Search

Tad is a LINE-like retrotransposon of the filamentous fungus Neurospora crassa. We have analyzed both expression and transposition of this element using strains with a single copy of Tad located in the 5' noncoding sequences of the am (glutamate dehydrogenase) gene. Tad in this position has been shown to carry a de novo cytosine methylation signal which causes reversible methylation of both Tad and am upstream sequences. Here we find that methylation of the Tad sequences inhibits both Tad expression and transposition. This inhibition can be relieved by the use of 5-azacytidine, a drug which reduces cytosine methylation, or by placing the Tad/am sequences in a dim-2 genetic background.

Tad is an active LINE-like retrotransposon isolated from the Adiopodoumé strain of Neurospora crassa. Extensive analysis of other Neurospora strains has revealed no other strain with active Tad, but all strains tested have multiple copies of defective Tad elements. We have examined the ability of Tad to survive during the sexual cycle of Neurospora and find that active Tad is rapidly eliminated. The characteristics of this elimination suggest that the repeat-induced point mutation (RIP) mechanism was responsible. By the use of transformation to switch the mating type of the Adiopodoumé strain we concluded that this strain is not defective in the RIP process. Analysis of defective Tad elements isolated from a variety of strains indicates that the major difference between these elements and active Tad is due to the presence of a large number of G-C to A-T transition mutations. This would be expected if the changes were due primarily to the RIP process. Mapping of a selection of defective Tad elements reveals that they are present on all of the chromosomes; however, many of the elements are not widely shared among strains. This suggests that repeated introduction and elimination of Tad elements has occurred. Mechanisms that might be responsible for this repeated introduction are discussed., (Copyright 2001 Academic Press.)

The expression of the am (glutamate dehydrogenase) gene is dependent upon two upstream activating sequences, designated URSam(alpha) and URSam(beta). A heteromeric nuclear protein Am Alpha Binding protein (AAB) binds specifically to a CCAAT box within the URSam(alpha) element. AAB appears to be composed of three components. We used polyclonal antiserum raised against the highly purified AAB1 subunit to isolate a partial aab-1 cDNA clone, which was then used to isolate a full-length cDNA and a genomic clone. The full-length cDNA has the potential to encode a 272 amino acid protein with a calculated molecular weight of 30 kD. Amino acid sequence obtained by Edman analysis of the AAB1 protein confirmed that the aab-1 gene had been cloned. AAB-1 shows similarity to the HAP5 protein of yeast and the CBF-C protein of rat. Each of these proteins is an essential subunit of their respective heteromeric CCAAT binding proteins. The aab1 gene maps on linkage group III of Neurospora crassa near the trp-1 locus. Disruption of the aab-1 gene results in pleiotropic effects on growth and development as well as a 50% reduction in glutamate dehydrogenase levels. Transformation of the aab-1 disruption mutant strain with the cloned genomic copy of the aab-1 gene rescued all of the phenotypic alterations associated with the aab-1 mutation.

Tad is a LINE-like element from Neurospora. Inactive Tad elements have been found in all strains of Neurospora examined; however, active copies of Tad were initially found in a single strain (Adiopodoumé) collected from nature. Active Tad elements can be transferred to laboratory strains by means of crosses, heterokaryons or by transformation with cloned active elements. We have analyzed the pattern of transcripts from Tad elements present in strains derived from Adiopodoumé containing either high or low copy numbers of active Tad. In both kinds of strains we found a complex set of discrete polyadenylated transcripts. As expected, a full-length transcript was detected with 5' and 3' ends that coincide with the consensus 5' and 3' ends of the Tad element. Two additional transcripts with the same polarity as the full-length transcript were identified. One of these plus-strand transcripts is 2.0 kb in length with a 5' end identical to that of the full-length transcript and an internal polyadenylation site. The other (4.2 kb) has an internal start site with termination occurring at the 3' consensus end of the element. The 2.0-kb transcript encodes all of ORF1. The 4.2 kb transcript could encode a 5' truncated version of ORF2. Two minusstrand transcripts were also consistently identified. These represent overlapping transcripts from the 3' tail of Tad. These transcripts share a small intron and 3' polyadenylation site. Their 5' start sites are both internal to the consensus 3' end of Tad. Possible functions of this complex array of transcripts are discussed.

An unstable allele of the Neurospora am (GDH) gene resulting from integration of the retrotransposon Tad3-2 into 5' noncoding sequences was found in previous work. We report that reversion to Am+ depends on DNA methylation within and upstream of Tad. Levels of methylation were correlated with the proportion of Am+ conidia, whether the cultures were derived from Am- or Am+ isolates. Reversion to Am+ did not occur when conidia were plated on 5-azacytidine, which reduces DNA methylation. The mutation dim-2, which appears to abolish DNA methylation, also prevented reversion to Am+. The native am allele, in a strain that lacked Tad elements, was replaced with am::Tad3-2 or with a deletion derivative that prevents transposition of Tad. Transformants of both classes showed instability comparable with that of the original isolates, which contain multiple Tad elements. Deletion of the upstream enhancer-like sequences, URSam alpha and beta, did not prevent the instability of am::Tad3-2. The results suggest that am expression is dependent on DNA methylation but not on proliferation or transposition of the Tad element and that the instability does not require the upstream sequences of am.

Expression of the Neurospora crassa am (NADP-specific glutamate dehydrogenase) gene is controlled by two upstream enhancer-like elements designated URSam alpha and URSam beta. URSam alpha is localized between - 1.3 and - 1.4 kb with respect to the major transcriptional start site. Deletion of a 90 bp sequence containing this element resulted in the loss of approximately 50% of normal glutamate dehydrogenase expression. Gel mobility shift analysis indicated that a nuclear protein from Neurospora binds in a specific manner to sequences within the 90 bp fragment. We have now used a combination of ion-exchange and affinity chromatography to purify this nuclear protein, which we call Am Alpha Binding protein (AAB). The activity was monitored by gel shift analysis. The protein was purified more than 14,000-fold with a yield of approximately 7%. The purified protein appears as a heteromer on denaturing polyacrylamide gel electrophoresis, with only two strong bands visible in silver-stained preparations. One band has an apparent molecular mass of 40 kDa, the other appears as a doublet with an apparent molecular mass of 30 kDa. DNAse I protection analysis indicated a protected region consisting of 30 bp, which contains a CCAAT pentanucleotide motif. Mutagenesis of the CCAAT motif abolished the binding of AAB to the DNA fragment.

RIP (repeat-induced point mutation) efficiently mutates repeated sequences in the sexual phase of the Neurospora crassa life cycle. Nevertheless, an active LINE-like retrotransposon, Tad, was found in a N. crassa strain from Adiopodoumé. The possibility was tested that Tad might be resistant to RIP, or that the Adiopodoumé strain might be incompetent for RIP. Tad elements derived from the Adiopodoumé strain were found to be susceptible to RIP. In addition, strains lacking active Tad elements, including common laboratory strains and strains representing seven species of Neurospora, were found to have sequences closely related to Tad but with numerous mutations of the type resulting from RIP (G:C to A:T). Even the Adiopodoumé strain showed Tad-like elements with mutations characteristic of RIP. Results of crossing of an Adiopodoumé transformant with progeny of Adiopodoumé suggest that the Adiopodoumé strain is proficient at RIP. We conclude that Tad is an old transposable element that has been inactivated by RIP in most strains. Finding relics of RIP in both heterothallic and homothallic species of Neurospora implicates RIP across the genus.

We have developed a system of recipient strains and donor plasmids that allows targeting of DNA sequences to the am locus of Neurospora crassa. A recipient strain was constructed that contains the 3' two-thirds of the bacterial Hygromycin B (Hy) phosphotransferase-encoding gene (hph) in single copy downstream from the am structural gene encoding NADP-specific glutamate dehydrogenase. Plasmids have been constructed that contain am, but with the 5' end of hph downstream from am. Resistance to Hy can occur during transformation only if recombination occurs within the region of overlap of the hph sequences from the transforming DNA and of the chromosomal hph sequences. A lacZ alpha gene with a multiple cloning site has been inserted into the region between the am and hph sequences to facilitate subcloning into the targeting vector. Additionally, unique sites at the junctions of the vector and targeting DNA allow easy removal of vector sequences prior to transformation. A second recipient strain has been constructed that has had the am structural gene deleted. This strain may be used to provide a double selection for homologous integration. Additionally, this strain facilitates replacement of am with heterologous constructs, including constructs that drive transcription of am with different promoters. The insertion of hph downstream from am has no apparent effect on am expression, and the resistance segregates as a Mendelian factor in crosses. Transformation efficiency is reduced two to three orders of magnitude, but interestingly, nearly all of the resistant strains have only a single copy of the transforming DNA.

Tad is a LINE-like retrotransposon of Neurospora crassa. The element was originally detected and cloned using the am gene as a transposon trap in hybrid strains derived from a cross of Adiopodoume (a wild collected strain) and a laboratory strain devoid of Tad elements. We report the cloning and sequencing of an active Tad element, Tad1-1, which is capable of independent transposition. Transposition was demonstrated by screening for transfer of the element from a donor nucleus that contained the Tad1-1 element as the only active Tad, into a naive nucleus within a forced heterokaryon. We also report here the sequence analysis of Tad1-1, and its comparison with the sequence of another active element, Tad3-2. These elements are approximately 7 kb in length. They contain two long open reading frames (ORFs) encoded on the strand of the same polarity as the full-length transcript. ORF1 encodes a putative protein of 486 amino acids. Homology to the first ORF of other LINE elements is confined to three cysteine-rich motifs, located near the carboxy-terminus, that are thought to be involved in binding nucleic acids. The second ORF is 1156 amino acids in length and shows homology to the reverse transcriptase domains of various retroviruses and retrotransposons. Tad1-1 and Tad3-2 differ in only ten positions over their whole length.

We have used gel mobility shift assays to scan 1.7 kb of 5' non-coding sequence of the am (glutamate dehydrogenase) gene of Neurospora crassa for binding by partially fractionated Neurospora proteins. Using genetic analysis this region had been shown to play an important role in the control of glutamate dehydrogenase (GDH) expression. Gel mobility shift analysis identified three regions to which Neurospora proteins bind specifically. Two of these corresponded to the two elements previously defined by genetic analysis (URSam alpha and URSam beta). The third protein binding site appears to be unrelated to am gene expression. Competition experiments showed that the proteins that bind to the URSam alpha and URSam beta elements are different. The URSam alpha element was shown to contain two independent binding sites for the URSam alpha binding protein(s). Both fragments contain a CCAAT motif, suggesting that URSam alpha binding protein(s) may be members of one of the CCAAT-binding protein families. The effect of deletion of either the URSam alpha or URSam beta elements on catabolite induction of am expression was also determined. Both elements appear to act as constitutive enhancers of gene expression.

Tad is a LINE-like DNA element found in Neutrospora crassa. A Neurospora artificial intron based on the first intron of the am (glutamate dehydrogenase) gene was constructed and introduced, in the correct orientation, into a unique Nru I site in open reading frame 1 of an active Tad element, Tad1-1. Transformants containing the Tad element with the artificial intron were placed in forced heterokaryons with strains lacking Tad elements. Tad was shown to transpose between nuclei in these heterokaryons. Examination of the transposed Tad elements showed that the intron had been precisely removed in all cases. This confirms that Tad is a retrotransposon and that there is a cytoplasmic phase in these retrotransposition events.

New chromosome rearrangements were found in 10% or more of mitotically stable transformants. This was shown for transformations involving a variety of different markers, vectors and recipient strains. Breakpoints were randomly distributed among the seven linkage groups. Controls using untransformed protoplasts of the same strains contained almost no rearrangements. A study of molecularly characterized Am+ transformants showed that rearrangements are frequent when multiple ectopic integration events have occurred. In contrast, rearrangements are absent or infrequent when only the resident locus is restored to am+ by a homologous event. Sequences of the transforming vector were genetically linked to breakpoints in 6 of 10 translocations that were examined using Southern hybridization or colony blots.

We have analyzed the junctions involved in two examples of ectopic integration of plasmids containing the am+ (glutamate dehydrogenase) gene into a strain of Neurospora crassa bearing a complete deletion of the am locus. In one transformed strain a single copy of plasmid DNA had been integrated into linkage group (LG) III DNA without the loss of chromosomal DNA. In contrast, 450 bp had been lost from plasmid sequences at the site of integration. The transforming DNA used was circular, so we postulate that the plasmid was linearized and truncated prior to its integration by end joining into a break in LG III DNA. There was no significant homology between the incoming DNA and DNA at the site of integration. The second transformed strain resulted from transformation with a linearized plasmid. It contained multiple integrated copies of plasmid DNA, one of which was recloned, together with adjacent chromosomal DNA, by plasmid rescue in Escherichia coli. Prior to integration into chromosomal DNA, the linear plasmid had been truncated by 64 bp on one end and 3.2 kbp on the other end. One end of the integrated DNA was adjacent to DNA from the right arm of LG I, while the other end was integrated into a copy of a repetitive sequence. Restriction fragment length polymerism mapping showed that integration was in a copy of the repetitive sequence that is linked to the previously unassigned telomere M11 and is distantly linked to the LG VI marker con-11. Genetic analysis revealed that a long segment of LG I containing all markers from un-1 to the right tip has been translocated to the right end of LG VI. Tetrad analysis showed that the integrated DNA was closely linked to the translocation. We conclude that the transforming DNA was truncated and joined to DNA from two different chromosomes by end joining during the formation of a quasiterminal translocation, T(IR----VIR) UK-T12. We also conclude that the previously unassigned telomere, M11, is the right end of LG VI.

The process designated RIP (repeat-induced point mutation) alters duplicated DNA sequences in the sexual cycle of Neurospora crassa. We tested whether non-Neurospora sequences are susceptible to RIP, explored the basis for the observed immunity to this process of a diverged tandem duplication that probably arose by a natural duplication followed by RIP (the Neurospora zeta-eta region), and investigated whether RIP extends at all into unique sequences bordering a duplicated region. Bacterial sequences of the plasmid pUC8 and of a gene conferring resistance to hygromycin B were sensitive to RIP in N. crassa when repeated in the genome. When the entire 1.6-kb zeta-eta region was duplicated, it was susceptible to RIP, but was affected by it to a lesser extent than other duplications. Only three of 62 progeny from crosses harboring unlinked duplications of the region showed evidence of changes. We attribute the low level of alterations to depletion of mutable sites. The stability of the zeta-eta region in strains having single copies of the region suggests that the 14% divergence of the tandem elements is sufficient to prevent RIP. DNA sequence analysis of unduplicated pUC8 sequences adjacent to a duplication revealed that RIP continued at least 180 bp beyond the boundary of the duplication. Three mutations occurred in the 200-bp segment of bordering sequences examined.

The Tad transposon of Neurospora crassa appears to be a LINE-like element with very restricted distribution within the genus Neurospora. When forced heterokaryons were constructed between strains which did and did not contain Tad, the nuclei of the naive nuclear type rapidly acquired Tad elements. The elements acquired by naive nuclei are active, since they can pass Tad to other naive nuclei in subsequent heterokaryons. When heterokaryons are passaged by serial transfer, the load of acquired Tad elements appears to increase, indicating that transposition is continuing in these heterokaryons, even after all of the naive nuclei have acquired Tad. In normal heterokaryons of Neurospora, nuclei do not fuse. An experiment to test for the possibility that Tad promotes nuclear fusion gave negative results. Thus Tad appears to have a cytoplasmic intermediate in its transposition. When heterokaryon incompatible strains were cocultured, there was no indication that Tad elements could be transferred to the naive strain, suggesting that Tad is not a virus. These data are consistent with the transposition of Tad via RNA and cDNA intermediates, as has been postulated to occur with LINE-like elements.

We have constructed deletions in the 5' noncoding sequences of the cloned Neurospora crassa am gene. Vectors with a truncated fragment of the am gene were used in transformation experiments to introduce the deletions into the chromosome by homologous recombination. Analysis of glutamate dehydrogenase (GDH) expression by enzyme assay and immunoblots, as well as Northern and dot blots of poly (A)+ RNA, in the deletion strains indicates that there are two upstream regulatory sequences that control the level of gene expression. The closer of these two elements (URSam alpha) is at approximately 1.4 kb upstream of the transcriptional start site. The second elements (URSam beta) is located between 2.1 and 3.2 kb upstream of the transcription start site. Deletion of either of these two elements reduces am expression to about 50% of the wild-type level. Deletion of both elements reduce am expression to from 5-16% of the wild-type level. Deletion of 1.1 kb of sequence just downstream of URSam alpha, which brings this element to within 300 bp of the transcription start site, had no effect on am expression. Likewise, deletion of 3.5 kb of sequence upstream of URSam beta had no effect on expression. None of these deletions had any effect on the expression of usg-1, a gene of unknown function that is transcribed in the same direction as the am gene, and which terminates about 3.5 kb upstream of the URSam beta element.

We have constructed a series of deletions in the 5' non-coding sequences of the cloned Neurospora crassa am gene which specifies NADP specific glutamate dehydrogenase. All of the deletions begin at -4.4 kb with respect to the am transcription start site and extend for various distances toward the am gene. Using vectors with a truncated fragment of the am gene, we introduced these deletions into the chromosome upstream of am by transformation. Analysis of glutamate dehydrogenase expression in strains with the deletion mutations confirmed that there are two upstream regulatory sequences (URS) that control the expression of the am gene. The more distal of these elements (URSam beta) has been limited to the 157 bp between -1924 and -2081 with respect to the start of am transcription. The proximal element (URSam alpha) was limited to the 97 bp between -1296 and -1393. The DNA sequence of the entire region was determined. Within the sequences that contain the URS elements several regions of homology with yeast UAS sequences were found. Gel mobility assays with DNA fragments containing the URS elements indicated that sequences in both elements are bound by nuclear proteins from Neurospora. The interaction of these proteins and the DNA fragments was found to be specific.

We used lambda and plasmid vectors containing the am+ gene in an insert of from 2.7 to 9.1 kb, to transform am point mutant and deletion strains. A total of 199 transformants were examined with the potential to yield am+ transformants by homologous recombination. When we used vectors that had 9.1 kb of homology with the chromosomal DNA, 30% of the transformants obtained were the result of homologous recombination regardless of whether the vector was a lambda molecule, a circular plasmid, or a plasmid that had been linearized prior to transformation. When vectors with up to 5.1 kb of homology were used, very few transformants (1 of 89 tested) resulted from homologous recombination. Of a sample of 29 ectopic integration events obtained by transformation with the 9.1 kb fragment cloned in a lambda vector, 18 included a major part (usually almost all) of both arms of lambda with the entire Neurospora 9.1 kb insert between them. Four included only lambda long arm sequence together with an adjacent segment of the insert containing the am gene. The remaining seven were the result of multiple integrations. There was no evidence of circularization of the lambda vector prior to integration. All transformants that had multiple copies of the am gene appeared to be subject to the RIP process, which causes multiple mutations in duplicated sequences during the sexual cycle.

We have characterized Neurospora crassa transformants obtained with plasmid pJR2, which consists of the Neurospora glutamate dehydrogenase (am) gene cloned in pUC8 and an am132 host strain which contains a deletion encompassing the cloned fragment. Every one of 33 transformants tested showed extreme meiotic instability: less than 1 or 2% am+ progeny were obtained in initial or successive backcrosses between am+ transformants and am132 or in intercrosses between am+ progeny. Furthermore, am+ progeny from backcrosses gave a high proportion of auxotrophic (am) mitotic segregants during vegetative growth. These results indicate that the am+ character is not stably integrated into chromosomal DNA in any of the transformants tested. Nuclear DNAs from six transformants were analyzed by Southern hybridization. All six transformants contained sequences homologous to pJR2. Four showed restriction fragments expected for unmodified pJR2, but most showed additional bands. Southern blots of undigested DNAs showed that the plasmid sequences are present predominantly in high-molecular-weight form (larger than 20 kilobases). Southern blots showed that auxotrophic (am) progeny from a backcross to am132 had lost restriction bands corresponding to free plasmid but retained additional bands, apparently integrated into chromosomal DNA in a nonfunctional manner. Considered together, these results are most reasonably interpreted as follows: recombinant plasmids containing the am+ gene can replicate autonomously in N. crassa, the free plasmids are present in oligomeric or modified form or both, and plasmid sequences also integrate at multiple sites in the deletion host but in a nonfunctional manner. An alternate interpretation--that tandem repeats of the plasmid are integrated into chromosomal DNA but eliminated during meiosis--cannot be completely excluded. However, stable integration of the am gene can be obtained under a variety of other conditions, viz., using the am gene cloned in a phage lambda vector (J. A. Kinsey and J. A. Rambosek, Mol. Cell. Biol. 4:117-122, 1984), using derivatives of pJR2, or using pJR2 to transform a frameshift mutant.

In a previous study the alteration in the amino acid sequence of Neurospora crassa NADP-specific glutamate dehydrogenase (GDH) resulting from two mutually compensating frameshift mutations was used to deduce the first 17 nucleotides of the coding sequence of the am gene. In the work reported here, a synthetic 17-mer corresponding to the deduced sequence was shown to hybridize strongly to a 9-kb HindIII fragment from N. crassa wild-type DNA but not to any corresponding fragment from the DNA of a mutant strain known to be deleted for most or all of the gene. Wild-type HindIII fragments were fractionated for size and a fraction centering around 9 kb was cloned in vector lambda L47. Two clones carrying the strongly hybridizing fragment were identified. The hybridization to the 17-mer was localized within a 2.7-kb BamHI fragment and, within this, to a 700-bp BamHI-Bg/II subfragment. 5' end-labelled polyadenylated RNA isolated from wild-type mycelium hybridized to the 2.7-kb BamHI fragment and not appreciably to flanking fragments. The partial sequence analysis of the BamHI-Bg/II fragment has confirmed that the 17-mer probe matches the coding sequence at the 5' end of the gene and has also revealed an intervening sequence 67 bp in length, interrupting codon 15. Both the 9-kb HindIII fragment and the 2.7-kb BamHI fragment have been shown to be capable of transforming the deletion mutant to prototrophy and ability to produce GDH. Analysis of one transformant showed that the am gene was integrated, together with a part of the long arm of the lambda vector, at an unusual locus. This transformant, in which the am gene does not show its normal linkage to the linkage group 5 marker inl, was found to produce GDH to about 20% of the normal level.

We have cloned the unstable am mutant gene, am126, as well as the am gene from an am126 revertant. The mutation is a result of a 33-bp duplication that repeats a sequence starting 13 bp upstream of the 3' splice junction between intron 1 and exon 2 and extends 20 bp into exon 2. In addition, there is a G----A transition 2 bp upstream of the first copy of the duplicated sequence. In the revertant gene the wild-type sequence is precisely recovered, involving both the loss of the duplication and a reversion (A----G) of the associated transition. Our data suggest that only the more 5' of the two 3' splice junctions present in the duplicated version of the gene is used. This favors a 5'----3' scanning mechanism for exon splicing.

A Neurospora crassa strain from Adiopodoumé, Ivory Coast, contains multiple copies of a transposable element, Tad. The element was detected as a 7-kilobase insertion in two independently isolated spontaneous forward mutants of the am (glutamate dehydrogenase) gene. Laboratory strains do not contain Tad. All progeny from crosses of the Adiopodoumé strain to laboratory strains contain multiple copies. When the element was inserted in am, target sequences of 14 and 17 base pairs were duplicated in the two cases analyzed. One mutation, caused by the insertion of Tad at the beginning of the am coding sequence, is genetically stable. The other mutation, caused by insertion upstream of the transcriptional start site, has a reversion frequency of 2.5 x 10(-3). Precise excisions of Tad have not been found.

The mutant strain am126 was isolated, using the direct selection procedure, after nitrous acid mutagenesis. It produced neither measurable NADP-dependent glutamate dehydrogenase (GDH) nor immunologically cross-reacting material. That the am126 strain produced some form of GDH product was shown by the fact that it complemented several other am mutant strains. The GDH formed by complementation between am126 and each of two other am mutants was relatively thermolabile, but could not be distinguished from wild-type GDH formed by electrophoresis in polyacrylamide gels. This, together with the relatively high yield of the complementation enzymes, suggest that the am126 product is a polypeptide chain not grossly abnormal in structure. The spontaneous revertant frequency was between 0.3 and 3 prototrophic revertants per 10(5) live cells. This frequency was at least 40 times greater than that for am19, which had the second highest spontaneous revertant frequency among the mutants tested. Neither meiosis nor mutagenesis increased the revertant frequency, nor did incubation at elevated temperatures lower it. Sixty-eight revertant strains were examined for thermostability of their GHD. All appeared to be identical to wild type. Seven of the revertant strains were also tested for instability with regard to forward mutation to am auxtrophy. None was found to be unstable. Models for the genetic instability of the am126 mutation are discussed.

Forty-eight new mutations at the am locus of Neurospora crassa have been characterized. Nineteen mutations were induced by UV; of these, eight were missense, two were frameshifts, two were nonsense, three were deletions and four were unidentified. Twenty-nine mutations were induced with nitrous acid; of these, twenty-one were missense, three were frameshifts, one was nonsense, two were deletions and one was genetically unstable.

A method was developed for the isolation of Neurospora bradytrophs. The bradytrophs (representing lesions in a number of pathways) were resistant to DL-p-fluorophenylalanine when growing in a leaky fashion but were sensitive when grown in the presence of their stimulating supplement.

We used DNA containing the am gene of Neurospora crassa, cloned in the lambda replacement vector lambdaL-47 (this clone is designated lambdaC-10), and plasmid vector subclones of this DNA to transform am deletion and point mutant strains. By means of subcloning, all sequences required for transformation to am prototrophy and expression of glutamate dehydrogenase have been shown to reside on a 2.5-kilobase BamHI fragment. We also characterized several am+ strains that were obtained after transformation with lambdaC-10. These strains showed Mendelian segregation of the am+ gene, although less than 50% of the transformed strains showed the normal linkage relationship of am with inl. In all cases tested, the strains had incorporated lambda DNA as well. The lambda DNA also showed a Mendelian segregation pattern. In one case, the incorporation of am DNA in a novel position was associated with a mutagenic event producing a strain with a very tight colonial morphology. In all cases in which the am+ gene had become the resident of a new chromosome, glutamate dehydrogenase was produced to only 10 to 20% of the wild-type levels.

Specific in vitro-generated insertion, replacement, and deletion mutations have been integrated near the chromosomal locus of am (NADP-specific glutamate dehydrogenase) of Neurospora crassa. Two approaches have been successful. One approach used am+-containing vectors capable of integrating at any site in the genome. This technique was used to introduce a specific 700 bp insertion near the am locus and to replace chromosomal sequences near am with plasmid DNA. Efficiency was low, however, and many transformants had to be screened to find the desired alterations among the ectopic insertions unless the incoming DNA had a large region of homology with the am region. A second approach increased the efficiency by using vectors containing a truncated am gene, so that prototrophs could arise only by homologous recombination. Overall transformation frequency was reduced relative to the first method, but a large fraction of the transformations involved specific alterations of the am region.

The am locus of Neurospora codes for NADP-dependent glutamate dehydrogenase (GDH). Four new am mutants that produced mutationally altered GDH have been characterized. Mutant am119 is a CRM-negative, complementing mutant that maps between am2 and am1. The other three mutants are CRM formers that produce varieties of GDH that can be activated by glutamate or succinate. The GDH of am130 and am131 is similar in terms of activation properties to that of am3. The GDH of am122 requires very high concentrations of dicarboxylate for activity. The mutation in am130 maps between am14 and am2 and resulted in a replacement at residue 75 of the GDH (pro leads to ser). The mutation in am122 maps near am11 and apparently resulted in the replacement of the tryptophan residue at position 389 with an unknown amino acid. The mutation in am131 maps between am2 and am1.