- THIS ARTICLE
-
Abstract
- Full Text (PDF)
- Alert me when this article is cited
- Alert me if a correction is posted
- SERVICES
- Similar articles in this journal
- Similar articles in PubMed
- Alert me to new issues of the journal
- Download to citation manager
- Reprints & Permissions
- CITING ARTICLES
- Citing Articles via HighWire
- Citing Articles via Google Scholar
- GOOGLE SCHOLAR
- Articles by Butchko, R. A. E.
- Articles by Keller, N. P.
- Search for Related Content
- PUBMED
- PubMed Citation
- Articles by Butchko, R. A. E.
- Articles by Keller, N. P.
Aspergillus nidulans Mutants Defective in stc Gene Cluster Regulation
Robert A. E. Butchkoa, Thomas H. Adams1,b, and Nancy P. Kelleraa Department of Plant Pathology and Microbiology, Texas A&M University, College Station, Texas 77843-2132
b Department of Biology, Texas A&M University, College Station, Texas 77843-2132
Corresponding author: Nancy P. Keller, Department of Plant Pathology and Microbiology, Texas A&M University, 120 Peterson Bldg., College Station, TX 77843-2132., n-keller{at}tamu.edu (E-mail)
Communicating editor: M. E. ZOLAN
| ABSTRACT |
|---|
The genes involved in the biosynthesis of sterigmatocystin (ST), a toxic secondary metabolite produced by Aspergillus nidulans and an aflatoxin (AF) precursor in other Aspergillus spp., are clustered on chromosome IV of A. nidulans. The sterigmatocystin gene cluster (stc gene cluster) is regulated by the pathway-specific transcription factor aflR. The function of aflR appears to be conserved between ST- and AF-producing aspergilli, as are most of the other genes in the cluster. We describe a novel screen for detecting mutants defective in stc gene cluster activity by use of a genetic block early in the ST biosynthetic pathway that results in the accumulation of the first stable intermediate, norsolorinic acid (NOR), an orange-colored compound visible with the unaided eye. We have mutagenized this NOR-accumulating strain and have isolated 176 Nor- mutants, 83 of which appear to be wild type in growth and development. Sixty of these 83 mutations are linked to the stc gene cluster and are likely defects in aflR or known stc biosynthetic genes. Of the 23 mutations not linked to the stc gene cluster, 3 prevent accumulation of NOR due to the loss of aflR expression.
THE filamentous fungus Aspergillus nidulans produces a carcinogenic polyketide called sterigmatocystin (ST), the penultimate precursor in the aflatoxin (AF) biosynthetic pathway found in the related species A. parasiticus, A. flavus, and A. nomius (![]()
![]()
![]()
![]()
The ST biosynthetic genes (stc genes) are clustered on chromosome IV of A. nidulans (![]()
![]()
![]()
![]()
![]()
![]()
![]()
![]()
![]()
![]()
![]()
Although we have identified two members of a G protein signal transduction pathway that are required for normal aflR expression and activity and hence ST biosynthesis (![]()
![]()
![]()
![]()
![]()
![]()
![]()
and ß subunits, respectively, of a fatty acid synthase (![]()
| MATERIALS AND METHODS |
|---|
Fungal strains and growth conditions:
All fungal strains are listed in Table 1. Fungal strains were maintained on appropriately supplemented minimal medium (![]()
![]()
|
Nucleic acid manipulations:
stcE was disrupted by homologous integration of plasmid pJH3. A 3.98-kb EcoRV/HindIII fragment from pAHK25 (![]()
![]()
stcE background. One para-aminobenzoic acid auxotrophic progeny, RJH007 (Table 1) was used for diploid analysis. Disruption of stcE in RJH007 was confirmed by visualization of NOR and Southern analysis.
For mRNA extraction, wild-type and mutant strains were grown in 5 ml of liquid stationary complete medium (minimal medium with 2.0% dextrose, 0.2% peptone, 0.1% yeast extract, and 0.1% casein hydrolysate) in a 20-ml screw-cap vial. Cultures were set at a 45° angle to increase surface area and then incubated at 37° for 3 days after which fungal mycelium was harvested and frozen in liquid nitrogen. RNA was extracted with Trizol (GIBCO, Grand Island, NY) following the manufacturer's instructions. Total RNA (20 µg) was run on a 1.2% agarose gel containing 4% formaldehyde (v/v) and transferred to Hybond-N+ (Amersham, Arlington Heights, IL) nylon membrane following the manufacturer's instructions. DNA fragments from pAHK25 (a 1.29-kb XbaI/EcoRV fragment containing the aflR gene) and from pAHK27 (a 2.27-kb KpnI fragment containing the stcF, stcG, and stcH genes) were labeled with 32P and hybridized to the nylon membranes at 60° for 16 hr. Membranes were washed for 30 min each in 2x SSC 0.1% SDS, 1x SSC 0.1% SDS, and 0.1x SSC 0.1% SDS at 60° and X-ray film was exposed overnight at -80°.
Mutagenesis:
Conidia of TJH3.40 were mutagenized with 4-nitro-1-quinoline oxide (NQO) as previously described by ![]()
Genetic techniques:
Linkage of NQO-generated mutations to the stc cluster was determined by crossing each MRB strain to a stcE wild-type strain (FGSC89) and analyzing the meiotic progeny for accumulation of NOR. Linkage of the new mutation to the stc gene cluster would be indicated by an absence or low percentage of NOR-accumulating progeny due to a low frequency of recombination between the mutated gene and the stcE deletion. Mutations were determined to be unlinked to the stc gene cluster if 25% of the progeny were able to accumulate NOR. Approximately 200 meiotic progeny were analyzed from each cross.
The number of linkage groups or loci represented in the mutations unlinked to the stc cluster was determined by crossing each mutant in every pairwise combination. Each mutant was first crossed to FGSC410 to introduce a new genetic marker, in this case a mutation in the riboB gene, into the mutant background. Progeny from crosses between the original mutants and the recombinants were scored for NOR production. Two mutations were considered to be in the same or closely linked genes if none or a low percentage (<6%) of progeny accumulated NOR. If ~25% of the progeny accumulated NOR, the mutations were considered to be in different genes.
Diploids were generated between MRB strains and RJH007 to determine the dominant or recessive nature of each mutation. Diploids were tested on solid oatmeal medium for the accumulation of NOR and haploidization was induced by incorporation of 9 µl benomyl (2 mg/ml) into solid media.
NOR assays:
NOR production was assessed by growing all strains at 37° in liquid and solid media. All mutants were visually screened on solid oatmeal medium containing 1.0% oatmeal and appropriate supplements. The mutants not linked to the stc gene cluster were further assessed by growing them on solid oatmeal media at 30°, 37°, and 42° and on pH 4 and pH 8 media at 37°. This same subset of mutants was cultured on three different liquid media: (1) complete medium, (2) complete medium plus oatmeal, and (3) oatmeal. Mycelium was harvested after 3 days of stationary culture, frozen, lyophilized, crushed in a 1.7-ml centrifuge tube, and extracted with 750 µl chloroform for 10 min. An equal volume of water was added and the samples were vortexed and spun at 13,000 rpm for 10 min to separate the aqueous and organic phases. The organic phase was removed, dried, and resuspended in 50 µl chloroform. Twenty-five microliters of each extract was separated on a silica gel TLC plate in toluene:ethyl acetate:acetic acid (80:10:10) and visualized with long-wave UV light.
| RESULTS |
|---|
Isolation of Nor- mutants:
We employed a visual screen for the identification of mutants defective in stc gene cluster activity. The mutants identified by this screen for NOR accumulation fell into two broad classes; 93 showed gross morphological defects and 83 exhibited growth and development resembling the original TJH3.40 strain. The 83 mutants whose growth and development resembled that of the original TJH3.40 strain were chosen for further analysis as we know that morphological mutants in A. nidulans are also defective in stc gene cluster activity (![]()
Linkage of mutations to the stc gene cluster:
To date it is known that four stc genes are required for NOR production, aflR, stcA, stcJ, and stcK. Mutations in any one of these four genes would lead to the loss of NOR production. Also, other genes have been found in the A. parasiticus and A. flavus aflatoxin clusters that are involved in aflatoxin production in these species and we believe it is possible that homologous genes that may be required for NOR production in A. nidulans exist in the region of the stc cluster. Therefore, MRB strains containing mutations linked to the cluster could have mutations in one of the four known stc cluster genes or in unidentified genes such as homologs of genes in the aflatoxin gene cluster. Linkage to the stc gene cluster was determined by crossing each MRB strain to a stcE wild-type strain (ST accumulating) and analyzing meiotic progeny for accumulation of NOR. MRB strains containing aflR, stcA, stcJ, and stcK mutations should produce a low percentage of NOR-accumulating progeny as such mutations would be closely linked to the stcE::argB locus. MRB strains containing mutations unlinked to the stc gene cluster would produce 25% NOR-accumulating progeny when crossed to a wild-type strain. Examination of 100200 ascospore progeny of 60 of the 83 mutants showed between 0.99 and 6.2% recombination, suggesting that these 60 are linked to the stc gene cluster and represent candidate mutations in aflR, stcA, stcJ, stcK, or another linked locus.
Determination of linkage groups among the 23 unlinked mutants:
To determine the number of genes or loci represented in the 23 unlinked mutations, the mutants were crossed in every pairwise combination. This required first crossing each MRB strain to FGSC410 to introduce another genetic marker, a riboflavin mutation, into the mutant strains, thus generating RRB1 through RRB23. Each RRB strain, containing the stcE::argB2 disruption, the new mutation, and riboB2, was crossed to the 23 MRB mutants, and progeny were analyzed for the accumulation of NOR. As in the above test, NOR accumulation in 25% of the MRB x RRB progeny indicated that the mutations were in two separate linkage groups. If none of the progeny accumulated NOR, this suggested that the mutations were in the same or closely linked loci. From this analysis we determined that 5 mutants, MRB246, 278, 288, 308, and 379, contain mutations that are closely linked (and are therefore possibly allelic), whereas the remaining 18 mutants represented unique loci (Fig 1). In certain cases (indicated by an asterisk) viable ascospores were never generated after multiple attempts to cross two strains. We are unable to determine the genetic interaction in such cases.
|
mRNA and NOR analysis:
To determine whether the 23 unlinked mutations could define potential regulators/activators of aflR expression or AflR activity, we grew the mutants in three different media and analyzed culture extracts for NOR production and mRNA for aflR and stc gene transcripts. Cultures grown on liquid oatmeal in this manner did not produce enough mycelium for mRNA extraction, but organic extraction of NOR was possible. For 20 of the strains, transcript accumulation varied depending on growth medium. TLC analysis showed that all of these 20 mutants produced trace amounts of NOR that was visually undetectable in the oatmeal plate assay (Table 2). Only 3 mutants, MRB230, MRB234, and MRB300, consistently failed to produce NOR on 1.0% oatmeal plates or detectable aflR and stc transcripts (Fig 2). These 3 mutants represent the best candidates for nonconditional regulators of aflR.
|
|
In addition, all 23 mutants were tested for temperature and pH sensitivity. ST and AF biosynthesis has been shown to be pH regulated (![]()
![]()
|
|
Dominant/recessive nature of mutations:
Diploids between MRB strains and another stcE disruption strain, RJH007, were created to determine the dominant or recessive nature of each mutation. For the 23 unlinked mutations, only one diploid (MRB265 x RJH007) did not produce NOR, indicating that MRB265 contains a dominant mutation. aflR and stc gene cluster transcripts were detectable in MRB265 (data not shown).
| DISCUSSION |
|---|
In an effort to reduce or eliminate AF contamination from food and feed supplies we are concentrating on understanding the genetic regulation of the Aspergillus gene cluster required for mycotoxin biosynthesis. A positively acting transcription factor present in the gene cluster, AflR (![]()
![]()
![]()
![]()
![]()
The design of the screen developed here ensured that we would obtain several different categories of Nor- mutants, some of which would be defective in genes already found to be required for NOR production, including flbA, fadA, aflR, and three ST biosynthetic genes, stcA (encoding a polyketide synthase) and stcJ and stcK (encoding the
and ß subunits, respectively, of a fatty acid synthase). Through genetic analysis, the mutations were determined to be linked or unlinked to the stc gene cluster. We are interested in characterizing those that are unlinked and thus represent the best possibilities for novel trans-acting regulators of ST/AF biosynthesis.
To assess how many loci are represented in the group of 23 unlinked mutants, meiotic progeny from each pairwise combination of mutants were analyzed for NOR production. This analysis revealed one linkage group that contained 5 mutants (MRB246, 278, 288, 308, and 379). MRB246, in contrast to the other four strains, was not able to grow at 42°, MRB308 did not express aflR in complete medium, and MRB278 grew poorly at pH 8. These differences could reflect distinct types of mutations in one gene or mutations in different, tightly linked genes. Interestingly, the remaining 18 mutants each fell into unique linkage groups, thus suggesting that our screen was not saturated. The high number of linkage groups may reflect the fact that a vast number of factors have been reported to influence AF production in A. flavus and A. parasiticus (![]()
![]()
![]()
Finally, to further characterize the unlinked mutants in terms of the regulation of expression of the stc gene cluster, we looked for aflR and other stc gene cluster transcripts in these mutants. Only 3 of the 23 mutants lacked aflR and stc gene expression in all conditions tested. These 3 mutants fall into different linkage groups and appear to be specific regulators of aflR. The other 20 mutations likely affect some other aspect of ST biosynthesis, as aflR and other stc cluster genes were transcribed in one or more of the three media tested (Table 2). Perhaps these 20 mutations represent unlinked genes that are necessary for allocation of resources (e.g., acetyl CoAs) for polyketide formation or represent independent regulators of aflR. The mutations may block the availability of resources for the ST pathway or the localization of stc gene products to subcellular sites as is described for the penicillin biosynthetic pathway (![]()
Aflatoxin contamination remains a problem worldwide and we believe that logical AF control strategies will be developed only through understanding the genetic regulation of the biosynthesis of this mycotoxin. To date we have found that all aspects of ST regulation in A. nidulans have been conserved with regard to AF regulation in the aflatoxigenic Aspergillus spp. including aflR function. Therefore it is likely that factors involved in the regulation of aflR should also be conserved in all species. If a novel approach to regulating ST production by regulating the expression of aflR exists, this might be applicable to controlling AF contamination in agricultural products.
| FOOTNOTES |
|---|
1 Present address: Cereon Genomics, LLP, Cambridge, MA 02139. ![]()
| ACKNOWLEDGMENTS |
|---|
We thank Julie K. Hicks and Tiffany Sanford for construction of TJH3.40 and Elspeth Sharp for technical assistance. This work was supported by grant no. 96-35303-3415 from U.S. Department of AgricultureCooperative State Research Service to N.P.K. and T.H.A., by grant no. DGE-9354891 from National Science Foundation to N.P.K., T.H.A., and R.A.E.B., and by the Tom Slick Fellowship, Texas A&M College of Agriculture and Life Science, to R.A.E.B.
Manuscript received January 20, 1999; Accepted for publication June 7, 1999.
| LITERATURE CITED |
|---|
ANDERSON, J. A., C. H. CHUNG, S. H. CHO, and R. H. ADAMSON, 1990 Versicolorin A hemiacetal, hydroxydihydrosterigmatocystin, and aflatoxin G2 alpha reductase activity in extracts from Aspergillus parasiticus. Induction of hepatocellular carcinoma in nonhuman primates by chemical carcinogens. Mycopathologia 111:39-45[Medline].
BRESSAC, B., M. KEW, J. WANDS, and M. OZTURK, 1991 Selective G to T mutations of p53 gene in hepatocellular carcinoma from Southern Africa. Nature 350:429-431[Medline].
BROWN, D. W., T. H. ADAMS, and N. P. KELLER, 1996a Aspergillus has distinct fatty acid synthases for primary and secondary metabolism. Proc. Natl. Acad. Sci. USA 93:14873-14877
BROWN, D. W., J.-H. YU, H. S. KELKAR, M. FERNANDES, and T. C. NESBITT et al., 1996b Twenty-five coregulated transcripts define a sterigmatocystin gene cluster in Aspergillus nidulans.. Proc. Natl. Acad. Sci. USA 93:1418-1422
BURSE, B., A. DMOCHOWSKA, M. SKRZYPEK, P. WEGLENSKI, and M. A. BATES et al., 1983 Cloning and characterization of the ornithine carbamoyltransferase gene from Aspergillus nidulans.. Gene 25:109-117[Medline].
CARDWELL, K. and J. D. MILLER, 1996 Mycotoxins in foods in Africa. Natural Toxins 4:103-107[Medline].
CHANG, P.-K., J. W. CARY, D. BHATNAGAR, T. E. CLEVELAND, and J. W. BENNETT et al., 1993 Cloning of the Aspergillus parasiticus apa-2 gene associated with the regulation of aflatoxin biosynthesis. Appl. Environ. Microbiol. 59:3273-3279
COLE, R. J., and R. H. COX, 1981 Handbook of Toxic Fungal Metabolites, pp. 6772. Academic Press, New York.
FERNANDES, M., N. P. KELLER, and T. H. ADAMS, 1998 Sequence-specific binding by Aspergillus nidulans AflR, a C6 zinc cluster protein regulating mycotoxin biosynthesis. Mol. Microbiol. 28:1355-1365[Medline].
HICKS, J. K., J.-H. YU, N. P. KELLER, and T. H. ADAMS, 1997 Aspergillus sporulation and mycotoxin production both require inactivation of the FadA G-alpha protein-dependent signaling pathway. EMBO J. 16:4916-4923[Medline].
KÄFER, E., 1977 Meiotic and mitotic recombination in Aspergillus and its chromosomal aberrations, pp. 3367 in Advances in Genetics, edited by E. W. CASPARI. Academic Press, New York.
KELKAR, H. S., T. W. SKLOSS, J. F. HAW, N. P. KELLER, and T. H. ADAMS, 1997 Aspergillus nidulans stcL encodes a putative P-450 monooxygenase required for bisfuran desaturation during aflatoxin/sterigmatocystin biosynthesis. J. Biol. Chem. 272:1589-1594
KELLER, N. P. and T. M. HOHN, 1997 Metabolic pathway gene clusters in filamentous fungi. Fungal Genet. Biol. 21:17-29[Medline].
KELLER, N. P., R. A. E. BUTCHKO, B. SARR, and T. D. PHILLIPS, 1994 A visual pattern of mycotoxin production in maize kernels by Aspergillus spp. Phytopathology 84:483-488.
KELLER, N. P., T. C. NESBITT, B. SARR, T. D. PHILLIPS, and G. B. BUROW, 1997 pH regulation of sterigmatocystin and aflatoxin biosynthesis in Aspergillus spp. Phytopathology 87:643-648[Medline].
LEE, L. S., J. W. BENNETT, L. A. GOLDBLATT, and R. E. LUNDIN, 1971 Norsolorinic acid from a mutant strain of Aspergillus parasiticus.. J. Am. Oil Chem. Soc. 48:93-94[Medline].
MULLER, W. H., R. A. L. BOVENBERG, M. H. GROOTHUIS, F. KATTEVILDER, and E. B. SMAAL et al., 1992 Involvement of microbodies in penicillin biosynthesis. Biochim. Biophys. Acta 1116:210-213[Medline].
PAPA, K. E., 1982 Norsolorinic acid mutant of Aspergillus flavus.. J. Gen. Microbiol. 128:1345-1348.
PAYNE, G. A., G. J. NYSTROM, D. BHATNAGAR, T. E. CLEVELAND, and C. P. WOLOSHUK, 1993 Cloning of the afl-2 gene involved in aflatoxin biosynthesis from Aspergillus flavus.. Appl. Environ. Microbiol. 59:156-162
PONTECORVO, G., J. A. ROPER, L. M. HEMMONS, K. D. MACDONALD, and A. W. J. BUFTON, 1953 The genetics of Aspergillus nidulans.. Adv. Genet. 5:141-238[Medline].
SCHROEDER, H. W. and H. HEIN, 1967 Aflatoxins: production of the toxins in vitro in relation to temperature. Appl. Microbiol. 15:441-445[Medline].
WIESER, J., B. N. LEE, J. W. FONDON, and T. H. ADAMS, 1994 Genetic requirements for initiating asexual development in Aspergillus nidulans.. Curr. Genet. 27:62-69[Medline].
YU, J.-H. and T. J. LEONARD, 1995 Sterigmatocystin biosynthesis in Aspergillus nidulans requires a novel type I polyketide synthase. J. Bacteriol. 177:4792-4800
YU, J.-H., R. A. E. BUTCHKO, M. FERNANDES, N. P. KELLER, and T. J. LEONARD et al., 1996 Conservation of structure and function of the aflatoxin regulatory gene aflR from Aspergillus nidulans and A. flavus.. Curr. Genet. 29:549-555[Medline].
ZAIKA, L. L. and R. L. BUCHANAN, 1987 Review of compounds affecting the biosynthesis or bio-regulation of aflatoxins. J. Food Prot. 50:691-708.
This article has been cited by other articles:
![]() |
T. M. Hammond, J. W. Bok, M. D. Andrewski, Y. Reyes-Dominguez, C. Scazzocchio, and N. P. Keller RNA Silencing Gene Truncation in the Filamentous Fungus Aspergillus nidulans Eukaryot. Cell, February 1, 2008; 7(2): 339 - 349. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. V. Roze, R. M. Beaudry, A. E. Arthur, A. M. Calvo, and J. E. Linz Aspergillus Volatiles Regulate Aflatoxin Synthesis and Asexual Sporulation in Aspergillus parasiticus Appl. Envir. Microbiol., November 15, 2007; 73(22): 7268 - 7276. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. K. Shwab, J. W. Bok, M. Tribus, J. Galehr, S. Graessle, and N. P. Keller Histone Deacetylase Activity Regulates Chemical Diversity in Aspergillus Eukaryot. Cell, September 1, 2007; 6(9): 1656 - 1664. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Marquez-Fernandez, A. Trigos, J. L. Ramos-Balderas, G. Viniegra-Gonzalez, H. B. Deising, and J. Aguirre Phosphopantetheinyl Transferase CfwA/NpgA Is Required for Aspergillus nidulans Secondary Metabolism and Asexual Development Eukaryot. Cell, April 1, 2007; 6(4): 710 - 720. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. M. Hammond and N. P. Keller RNA Silencing in Aspergillus nidulans Is Independent of RNA-Dependent RNA Polymerases Genetics, February 1, 2005; 169(2): 607 - 617. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. H. Wilkinson, A. Ramaswamy, S. C. Sim, and N. P. Keller Increased conidiation associated with progression along the sterigmatocystin biosynthetic pathway Mycologia, November 1, 2004; 96(6): 1190 - 1198. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. W. Bok and N. P. Keller LaeA, a Regulator of Secondary Metabolism in Aspergillus spp. Eukaryot. Cell, April 1, 2004; 3(2): 527 - 535. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Kato, W. Brooks, and A. M. Calvo The Expression of Sterigmatocystin and Penicillin Genes in Aspergillus nidulans Is Controlled by veA, a Gene Required for Sexual Development Eukaryot. Cell, December 1, 2003; 2(6): 1178 - 1186. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Shimizu and N. P. Keller Genetic Involvement of a cAMP-Dependent Protein Kinase in a G Protein Signaling Pathway Regulating Morphological and Chemical Transitions in Aspergillus nidulans Genetics, February 1, 2001; 157(2): 591 - 600. [Abstract] [Full Text] |
||||
![]() |
N. P. Keller, C. M. H. Watanabe, H. S. Kelkar, T. H. Adams, and C. A. Townsend Requirement of Monooxygenase-Mediated Steps for Sterigmatocystin Biosynthesis by Aspergillus nidulans Appl. Envir. Microbiol., January 1, 2000; 66(1): 359 - 362. [Abstract] [Full Text] |
||||
- THIS ARTICLE
-
Abstract
- Full Text (PDF)
- Alert me when this article is cited
- Alert me if a correction is posted
- SERVICES
- Similar articles in this journal
- Similar articles in PubMed
- Alert me to new issues of the journal
- Download to citation manager
- Reprints & Permissions
- CITING ARTICLES
- Citing Articles via HighWire
- Citing Articles via Google Scholar
- GOOGLE SCHOLAR
- Articles by Butchko, R. A. E.
- Articles by Keller, N. P.
- Search for Related Content
- PUBMED
- PubMed Citation
- Articles by Butchko, R. A. E.
- Articles by Keller, N. P.






