In vivo mutation at the human HPRT locus.

The molecular nature of mutations that arise in vivo at the human hypoxanthine-guanine phosphoribosyltransferase (HPRT) locus can be determined. A wide variety of such mutations can be detected, including large and small deletions, frameshift mutations and single-base substitutions, as well as alterations that cause aberrant mRNA splicing. Here, we review the available information on mutations at this locus.

In vitro mutational spectrum of aflatoxin B1 in the human hypoxanthine guanine phosphoribosyltransferase gene.

The in vitro mutational spectrum of aflatoxin B1 (AFB1) in exon 3 of the human hypoxanthine guanine phosphoribosyltransferase gene in B-lymphoblasts was examined by a combination of polymerase chain reaction and denaturing gradient gel electrophoresis. The cell line used in this study contained an expression vector that produced high levels of human cytochrome P450 CYP1A1. CYP1A1 metabolizes AFB1 to form an epoxide intermediate which can react with DNA. About 1200 independent mutants were induced at the hypoxanthine guanine phosphoribosyltransferase locus by AFB1 and were selected en masse by addition of 6-thioguanine to the bulk culture. Two independent cultures were treated with AFB1. Polymerase chain reaction was used to amplify exon 3 from the complex mutant population, and denaturing gradient gel electrophoresis was used to separate wild-type DNA sequences from mutant sequences. Mutational hotspots were visible as discrete bands on the denaturing gradient gel. Scanning densitometry was used to determine the fraction of the complex population that was represented in each non-wild-type band. The bands containing the mutations were excised from the denaturing gradient gel and sequenced. In this way, the nature and frequency of mutational hotspots in a population of > 1000 mutants were determined. AFB1 produced one strong mutational hotspot in exon 3. Between 10 and 17% of the AFB1-induced mutants contained a single GC-->TA base substitution at base pair 209. This hotspot occurred in a GGGGGG sequence (the mutated base is underlined). This mutation was observed reproducibly in two independently treated cultures. Several other mutations were observed in only one culture but at a lower frequency. Our results are the first report of the mutational spectrum of AFB1 in a native human gene.

Database and software for the analysis of mutations in the human p53 gene.

Mutations of the human p53 gene are of importance in the development of cancer. Perhaps 50% of all human cancers contain a mutation in the p53 oncogene and many laboratories are investigating mutations at this locus. In an effort to centralize and standardize the information regarding human p53 mutations, we have created a computerized database that contains information about DNA sequence alterations for > 3000 p53 mutants. Information on the cancer type, the origin of the cells, the specific mutation, the amino acid change, the literature citation, and other data are provided for each mutant. We have also produced a software package for the analysis of the p53 database. Routines have been developed for the analysis of single-base substitutions, including programs to (a) determine whether two mutational spectra are different, (b) display the number of mutations and mutable sites in each exon, (c) determine whether mutations show a DNA strand bias, (d) determine the frequency of transitions and transversions, (e) display the number and kind of mutations observed at each base in the coding region, (f) perform nearest neighbor analysis, and (g) display mutable amino acids in the p53 protein. The software runs only on IBM-compatible machines with MS-DOS. The software and p53 database are freely available via the Internet, using the remote file transfer protocol. These programs simplify the analysis of the rapidly increasing body of information about p53 mutations. The programs permit facile comparison between different p53 data sets, as well as the identification of mutational patterns that may be of importance to experimenters studying the mechanisms of mutation and the etiology of cancers.

In vitro mutational specificity of cisplatin in the human hypoxanthine guanine phosphoribosyltransferase gene.

The in vitro mutational spectra of cisplatin [cis-diamminedichloroplatinum(II)] in exon 3 of the human hypoxanthine guanine phosphoribosyltransferase gene in B-lymphoblasts was examined by a combination of polymerase chain reaction and denaturing gradient gel electrophoresis. Several thousand independent mutants were induced at the hypoxanthine guanine phosphoribosyltransferase locus by cisplatin and were selected en masse by addition of 6-thioguanine to the bulk culture. Polymerase chain reaction was used to amplify exon 3 from the complex mutant population, and denaturing gradient gel electrophoresis was used to separate wild-type DNA sequences from mutant sequences. Mutational hotspots were visible as discrete bands on the denaturing gradient gel. Scanning densitometry was used to determine the fraction of the complex population represented by the novel bands. The mutant bands were excised from the denaturing gradient gel and sequenced. In this way, the nature and frequency of mutational hotspots in a population of several thousand mutants were determined. Cisplatin produced several mutational hotspots in exon 3. About 9-10% of the cisplatin-induced mutants had mutations in a GGGGGG sequence (base pairs 207-212). GC----AT substitutions at the second and third guanines in the 5'-GGGGGG-3' run made up about 2 and 4% of the induced mutants, respectively. About 4% of the induced mutants contained a GC----TA substitution at the sixth guanine. About 1% of the cisplatin-induced mutants had an AT----TA transversion in a TAGA sequence (base pair 271; mutated base is underlined). Our results are consistent with mutations occurring at GpG and ApG sites. These nucleotide sequences have been identified as the primary sites of cisplatin adduction.

Analysis of mutations occurring at the human hprt locus.

We have recently established a computerized database containing information on mutants at the human hypoxanthine guanine phosphoribosyl transferase (hprt) locus. The database contains sequence information on over 1000 mutants. We now present an analysis of the information in the database. 542 single base substitution mutants in the hprt coding region exist, and we have examined (1) the number of mutations and the number of mutable sites in each exon, (2) transcribed versus non-transcribed strand bias for mutations, (3) the frequency of the 5' and 3' nearest neighbors to a mutated base, and (4) the distribution of amino acid substitutions. The distribution of both DNA mutations and amino acid mutations was not uniform, several clusterings of mutations were observed and we propose several possible mechanisms to account for the hotspots. We also examined mRNA splicing mutants, mutants with small deletions, and frameshift mutants.

Single base mutations in the human androgen receptor gene causing complete androgen insensitivity: rapid detection by a modified denaturing gradient gel electrophoresis technique.

Mutations of the human androgen receptor (AR) gene impair normal sexual differentiation and development in karyotypic males, resulting in a spectrum of external genital phenotypes ranging from complete female to nearly complete male. Identification and characterization of these mutations can provide valuable information regarding the functional importance of specific amino acids of the AR. To screen for point mutations in the AR gene underlying the phenotypic abnormalities in the androgen insensitivity syndrome (AIS), the eight exons of the AR gene were amplified from genomic DNA using the polymerase chain reaction (PCR) and analyzed by denaturing gradient gel electrophoresis. A computer program, MELTMAP, was used to identify optimum sequences for denaturing gradient gel electrophoresis, and mutation detection sensitivity was enhanced by forming heteroduplexes between control and subject PCR products to create base mismatches. In seven families with complete AIS, single base mutations were found in the region of the AR gene encoding the steroid-binding domain of the receptor. The mutations that converted amino acid 774 from Arg to His and amino acid 864 from Asp to Gly were recreated using site-directed mutagenesis and the mutant ARs expressed in COS 7 and CV1 cells. In both cases, abnormalities of androgen binding and transcriptional activation were consistent with the observed sex phenotype. These results together with others reported previously demonstrate that single amino acid changes within the region encoded by exons D to H of the AR gene can alter androgen binding and are a common cause of complete androgen resistance. The strategy used herein, employing denaturing gradient gel analysis of heteroduplex PCR products, provides a valuable aid to rapid detection of single base mutations in AIS.

Deletion mutagenesis during polymerase chain reaction: dependence on DNA polymerase.

Polymerase chain reaction (PCR) was performed with two polymerases. Thermus aquaticus DNA polymerase (Taq), and modified T7 DNA polymerase (Sequenase). Both polymerases were used to amplify the same portion of the human 18S rRNA gene. We report a PCR artifact, namely a deletion of 54 bp, when Taq polymerase was used to amplify a portion of the human 18S rRNA gene. PCR performed with Sequenase did not produce this artifact. The deletion eliminated a potentially stable hairpin loop. Our data are consistent with the following model for generation of the deletion: (i) the formation of an intrastrand hairpin, and (ii) polymerization across the base of the hairpin, thus deleting the nucleotide sequence in the hairpin. Furthermore, we show that the deletion occurs mainly during synthesis of the (-)DNA strand. Our observations suggest that similar artifacts may occur in other sequences containing stable secondary structures.

Visualization and quantitation of peroxisomes using fluorescent nanocrystals: treatment of rats and monkeys with fibrates and detection in the liver.

Peroxisome proliferation in the liver is a well-documented response that occurs in some species upon treatment with hypolipidemic drugs, such as fibrates. Typically, liver peroxisome proliferation has been estimated by direct counting via electron microscopy, as well as by gene expression, enzyme activity, and immunolabeling. We have developed a novel method for the immunofluorescent labeling of peroxisomes, using an antibody to the 70-kDa peroxisomal membrane protein (PMP70) coupled with fluorescent nanocrystals, Quantum Dots. This method is applicable to standard formalin-fixed, paraffin-embedded tissues. Using this technique, a dose-dependent increase in PMP70 labeling was evident in formalin-fixed liver sections from fenofibrate-treated rats. In formalin-fixed liver sections from cynomolgus monkeys given ciprofibrate, quantitative image analysis showed a statistically significant increase in PMP70 labeling compared to control; the increase in hepatic PMP70 protein levels was corroborated by immunoblotting using total liver protein. An increase in hepatic peroxisome number in ciprofibrate-treated monkeys was confirmed by electron microscopy. An advantage of the Quantum Dot/PMP70 method is that a single common protocol can be used to label peroxisomes from several different species, and many of the common problems that arise with immunolabeling, such as fading and low signal strength, are eliminated.

Database and software for the analysis of mutations at the lacZ locus in transgenic rodents.

The use of transgenic rodents is becoming increasingly widespread in genetic toxicology. In an effort to centralize and standardize the information regarding mutations in rodents bearing the lacZ transgene, we have created a computerized database that contains published information about DNA sequence alterations on over 100 mutants. Information on the literature citation, mutagenic conditions, organs from specific animals, mutation frequency in each organ, specific mutation, amino acid change, and other data are provided for each mutant. We have also produced a software package for the analysis of the lacZ database. Routines have been developed for the analysis of single base substitutions, including programs to 1) determine whether two mutational spectra are statistically different, 2) determine whether mutations show a DNA strand bias, 3) determine the frequency of transitions and transversions, 4) display the number and kind of mutations observed at each base in the coding region, 5) perform nearest-neighbor analysis, and 6) display mutable amino acids in the lacZ protein. The software runs only on IBM-compatible machines running Microsoft Windows. The software and lacZ database are freely available via the Internet (http:@sunsite.unc.edu/dnam/mainpage.ht ml). These programs simplify the analysis of the rapidly increasing information about lacZ mutation. The programs permit the facile comparison between different lacZ data sets as well as the identification of mutational patterns that may be of importance to experimenters studying the mechanisms of mutation and mutational spectra in transgenic animals.

Database and software for the analysis of mutations at the lacI gene in both transgenic rodents and bacteria.

The use of transgenic rodents for the study of genetic toxicology has increased dramatically in the past several years. A great deal of the recent work has employed the lacI locus in transgenic mice. In addition to the transgenic data, a substantial amount of information exists regarding mutation of the lacI gene in bacteria. In an effort to centralize the information regarding mutations in the lacI gene in both rodents and bacteria, we have created a computerized database that contains information about DNA sequence alterations on about 500 mutations in transgenic rodents and 8,000 mutations in bacteria. We have also produced a software package for the analysis of the lacI database. Routines have been developed for the analysis of single base substitutions, including programs to (i) determine if two mutational spectra are different; (ii) determine if mutations show a DNA strand bias; (iii) determine the frequency of transitions and transversions; (iv) display the number and kind of mutations observed at each base in the coding region; (v) perform nearest neighbor analysis; and (vi) display mutable amino acids in the lacI protein. The software runs only on IBM-compatible machines running Microsoft Windows. The software and lacI database are freely available via the internet (http:/(/)sunsite.unc.edu/dnam/mainpage.++ +html).

The Ames miniscreen assay: volatility of sodium azide can cause an increase in the reversion frequencies of adjacent, untreated wells.

Sodium azide, when added to wells adjacent to untreated wells, caused an increase in the reversion rate of Salmonella typhimurium TA100 in a 12-well plate format. Increases in the reversion frequency in adjacent, untreated wells were observed when a single well on the plate was treated with as little as 1 microg of sodium azide. This effect is probably caused by the hydrolysis of sodium azide to form hydrazoic acid. Hydrazoic acid has a boiling point of 37 degrees C and, thus, would become a diffusible gas during the incubation of the plates. Our findings suggest that a diffusible gas is present and that this gas has the ability to contaminate nearby wells when using the multiwell version of the Ames assay. Furthermore, it may be prudent to isolate all positive controls and negative controls on separate plates with no test material since a volatile test material could produce spurious results in the Ames miniscreen.

Analysis of mutations using PCR and denaturing gradient gel electrophoresis.

Denaturing gradient gel electrophoresis (DGGE) separates DNA molecules based on primary sequence. Under the appropriate conditions, all base pair (bp) substitutions, frame-shifts, and deletions less than about 10 bp can be resolved from the wild type sequence using DGGE. Polymerase chain reaction (PCR) permits facile amplification of a given region of the genome. We have combined PCR and DGGE to: (i) Localize mutations in the X-linked human androgen receptor gene. PCR/DGGE was used to screen the individual exons in the 2757-bp coding region of the gene in afflicted individuals as well as in potential carriers. Inheritance of a mutant allele has been demonstrated in several cases; (ii) Analyze thousands of thioguanine-resistant mutants simultaneously. The in vitro mutational spectra of MNNG, ICR-191, and cisplatin at the human HPRT locus have been examined by this method. The compounds all have mutational hotspots in a GGGGGG sequence in exon 3; however, the particular mutations induced by the agents were different; (iii) Examine the fidelity of several DNA polymerases used in PCR. The fidelity of Thermus aquaticus DNA polymerase (Taq) is 1-2 x 10(-4) misincorporations/bp/replication. Problems with Taq polymerase arise in the analysis of complex mutant populations by DGGE because the Taq-induced errors reduce the sensitivity of the system. To circumvent this, it had been necessary to use Sequenase, a modified T7 DNA polymerase with a higher fidelity. However, Sequenase is not thermostable and must be added every PCR cycle. A thermostable DNA polymerase from Thermococcus litoralis (Vent) is now available, and we have examined the fidelity of Vent, Taq, and Sequenase polymerase in PCR using DGGE. The fidelity of Vent, Taq, and Sequenase polymerase was 2.4 x 10(-5), 8.9 x 10(-5), and 4.4 x 10(-5) errors/bp, respectively. Vent polymerase had the highest fidelity of the three enzymes tested.

Human HPRT mutant database: software for data entry and retrieval.

We have developed a computer database containing information on over 1,000 human hypoxanthine guanine phosphoribosyl transferase (HPRT) mutants. Both published and unpublished data are present. The database itself is maintained in a dBASE format (.DBF) and we provide a set of programs to examine and extract information from the database. A program to input information into the database is also supplied. The database and programs are available directly from us or via remote FTP (file transfer protocol) using BITNET/INTERNET. All programs require an IBM-compatible computer, the MS-DOS operating system (version 3.3 or greater), and a hard disk with about 5 megabytes of free disk space. The purpose of the database is 1) to allow investigators to contribute their HPRT mutants directly to the database in a standardized fashion, and 2) to allow access to the entire database with a set of programs that allows manipulation and extraction of data. For example, using our programs it is possible to i) order the database by base pair position, ii) examine only information regarding mutagenesis by a particular agent, iii) search for a particular author, iv) create a report which contains selected portions of the database, the report can be printed or saved as a file. The database will be updated every several months and distributed.

Software for the analysis of mutations at the human hprt gene.

Mutations at the human hypoxanthine-guanine phosphoribosyl transferase gene (hprt) are currently of great interest because mutations at this locus are being used as a biomonitor of human mutagenic exposure. Not only can somatic hprt mutants arising in vivo in humans be recovered and sequenced, but there is also a considerable body of information about the in vitro mutational spectra of different carcinogens at this locus. Previously, we reported the creation of a computerized database containing DNA-sequence information on human hprt mutants (Cariello et al. (1992) Environ. Mol. Mutagen., 20, 81-83). In the present manuscript, software for the analysis of mutations in the hprt database is described. Numerous routines have been developed for the analysis of single-base substitutions, including programs to (i) determine if two mutational spectra are different, (ii) display the number of mutations and mutable sites in each exon, (iii) determine if mutations show a DNA-strand bias, (iv) determine the frequency of transitions and transversions, (v) display the number and kind of mutations observed at each base in the coding region, (vi) perform nearest-neighbor analysis and (vii) display mutable amino acids in the hprt protein. The software runs only on IBM-compatible machines with MS-DOS. The software and hprt database is freely available via the INTERNET using remote file-transfer protocol. These programs simplify the analysis of the rapidly increasing information about hprt mutation. The programs permit the facile comparison between in vitro and in vivo data, as well as the identification of mutational patterns that may be of importance to experimenters using hprt as a biomonitor and and of importance to researchers studying mechanisms of mutation.

Mutational analysis using denaturing gradient gel electrophoresis and PCR.

Denaturing gradient gel electrophoresis (DGGE) separates (DNA) molecules based on their sequence. Using the proper conditions, all base-pair substitutions can be resolved from the wild-type sequence using DGGE. Polymerase chain reaction (PCR) permits rapid amplification of a given region of the genome. In this paper, we demonstrate the utility of DGGE combined with PCR for mutation analysis by presenting different examples: (i) analysis of mouse p53 cDNA for mutations, (ii) simultaneous analysis of thousands of 4NQO-induced mutants for mutations in HPRT exon 3, (iii) examination of the fidelity of the thermostable DNA polymerase isolated from Pyrococcus furiosus (Pfu), (iv) purification of mutant DNA from contaminating wild-type DNA from mouse spleenic T-cell clones.

The Ames test: the two-fold rule revisited.

Mutagenicity in the Ames assay is evaluated by comparing the number of revertants observed in treated cultures to those in untreated cultures. Often, some form of the '2-fold rule' is employed, whereby a compound is judged mutagenic if a 2-fold or greater increase is seen in a treated culture. In order to understand the underpinnings of this approach, we study some of its statistical properties. We assume that the number of revertants on any plate from a given two-group experiment follows a Poisson distribution and we address the following questions: (1) what is the false-positive error probability of observing at least a doubling of the number of colonies from the control to the treatment group?; (2) if a given mean number of colonies is postulated for a control group, what number of colonies above the observed control mean provides a false-positive rate of 5%? We also present results for question 1 in the case where the number of revertants follows a negative binomial distribution.

Molecular analysis of complex human cell populations: mutational spectra of MNNG and ICR-191.

We describe a method to identify and enumerate mutants at the nucleotide level in complex cell populations. Several thousand different mutants were induced at the HPRT locus in human lymphoblastoid cultures by either MNNG, an alkylating agent, or by ICR-191, a substituted acridine. HPRT mutants were selected en masse by resistance to 6-thioguanine. The most frequent mutations (hotspots) in HPRT exon 3 were determined by a combination of denaturing gradient gel electrophoresis and polymerase chain reaction. MNNG predominantly produced GC----AT transitions at nucleotides in a GGGGGG sequence, while ICR-191 produced both +1 frameshifts in the same GGGGGG sequence and +1 frameshifts in a CCC sequence.

A novel bacterial reversion and forward mutation assay based on green fluorescent protein.

We report the first use of green fluorescent protein (GFP) for mutation detection. We have constructed a plasmid-based bacterial system whereby mutated cells fluoresce and non-mutated cells do not fluoresce. Fluorescence is monitored using a simple hand-help UV lamp; no additional cofactors or manipulations are necessary. To develop a reversion system, we introduced a +1 DNA frameshift mutation in the coding region of GFP and the resulting protein is not fluorescent in Escherichia coli. Treatment of bacteria containing the +1 frameshift vector with ICR-191 yields fluorescent colonies, indicating that reversion to the wild-type sequence has occurred. Site-directed mutagenesis was used to insert an additional cytosine into a native CCC sequence in the coding region of GFP in plasmid pBAD-GFPuv, expanding the sequence to CCCC. A dose-related increase in fluorescent colonies was observed when the bacteria were treated with ICR-191, an agent that induces primarily frameshift mutations. The highest dose of ICR-191 tested, 16 microg/ml, produced a mutant fraction of 16 x 10(-5) and 8.8 x 10(-5) in duplicate experiments. The reversion system did not respond to MNNG, an agent that produces mainly single-base substitutions. To develop a forward system, we used GFP under the control of the arabinose PBAD promoter; in the absence of arabinose, GFP expression is repressed and no fluorescent colonies are observed. When cells were treated with MNNG or ENNG, a dose-dependent increase in fluorescent colonies was observed, indicating that mutations had occurred in the arabinose control region that de-repressed the promoter. Treating bacteria with 100 microg/ml MNNG induced mutant fractions as high as 82 x 10(-5) and 40 x 10-5 in duplicate experiments. Treating bacteria with 150 microg/ml ENNG induced a mutant fraction of 2.1 x 10(-5) in a single experiment.

Correlation analysis of gene expression and clinical chemistry to identify biomarkers of skeletal myopathy in mice treated with PPAR agonist GW610742X.

Data from individual animals were used to identify genes in mouse skeletal muscle whose expression correlated with a known serum marker of skeletal myopathy, creatine kinase activity (CK), after treatment with a peroxisome proliferator-activated receptors (PPAR) agonist, GW610742X. Six genes had correlation coefficients of >or=0.90: Mt1a (metallothionein 1a), Rrad (Ras-related associated with diabetes), Ankrd1 (ankyrin repeat domain 1), Stat3 (signal transducer and activator of transcription 3), Socs3 (suppressor of cytokine signalling 3) and Mid1ip1 (Mid1 interacting protein 1). The physiological function of these genes provides potentially useful information relating to the mechanism of PPAR-induced skeletal myopathy, with oxidative stress and disruption of glycolysis most closely associated with myopathic damage. Some of the muscle genes most highly correlated with serum CK in mice also appear to be good indicators of PPAR-induced myopathy in rat skeletal muscle, demonstrating the translational potential of this approach. This study clearly shows the utility of using correlation analysis as a simple tool for identifying novel biomarkers and investigating mechanisms of toxicity.

Database and software for the analysis of mutations at the human hprt gene.

A computerized database containing DNA sequence information regarding human HPRT mutants has been created. The database itself is in the dBASE format and contains information on about 1500 mutants. In addition, an IBM PC compatible software package to analyze the information in the database has been developed. Both the database and software are freely available via the Internet.