deduced on the basis of consensus splice site sequences (2) and protein alignment with known NMT sequences. Transposons E8 and B7 are inserted in the third exon of the gene at 637 and 1,358 bp, respectively.
![]() View larger version (20K): [in a new window] |
FIG. 2. Complementing genomic DNA and swoF. (a) Schematic representation of the 7,385-bp genomic fragment that complements swoF. Three open reading frames were revealed by a search of GenBank, the full NMT gene and portions of a tetrahydrofolylpolyglutamate synthase (TPS) gene and a ß-fructofuranosidase gene. The locations of five transposon insertions are indicated by arrowheads. (b) Schematic representation of swoF (NMT), a 1,639-bp open reading frame containing three introns represented by black boxes. The light gray arrowhead shows the location of mutant lesion D369Y.
|
![]() View larger version (121K): [in a new window] |
FIG. 3. Transposon insertion verifies that the NMT region complements swoF. Colonies of swoF cells were grown at a permissive temperature (30°C) (top) or at restrictive temperature (42°C) (bottom). (a) swoF. (b to g) swoF transformed with p19c2 containing no transposon insertion (b) or transposon insertion C11 (c), E8 (d), B7 (e), F10 (f), or B11(g).
|
|
View this table: [in a new window] |
TABLE 1. Identity and similarity of SwoFp and other NMTs
|
![]() View larger version (82K): [in a new window] |
FIG. 4. Alignment of A. nidulans SwoFp (Anid), putative Nmtp from A. fumigatus (Afum), Nmt1p from S. cerevisiae (Scer), and Nmt1p from C. albicans (Calb). Black shading shows residue identity in at least three proteins. Gray shading shows similarity in at least two residues. Secondary structures are shown above the corresponding amino acid sequence as colored arrows (ß strands) or cylinders ( helices) coded blue (N terminus) to red (C terminus). The nomenclature for the secondary structures follows that used by Bhatnager et al. (3). Residues involved in binding myristoyl-CoA are indicated by gray stippled bars below the corresponding sequence, residues involved in binding the peptide substrate are indicated by red bars, and residues involved in binding both ligands are indicated by gray and red stippled bars. Tn, sites of transposon insertion disruption at K175 (E8) and at Q415 (B7). Solid black triangles below the sequence indicate residues interacting with D369. S. cerevisiae ts mutations reported by Zhang et al. (49) are indicated by a mutant residue enclosed by a black circle below the wild-type residue. The A. nidulans swoF mutation is denoted by a mutant residue (Y) enclosed by a red circle below the wild-type D residue.
|
Protein modeling.
The crystal structures for the C. albicans and S. cerevisiae NMTs have been determined with and without bound substrate peptide and myristoyl-CoA analogues (3, 16, 44). The Nmt1p fold contains a large ß sheet bisected by a pseudo-twofold symmetry axis and flanked by several
helices. The ligand binding sites are related by a pseudo-twofold axis, with the myristoyl-CoA binding to the N-terminal half of the protein and the peptide substrate binding to the C-terminal half. Using these structures as templates, we generated a homology model for A. nidulans SwoFp (Fig. 5). The N-terminal 76 amino acids are missing from the crystal structures and so are not included in our model. Most of the residues found to be important in myristoyl-CoA binding, peptide binding, or transferase activity are conserved in SwoFp (Fig. 4). Amino acids critical for structural stability are also conserved among all NMTs. The swoF D369Y mutation changes one of these structurally important residues, an aspartic acid that participates in stabilizing a turn in ß-strand ßk via hydrogen bonds and an ionic interaction (Fig. 5b).
![]() View larger version (52K): [in a new window] |
FIG. 5. Ribbon representation of the A. nidulans SwoFp homology model. (a) The SwoFp homology model lacking the first 76 residues is portrayed in a ribbon representation with secondary structure features colored blue (N terminus) through red (C terminus). The substrate molecules are shown in ball-stick representation with myristoyl-CoA colored gray and a peptide analogue colored red. The side chain of D369 projects from ß-strand ßk and is shown in ball-stick representation and colored black. Helix H is at the lower right of this view and is colored yellow. (b) Enlargement of the region near D369. The view is rotated about 90° relative to that in panel a, looking down the arrow toward D369. The side chains of T368, D369, T399, and R413 are shown in ball-stick representation. Broken lines represent hydrogen bonds. Bond distances are indicated.
|
![]() View larger version (105K): [in a new window] |
FIG. 6. Medium supplementation with myristate partially restores the growth of swoF cells. (a and c) Wild-type and swoF colonies, respectively, after 48 h of incubation in minimal medium at restrictive temperature. (b and d) Wild-type and swoF colonies, respectively, after 48 h of incubation in minimal medium supplemented with 500 µM myristate and 1% Brij 58 at restrictive temperature. (e and g) Wild-type and swoF germlings, respectively, after 16 h of incubation in minimal medium at restrictive temperature. (h and f) Wild-type and swoF germlings, respectively, after 16 h of incubation in minimal medium supplemented with 500 µM myristate and 1% Brij 58 at restrictive temperature. Bars: d (for a to d), 1 cm; f (for e to h), 10 µm.
|
|
|
|---|
Many NMT orthologues have been reported in the literature, including representatives from S. cerevisiae (15), C. albicans (46), Ajellomyces capsulatus (31), C. neoformans (30), and Homo sapiens (17). In addition, NMT orthologues from Schizosaccharomyces pombe and A. fumigatus are present in GenBank. NMT genes are single-copy genes in all fungi in which they have been characterized (30). Disruption of this single copy is lethal in S. cerevisiae (15), C. albicans (43), and C. neoformans (30). NMTs from C. albicans, C. neoformans, and H. sapiens functionally complement the lethal S. cerevisiae null allele (31). A preliminary search of the A. nidulans expressed sequence tag database (http://www.genome.ou.edu/asperblast.html) indicates only one NMT in A. nidulans. Additionally, in Southern analysis of A. nidulans genomic DNA, a swoF probe hybridizes to only one band (B. D. Shaw and M. Momany, unpublished data).
SwoFp homology model.
Using the X-ray crystal structures of NMTs from S. cerevisiae and C. albicans as templates, we homology modeled A. nidulans SwoFp (Fig. 5). Because most residues involved in substrate binding, catalysis, and structural stability among the NMTs are absolutely conserved, our model is likely very accurate. As can be seen in the template structures, the predicted SwoFp consists of two ß sheets related by a pseudo-twofold symmetry axis and flanked by
helices. Most of the ts mutations previously reported for NMTs are either in the central core or in residues very near the substrate binding sites and result in higher Km and/or lower Vmax values relative to those for the wild type (Fig. 4) (3, 49). We expected that the D369Y lesion in the swoF mutant would also fall within the core or near the pockets where myristoyl-CoA and peptide ligands bind. However, the mutated residue is at the amino-terminal end of ß-strand ßk, near the surface of the protein and on the face opposite that of the substrate binding sites. The carboxy-terminal end of ß-strand ßk interacts with the peptide substrate. Therefore, one explanation for the temperature sensitivity is that the D369Y mutation distorts the orientation of strand ßk and thus distorts peptide substrate binding. It seems more likely, however, that the D369Y mutation destabilizes the tertiary structure in the local region around strand ßk and thus has a more global effect. Figure 5b illustrates the local environment around D369. In the wild-type protein, D369 stabilizes a bend at the amino-terminal end of ßk by hydrogen bonding to T368 and bridges the side chains of neighboring secondary structures via an ionic interaction with R413 (helix
H) and hydrogen bonds with T368 and T399 (ß-strand ßi). T368 and D369 are absolutely conserved across species (49), suggesting that these residues play a critical role in NMTs. Alhough R413 and T399 are not conserved in higher eukaryotes, conservative replacements, such as K for R413 and S for T399, could maintain these interactions and substitute functionally. While modeling suggests that the bulky side chain of Y369 in the swoF mutant could be accommodated with some adjustment of the surrounding residues, the stabilizing effects of the ionic and hydrogen bonds would be completely lost. Destabilization of strand ßk in the mutant could result in lower SwoFp activity levels through conformational changes affecting substrate binding and catalytic function. Alternatively, destabilization might lead to lower steady-state levels of SwoFp either from improper folding or from increased proteolytic sensitivity. If improper folding is the main defect in SwoFp, it raises the intriguing possibility that myristate remediation acts by furnishing a scaffold around which the myrisotyl-CoA binding pocket can be folded or stabilized.
Conclusion. Polar growth is essential for the pathogenicity of A. fumigatus and many other fungal pathogens of animals and plants. Polar growth is a two-step process, and SwoFp plays a role in the first step (polarity establishment) and is absolutely required for the second step (polarity maintenance). Our current model for the involvement of SwoFp in polarity is that a target protein(s) required for polar growth must be myristoylated to be properly localized and/or fully active. At restrictive temperature, mutant SwoFp is unable to properly myristoylate its target. In this light, a recent report that an sgdD germination mutant is defective in a malonyl-CoA synthetase (36) is intriguing. This protein is involved in the formation of acetyl-CoA, a necessary precursor in the production of myristoyl-CoA. It is possible that sgdD is needed for the production of the myristoyl-CoA necessary for the myristoylation of an as-yet-unknown target responsible for the initiation of germination and polar growth.
NMTs are a highly conserved class of eukaryotic proteins; however, their substrate specificities have diverged, with fungal NMTs generally having targets different from those of mammalian NMTs (12, 13, 27, 30, 38). This target divergence, coupled with the fact that deletion of the single copy of the NMT gene in fungi is lethal (15, 30, 43), has led to considerable interest in NMTs as possible targets for a new class of antifungal drugs. This approach has met with success in C. albicans and C. neoformans, for which compounds that specifically inhibit myristoylation are being tested (11, 12, 26, 29, 42). To our knowledge, this option has not yet been explored for A. fumigatus, now the most common fungal pathogen found in autopsies of immunocompromised patients in European hospitals (19). This antifungal potential, coupled with the key role that myristoylation plays in polar growth (a critical aspect of fungal pathogenesis), is a fruitful area for further investigation.
We thank Greg May (University of Texas M. D. Anderson Cancer Center) for providing the genomic library used in this study, Xiaorong Lin (University of Georgia) for assistance in sequencing and crosses to create the swoF strain used in this study, and Greg Derda (University of Georgia) for computer assistance.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»