A Prolactin-Inducible T Cell Gene Product Is Structurally Similar to the Aspergillus nidulans Nuclear Movement Protein NUDC

Clone 15 ( c15 ) was originally identified as a PRL-inducible gene in activated T cells. Sequence analysis of c15 revealed that the last 94 amino acids of c15 are 68% identical and 78% similar to the filamentous fungus Aspergillus nidulans nuclear movement protein NUDC. The identification of the mammalian (rat) c15 protein suggests that the car-boxy-terminal NUDC-like region has been con-served over evolution for an important structure and/or function. To determine whether c15 is functionally analogous to NUDC, complementation studies were performed using the inducible/re-pressible pAL5 vector system. The results of the complementation experiments show that the full-length mammalian c15 protein is not only capable of rescuing the nuclear movement defect of the nudC3 mutants, but is also able to restore the expression of the downstream endogenous NUDF protein to near wild type levels. These results indicate that rat c15 and fungal NUDC not only share similar structures, but also serve similar functions. Taken together, the structural and functional con-servation between c15 and NUDC is consistent with the notion that c15 is the rat homolog of nudC and has therefore been given the name RnudC . (Molecular Endocrinology 11: 229–236, 1997)


INTRODUCTION
When T cells come in contact with antigen-presenting cells or antigen-bearing target cells, engagement of cell surface receptors results in the expression of a number of activation genes that together set the genetic program leading to T cell proliferation, differentiation, or apoptosis (1,2). In addition to the induction of gene expression in activated T cells, another early activation event involves a massive reorganization and polarization of the cytoskeleton that reorients the centrosome microtubule-organizing center (MTOC), its associated microtubules (MTs), and the Golgi apparatus toward the site of cell contact (3). This reorganization optimally positions the cell's MT-dependent vesicle transport system in alignment with the target cell to serve directed protein transport between cells, an important feature of T cell function. Understanding how T cells correctly position their nuclei and associated organelles during activation and proliferation is a critical step to a better understanding of the overall life cycle and effector functions of T cells.
Reorientation and movement of intracellular organelles, in particular, the centrosome MTOC, the nucleus, and the Golgi, depends on MT interactions with the cell cortex and is coordinated by forces acting on the centrosomes themselves (4)(5)(6). Perhaps the best evidence that there are forces acting directly on the centrosomes comes from studies of embryogenesis in Drosophila melanogaster. In the Drosophila embryo, nuclei migrate to the cortex before cellularization (7). If nuclear division is aborted, either by drug action or mutation, the centrosomes nevertheless continue to multiply and migrate to the cortex independently of the nuclei (8,9). This provides strong evidence that nuclear migration is mediated through a force acting directly on the centrosome itself, which in turn serves to move the nucleus.
In simple eukaryotes such as fungi and yeast, nuclear movement is a MT-dependent process (10,11). The main MT-dependent "motor" involved in centrosomal and nuclear positioning has been identified as cytoplasmic dynein. The evidence that cytoplasmic dynein is the main motor protein for nuclear positioning comes from the studies of the nudA, ro-1, and DYN1 genes, which encode the cytoplasmic dynein heavy chains of Aspergillus nidulans, Neurospora crassa, and Saccharomyces cerevisiae, respectively. Mutations in the nudA gene and the ro-1 gene cause a failure of nuclear movement through the germ tubes (12,13). In A. nidulans, the dynein heavy chain has been localized to the tip of the growing germ tube (14). A mutation in the 8-kDa A. nidulans dynein light chain, encoded by the nudG gene, causes the heavy chain to be lost from the tip and also affects nuclear movement (15). Similarly, disruption of DYN1 causes a partial failure of nuclear segregation between mother and daughter cells in yeast. This failure to segregate is caused by a defect in the ability of mitotic nuclei to migrate to the correct orientation before anaphase (16,17). Additionally, in yeast, MTs attached to the spindle pole body (SPB), the centrosome equivalent in lower eukaryotes, have been shown to be essential for the nuclear orientation process (18). Furthermore, several of the N. crassa ropy mutations have been shown to affect dynactin, a complex that is required for the coupling of dynein to its cargo, and result in a nuclear movement defect (13,19,20). In yeast, mutations have also been identified that affect nuclear positioning (21)(22)(23). Together these data support a model in which SPB orientation and nuclear movement are mediated via an interaction between SPB MTs and a dynactinassociated, minus end-directed dynein motor located at the cell cortex.
In filamentous fungi, e.g. A. nidulans and N. crassa, nuclear movement is required for normal colony formation (13,24). During the germination of conidia (asexual spores), the parental nucleus divides and the daughter nuclei move out in the germ tube of the developing germling. Temperature-sensitive mutations have been identified in A. nidulans that affect this initial stage of nuclear movement as well as A. nidulans' ability to maintain nuclei of additional nuclear divisions at a uniform distance from each other. Because of the observable defect in nuclear distribution, this class of mutants was call nud's (24). In addition to nudA and nudG, two other genes have been identified, nudC and nudF (12,(24)(25)(26). The nudC gene encodes a 22-kDa protein of unknown function (25). The nudF gene encodes a 49-kDa protein that is similar to the human lissencephaly (LIS-1) protein (26). LIS-1 is involved in controlling neuronal migration in the cerebral cortex (27). Through A. nidulans studies, NUDC has been shown to posttranscriptionally regulate NUDF (26). Additionally, NUDF appears to be an upstream regulator of cytoplasmic dynein/dynactin function, as a newly discovered mutation in the cytoplasmic dynein heavy chain acts as a bypass suppressor of the nudF deletion (D. A. Willins and N. R. Morris, unpublished data). Therefore, through genetic studies, NUDC can be placed upstream of NUDF, and in turn both NUDC and NUDF can be placed upstream of dynein and dynactin.
The rat Nb2 T cell line can be stimulated to proliferate by the addition of 1-10 ng/ml PRL to the cell culture medium (28). One of the PRL-inducible genes cloned from Nb2 T cells is clone 15 (c15) (29). The c15 gene encodes a 332-amino acid (aa) protein  in which the carboxy-terminal 94 aa (Gly 239 to Asn 332) are 68% identical to the carboxy-terminal portion of the A. nidulans nuclear movement protein NUDC (Gly 105 to Gly 198). This striking similarity suggests that the dynein/dynactin pathway that mediates nuclear movement in fungi may also be involved in reorientation of the centrosomal MTs and Golgi in T cells, and that the c15 and nudC gene products may share similar functions. The present work was designed to test the hypothesis that c15 and NUDC are functionally related by determining whether the rat T cell c15 protein could complement the temperature sensitivity of the A. nidulans nudC3 mutation.

c15 Complementation Strategy and Constructs
The temperature-sensitive A. nidulans used in this study contain the pyrG89 mutation. These A. nidulans lack orotidine 5Ј-P decarboxylase, an enzyme required for the uracil and uridine synthetic pathway (30). Therefore, to grow, the pyrG89 A. nidulans require exogenous uracil and uridine in their growth media. The pAL5 vector was generated by cloning the A. nidulans histone H2A 3Ј-fragment polyadenylation site downstream of the pAL3 multicloning site ( Fig. 1) (31,32). The H2A 3Ј-fragment allows for more efficient processing and export of the mature mRNA from the nucleus and greater stability of the mRNA transcript. The pAL5 vector contains, as does the pAL3 parental vector, the alcA alcohol dehydrogenase I gene promoter, which is inducible or repressible based on the growth media and carbon source present. Transcription of the gene of interest is repressed when A. nidulans containing this construct are grown on rich glucose media (MAG or YAG), and expression is allowed when A. nidulans are grown on glycerol-containing media (32,33). In addition, both pAL5 and pAL3 vectors contain the ampicillin resistance gene, to allow for selection in E. coli, and the N. crassa pyr4 gene. The pyr4 gene encodes for orotidine 5Ј-P decarboxylase and is capable of complementing the pyrG89 mutation in A. nidulans, thereby allowing A. nidulans to grow in the absence of exogenous uracil and uridine (30).
Complementation studies were performed using the pAL5 vector system. The entire c15 open reading frame (ORF), encoding the full-length 332-aa protein, was generated through PCR and was cloned down-stream of the alcA promoter ( Fig. 1). A. nidulans containing the nudC3 temperature-sensitive mutation were stably transformed with either the c15 vector or the pAL5 vector. After transformation, 11 c15 vectortransformed strains and 13 pAL5 vector-transformed strains were obtained and assayed for complementation of the nudC3 mutation at the restrictive temperature (42 C), on the appropriate media (glycerol).

c15 Complements nudC3 Mutants
The A. nidulans nudC3 mutant strain is temperature sensitive ( Fig. 2A). At the permissive temperature, 32 C, the colonies are similar to the wild type strain as they are large, white, have undergone normal growth and differentiation, and are able to generate spores. However, at the restrictive temperature, 42 C, the nudC3 colonies are much smaller than the wild type colonies grown at the restrictive temperature, brown in color, restricted in cellular growth, and unable to undergo normal differentiation to form conidia.
After transformation of the nudC3 mutant strain, the transformants were tested for their ability to complement the temperature-sensitive nudC3 mutation. All 11 c15 vector-transformed strains complemented the nudC3 mutation at the restrictive temperature on glycerol media. In contrast, all 13 pAL5 vector-transformed strains failed to complement the nudC3 mutation at 42 C on glycerol media. To illustrate the complementation phenotype, two representative transformants of each of the two constructs, c15 (designated c15 # 1 and c15 # 2) and pAL5 (designated pAL5 # 1 and pAL5 # 2), were grown on either MAG or glycerol minimal media at the permissive or restrictive temperatures (Fig. 2B). A. nidulans containing either construct grew well on either of the two different media at 32 C. However, when the transformants were grown at the restrictive temperature, only those on the glycerol minimal media containing the c15 construct grew. The c15 transformants grew in a similar manner on glycerol minimal media at either 32 C or 42 C. Therefore, the results of this experiment indicate that the rat c15 protein has the ability to complement the functions lost by the mutant A. nidulans NUDC protein and restore normal growth and differentiation.

c15 Protein Is Expressed in Complemented A. nidulans
To verify that the c15 protein was being expressed in the c15-transformed and -complemented A. nidulans,

Fig. 1. Schematic Diagram of the Complementation Vectors
The pAL5 vector contains the inducible/repressible alcA promoter, the histone H2A 3Ј-fragment polyadenylation signal, the ampicillin resistance gene, and the N. crassa pyr4 gene. The c15 complementation vector contains the entire c15 ORF, encoding the full-length 332-aa c15 protein, cloned downstream of the alcA promoter. The pAL5 vector was used as a negative control for the complementation experiments.

Fig. 2. c15 Complements nudC3 Mutants
A, Temperature-sensitive phenotype of the nudC3 mutants. Wild type A. nidulans and A. nidulans carrying the nudC3 temperature-sensitive mutation were plated on MAG ϩ UU at either 32 C or 42 C. The left panel shows the normal colony growth at 32 C, while the right panel shows the restricted overall development of nudC3 mutants at 42 C as compared with the wild type strain. Bar, approximately 10 mm. B, Phenotype of the transformed nudC3 mutants. Two independently isolated A. nidulans strains carrying either the c15 or pAL5 construct were plated on MAG (top) and glycerol minimal media plates (GLY, bottom) at 32 C (left) and 42 C (right) for 2 days. Only those strains carrying the c15 construct grew on the glycerol minimal media plates at 42 C. Bar, approximately 10 mm.
Western blot analysis of the total cellular proteins was performed using affinity-purified rabbit anti-c15-carboxy-peptide (c15-C) antibodies (S. M. Morris, in preparation). The expression of the 45-kDa c15 protein (arrow) was detected in the two c15-complemented strains tested (Fig. 3, lanes 2 and 3) and not in the control nudC3 strain (Fig. 3, lane 1) or those strains containing the pAL5 vector (Fig. 3, lanes 4 and 5). Additionally, the level of c15 protein expression in the two different c15-transformed strains varied. The c15 # 1 strain (Fig. 3, lane 2) consistently expressed more c15 protein than did the c15 # 2 strain (Fig. 3, lane  3). Furthermore, the c15-C antibodies specifically recognized a 22-kDa protein band (asterisk), corresponding to the mutant NUDC protein in all protein preparations. Interestingly, the overall levels of endogenous mutant NUDC in the c15-expressing strains was lower than the levels of mutant NUDC found in the nontransformed nudC3 strain or in the two pAL5-transformed strains. Both the 45-kDa and 22-kDa protein bands could be specifically competed by the addition of c15-C to the immunoblotting solutions (data not shown).

A. nidulans
To better characterize the complemented phenotype, the nuclei of the transformed A. nidulans were stained with 4Ј,6-diamidino-2-phenylindole (DAPI) to determine whether or not normal nuclear migration was occurring. A. nidulans were grown for 18 h on coverslips in minimal media containing glycerol at either 32 C or 42 C. When A. nidulans containing the nudC3 mutation were grown at 32 C, their nuclei migrated normally into the germ tube and were maintained at equal distances from each other (Fig. 4A). In contrast, the nuclei of A. nidulans containing the nudC3 mutation, when grown at 42 C, divided but failed to migrate out into the germ tube (Fig. 4C). When nudC3 mutant A. nidulans were transformed with the c15 complementation construct and were grown at 42 C on glycerol minimal media, their phenotype closely resembled that of the nontransformed nudC3 mutants grown at the permissive temperature (32 C). The c15-complemented A. nidulans extended and elongated their germ tube (Fig. 4E), and their nuclei migrated into the germ tube of the developing germling.
To further quantitate the percentage of A. nidulans that moved their nuclei under the various growth con- Total cellular proteins (50 g per lane) from two c15transformed strains (c15 # 1, c15 # 2) (lanes 2 and 3), two pAL5transformed strains (pAL5 # 1, pAL5 # 2) (lanes 4 and 5), and the parental nudC3 mutant strain (lane 1) were resolved by 12% SDS-PAGE and transferred to a nylon filter. Primary antibodies, affinity-purified rabbit anti-c15-C, were applied in a 1:500 dilution for 3 h. Secondary antibodies, donkey anti-rabbit IgG conjugated to horseradish peroxidase, were then added in a 1:2000 dilution for 1 h. The arrow indicates the 45-kDa c15 protein band, and the asterisk indicates the 22-kDa mutant NUDC protein band.

Fig. 4. Nuclear Movement Is Restored in Complemented A. nidulans
Spores from the nudC3 mutant strain were germinated and grown at 32 C and 42 C in glycerol minimal media ϩ UU, and spores from a c15-transformed strain were grown in glycerol minimal media at 42 C for 18 h. The germlings were stained with DAPI to allow for visualization of the nuclei. Panels A, C, and E show the results of the DAPI staining, while panels B, D, and F show the phase contrast view of the same field. Bar, approximately 10 m. Panel G shows the overall percentage of A. nidulans germlings that moved their nuclei under the growth conditions described above. DAPI-stained germlings were scored as "no movement" if none of their nuclei entered the germ tube or as "movement" if one or more of their nuclei entered the germ tube. N represents the number of germlings scored for each growth condition.
MOL ENDO · 1997 Vol 11 No. 2 232 ditions, DAPI-stained germlings were counted and scored for their ability to move their nuclei (Fig. 4G). In nudC3 mutants, grown at 32 C in glycerol media, more than 95% of the germlings moved their nuclei. In contrast, when nudC3 mutants were grown at 42 C in glycerol media, only approximately 18% of the germlings counted showed any form of nuclear movement. When the c15-transformed A. nidulans were grown at 42 C in glycerol media, nearly 90% of the germlings moved their nuclei. Therefore, c15 is capable of restoring normal nuclear migration.

c15 Protein Expression Restores NUDF Expression
Endogenous levels of NUDF decrease when nudC3 mutants are grown at the restrictive temperature of 42 C. Studies involving the use of an exogenous promoter to induce nudF gene expression revealed a similar decrease in overall NUDF expression when this construct was transformed into nudC3 mutants that were grown at 42 C. These results suggest that NUDC regulates NUDF levels in a posttranscriptional manner (26). To determine whether or not the expression of c15 at 42 C in nudC3 mutants could rescue NUDF expression, Western blot analysis of total cellular proteins was performed using affinity-purified anti-NUDF antibodies. The expression of the 49-kDa NUDF protein was detected at normal levels in the wild type strain (Fig. 5, lane 1) and in the nudA4 mutant strain (Fig. 5, lane 2) grown at 42 C. However, the expression of the NUDF protein was greatly decreased in the nudF7 mutant strain and the nudC3 mutant strain (Fig.  5, lanes 3 and 4) grown at 42 C. When the c15 protein was expressed at 42 C, in the two c15-complemented strains tested, the expression of NUDF was restored (Fig. 5, lanes 5 and 6). The level to which NUDF protein was restored correlates well with the level of c15 protein expression observed in the two c15-transformed strains (Fig. 3). As a control for the transformation, two control pAL5 vector-containing strains were tested and the levels of NUDF were decreased to levels observed in the nudC3 parental strain (Fig. 5, lanes 7 and  8). The lower band is nonspecific (dot). Therefore, the results of this experiment show that c15 expression can rescue the expression of NUDF at the restrictive temperature.

DISCUSSION
The c15 gene was originally cloned as a PRL-responsive gene from rat Nb2 T cells (29). GenBank searches revealed that the c15 protein exhibits high similarity over its carboxy-terminal 94 aa to the A. nidulans nuclear movement protein encoded by the nudC gene. Based on the similarity that exists between c15 and NUDC, complementation studies were performed to determine whether c15 could functionally rescue temperature-sensitive A. nidulans nudC3 mutants. The results of our experiments show that the expression of the 45-kDa c15 protein in an A. nidulans nudC3 mutant strain can complement the temperature-sensitive growth and nuclear movement phenotypes, as well as restore depleted endogenous NUDF protein to near wild type levels. These results indicate that c15 functions like NUDC in A. nidulans and suggest that c15 may play a functional role in Nb2 T cells similar to the role played by NUDC in A. nidulans. It is interesting to note that the levels of endogenous NUDC in the c15-complemented strains are lower than the levels seen in the nudC3 strain or vector control, pAL5transformed strains. Potentially, the expression of the transformed c15 protein may have the ability to feedback and down-regulate the expression of the endogenous mutant NUDC protein or in some way affect the stability of the mutant NUDC protein, thereby allowing the cell to maintain a "critical" overall level of NUDC/c15 expression.
The nudC3 allele of nudC, like other temperaturesensitive nud mutants (nudA, nudF, and nudG) grows very slowly at the restrictive temperature (12,25,26). The nuclei of these various nud mutants divide normally; however, nuclear migration through the germ tube is arrested. Biochemical studies have shown that the nudC3 mutation affects nuclear migration and growth rate by causing a decrease in the intracellular levels of NUDF, a protein that is essential for nuclear migration in A. nidulans (26). Mutations in the nudF gene that cause a reduction in NUDF protein level also inhibit nuclear movement. Previous work has shown that the effect of the nudC3 mutation on NUDF levels is posttranscriptional, and that NUDC and NUDF are not in complex with each other (26). Recently, a mutation in the A. nidulans cytoplasmic dynein heavy chain that acts as a bypass suppressor of the nudF deletion has been discovered (D. A. Willins and N. R. Morris, unpublished data). This finding suggests that NUDF may serve as an upstream regulator of dynein/ dynactin function. In A. nidulans, cytoplasmic dynein is the main motor for nuclear migration, and evidence derived from characterization of the nudC3 mutant phenotype indicates that the nudC gene product probably acts as an upstream modulator of dynein function. Interestingly, the Drosophila nudC gene shows a genetic interaction with the Drosophila Glued gene, which encodes a component of dynactin (J. Cunniff, R. Warrior, Y. Chiu, and N. R. Morris, unpublished data). As a result of these genetic studies, a hierarchy can be established in which NUDC can be placed upstream of NUDF, and both NUDC and NUDF can be placed upstream of dynein and dynactin. Whether or not the NUDC protein directly affects cytoplasmic dynein or its functional partner, dynactin, is not known (19,34).
The nudF gene encodes a 49-kDa WD-40 protein with a putative coiled-coil domain, similar to the ␤-subunit of heterotrimeric G proteins (26,35,36). However, NUDF most closely resembles (42% identity) the human lissencephaly-1 gene (LIS-1) (27). Lissencephaly is a neuronal migration disease characterized by the inability of neurons to migrate to their proper positions in the cerebral cortex (37). There is evidence to suggest that neurons may migrate by first extending a long process through which the nucleus and associated organelles then move to establish a new position for the cell body (38,39). These observations potentially link the genes and proteins involved in centrosomal positioning and nuclear migration with those involved in neuronal migration and brain development. LIS-1 is 99% identical to the 45-kDa regulatory subunit of bovine brain platelet-activating factor acetylhydrolase (PAFAH) (40). The PAFAH enzyme inactivates platelet-activating factor, a lipid second messenger, by removing the acetyl group at the sn-2 position (41,42). The structural similarities that exist between PAFAH, LIS-1, and NUDF suggest that these proteins may be involved in PAFAH-associated functions. Thus, NUDC may play a role in regulating the activity or the targeting of the PAFAH complex in mammalian cells.
Centrosome orientation in T cells appears to be mediated by a force on centrosomal MTs just as nuclear movement is mediated by a force on SPB MTs (8,11). We propose that c15 may play a role in the centrosome reorientation that occurs when a T cell meets an antigen-presenting cell or antigen-bearing target cell (3,43,44). In this way, the Golgi apparatus and its associated secretory vesicles are reoriented, allowing transport of vesicles along the centrosomal MTs toward the site of contact (3,44). Although maximal c15 gene induction in T cells by PRL occurs at the G1/S transition (29), a constitutive level of c15 protein already exists (data not shown), which may participate in the rapid centrosome MTOC reorganization process. A similar kinetics of c15 induction was observed in IL-3-stimulated premyeloid cells which suggests that c15 induction by cytokines is part of an activation response (29). Furthermore, the c15 gene product could be involved in other cytoplasmic dynein-mediated functions other than moving centrosome and nuclei, e.g. vesicle migration or mitosis (45,46), which occur later in the cell cycle as suggested by the kinetics of c15 induction in PRL-stimulated T cells (29). This additional functional complexity is also suggested by the large difference in size between the 45-kDa c15 protein and the 22-kDa NUDC protein. The size difference is due to a much larger amino-terminal domain in the rat c15 protein, which contains basic and acidic motifs that may be involved in protein-protein interactions.
In summary, we have identified a mammalian (rat) gene from PRL-stimulated T cells, c15, that when expressed in A. nidulans nudC3 mutants can functionally rescue the nuclear movement defect. The fact that the rat c15 protein can replace the nuclear movement function of the A. nidulans nudC gene suggests that c15 and NUDC have similar functions. Because of the structural and functional similarities between c15 and NUDC, c15 is likely a mammalian homolog of nudC, and we have therefore named it RnudC. Additionally, c15/RNUDC is widely represented throughout eukaryotes, as immunoreactive c15/RNUDC-like proteins have been detected in Drosophila (J. Cunniff and R. Warrior, personal communication), monkey, and man (data not shown).
Future studies will address the localization of c15/ RNUDC in the cell and its association with other proteins. We are particularly interested in determining whether c15/RNUDC localizes to the centrosome MTOC or to the cortex in PRL-stimulated Nb2 T cells. These studies will help determine the role of c15/ RNUDC in the events leading to T cell activation, proliferation, and differentiation.

Complementation Vectors
The 100A plasmid, containing the entire rat c15 ORF (29), was used as a template to generate a 1.0-kb DNA fragment by PCR using Pfu polymerase (Statagene, La Jolla, CA). The upstream primer contained a KpnI linker (lowercase): 5Ј-gcgggtaccgATGGGAGGGGAGCAG-3Ј (start codon underlined), and the downstream primer contained an XbaI linker (lowercase): 5Ј-gcgtctagaCTAGTTGAATTTGGC-3Ј (stop codon underlined). The KpnI/XbaI-digested PCR product, encoding c15, was cloned downstream of the alcA promoter in the pAL5 vector (32) and confirmed by sequencing. The c15 vector and pAL5 vector were used to transform A. nidulans containing the nudC3 mutation.

A. nidulans Transformation
Conidia (1 ϫ 10 9 ) were inoculated into 50 ml YAG without agar supplemented with UU. The conidia were allowed to germinate for about 5.5 h, or until the emerging germ tube was visible, at 32 C with shaking. The , and the cell wall was digested for 2 h at 32 C with shaking. The protoplasts were washed two times in Solution C (50 mM citric acid, pH 6.0, 0.4 M (NH 4 ) 2 SO 4 , 1% sucrose) and were resuspended in 1 ml Solution E (0.6 M KCl, 100 mM CaCl 2 , 10 mM Tris-HCl, pH 7.5). Protoplasts (100 l) were added to 6 g plasmid DNA, followed by the addition of 50 l Solution D (25% polyethylene glycol 8000, 100 mM CaCl 2 , 0.6 M KCl, 10 mM Tris-HCl, pH 7.5), and incubated on ice for 20 min. Next, 1 ml Solution D was added to the protoplasts and allowed to incubate at room temperature for 30 min. Aliquots (200 l) of the protoplast mixture were plated in 3 ml 45 C sucrose top agar (0.5% yeast extract, 20 mM glucose, 1 M sucrose, trace elements, vitamins, 1% agar) onto sucrose plates (0.5% yeast extract, 20 mM glucose, 0.2 M sucrose, trace elements, vitamins, 1.5% agar), and incubated at 32 C for 3-4 days.

Complementation of nudC3 Mutants with c15
To test for complementation, conidia containing either the c15 or pAL5 construct were streaked on either MAG or glycerol minimal media lacking UU at 32 C or 42 C. Independently isolated transformants of each of the two constructs (c15 and pAL5) were plated, and colonies were allowed to grow for 2 days before analysis. Conidia from the transformants were isolated and retested for complementation five times.

Protein Preparation and Western Blot Analysis
To prepare total cellular proteins, 5 ϫ 10 8 conidia were inoculated into 50 ml supplemented minimal media containing 10 ml/liter glycerol and incubated for 42 h at 42 C with shaking. The mycelia were harvested by centrifugation, washed in ice-cold H 2 O, collected by filtration through cheesecloth, and pressed dry. The mycelia were ground in a Tenbroek homogenizer (Fisher Scientific, Pittsburgh, PA) in 1-2 ml extraction buffer [50 mM Tris-HCl, pH 7.4, 200 mM NaCl, 5 mM EDTA, 5 mM benzamidine, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml each of soybean trypsin inhibitor, aprotinin, leupeptin, N-tosyl-L-phenylalanine chloromethyl ketone (Sigma)], and insoluble proteins were removed by centrifugation at 14,000 rpm for 5 min at 4 C. Protein concentration was determined by Bradford assay (Bio-Rad, Richmond, CA) using BSA as a standard. The cellular proteins were stored at Ϫ20 C.
For Western blotting, 50 g total cellular proteins were analyzed by 12% SDS-PAGE and transferred to Immobilon P (Millipore, Bedford, MA) or nitrocellulose (Bio-Rad) as previously described (29). To determine c15 expression, the blot was blocked with 5% nonfat milk, 0.2% Tween 20 (Sigma) in Tris-buffered saline, and affinity-purified rabbit anti-c15-C antibodies (S. M. Morris, in preparation) were applied in a 1:500 dilution, followed by the addition of donkey anti-rabbit IgG antibodies conjugated to horseradish peroxidase (Amersham, Arlington Heights, IL) in a 1:2000 dilution. c15 proteins were detected with the enhanced chemiluminescence system as suggested by the manufacturer (Amersham). To detect NUDF, affinity-purified rabbit anti-NUDF antibodies were used at a 1:100 dilution, followed by goat anti-rabbit IgG antibodies coupled to alkaline phosphatase in a 1:2000 dilution, and developed with 5-bromo-4-chloro-3-indoyl phospate p-toluidine salt and p-nitro blue tetrazolium chloride (26).

DAPI Staining of Nuclei
To stain the nuclei of the developing germlings, 5 ϫ 10 6 conidia were inoculated into Petri dishes containing sterile coverslips and 25 ml supplemented liquid minimal media with 10 ml/liter glycerol and were incubated at either 32 C or 42 C for 18 h. The coverslips with the attached germlings were rinsed in H 2 O and placed in methanol at Ϫ20 C for 10 min, rinsed well with H 2 O, and placed in acetone at Ϫ20 C for 10 min. The coverslips were rinsed again and placed in a 50 ng/ml DAPI (Sigma) solution for 10 min. After a final rinse, the coverslips were mounted in ProLong Antifade (Molecular Probes, Inc., Eugene, OR) and viewed at 1000ϫ using the Zeiss Axiophot system (Zeiss, Jena, Germany) (Baylor Integrated Microscopy Core, Baylor College of Medicine, Houston, TX).