Independent integration of rodent identifier (ID) elements into orthologous sites of some RT6 alleles of Rattus norvegicus and Rattus rattus

The two alleles of the rat T-cell differentiation alloantigen RT6 are highly divergent and their expression is distinctively regulated. While the majority of T cells of RT6a/RT6b heterozygous laboratory rats (Rattus norvegicus) expresses both alleles, a subpopulation expresses only RT6b. To identify cis-regulatory elements that potentially control monoallelic expression, we compared the sequences of both alleles. A striking difference is the presence or absence of a rodent identifier (ID) sequence in intron 7. All investigated inbred RT6a rat strains (n=7) had this integration, while it was absent in all investigated RT6b rats (n=9). An ID element was also identified at precisely the same integration site in one RT6 allele of the closely related species Rattus rattus. The ID elements of both species showed nucleotide substitutions characteristic of different subfamilies, and their flanking repeats differed in length, indicating that two independent integration events had occurred into the same site adjacent to a mammalianwide interspersed repeat. Analysis of the surrounding sequences did not disclose any motifs to explain preferential integration into one allele. Our data indicate that the RT6 alleles diverged about 1 myr ago and that the ID element integrated into the RT6a locus soon after this. We have previously shown that DNA methylation plays an important role in regulating monoallelic RT6 expression. The possibility that the ID element in the RT6a allele interferes with the required demethylation process and thus accounts for monoallelic expression is discussed.     

Restriction fragment length polymorphism of the murine C4 and Slp genes: two C4 groups.

We have examined the related H-2 genes coding for the fourth component of complement (C4) and the sex-limited protein (Slp) from 30 inbred mouse strains by Southern blot analysis. With four restriction enzymes, 11 RFLP patterns distributed among 26 different H-2 haplotypes have been identified. Strains of the same serologic H-2 haplotype were found to have identical RFLP patterns. It was confirmed that the number of C4-related genes in most haplotypes is two, Slp and C4; but H-2SWI6 (SWI6) and SWI9, which have the same RFLP pattern, have four and Sw7 has five. Although C4 and Slp have many similarities, they also were found to contain distinctive features: relative to Slp, each C4 allele examined has two insertions totaling 1.1 kb located in introns 14 and 15; and each Slp allele examined, excluding hybrids, has a provirus insertion upstream. No other large deletions or insertions were detected. The RFLP patterns are also due to 10 polymorphic restriction sites, which have been placed on standard maps; two are associated with Slp and eight are associated with C4.Sk strains, the only strains that express low serum levels of C4, have the same RFLP phenotype as Sw14, Sw18, and Swx; Sk may have arisen from a recent common ancestor of these strains. Homologous recombination has been important in the formation of existing C4 alleles. However, based on complete linkage disequilibrium between three RFLP internal to C4, the haplotypes have been divided into two groups that may have functional significance.     

Recovery and characterization of hybrid dysgenesis-induced mei-9 and mei-41 alleles of Drosophila melanogaster

X-Linked methyl methanesulfonate (MMS)-sensitive mutations were induced with hybrid dysgenesis using four P strains: pi 2, Harwich, T-007 and OK-1. Mutations were identified after two generations of backcrosses to M strain females to replace the autosomes. Among 51,471 X-chromosomes examined 10 carried stable MMS-sensitive mutations representing 8 independent events. Males of the mutant strains failed to induce gonadal dysgenesis in crosses to Oregon-R females at 28.5 degrees C. Complementation tests showed that 3 of the induced mutations were mei-9 alleles, 2 were mei-41 alleles, 1 was a mus102 allele, and 2 were alleles at a newly identified MMS-sensitive locus, mus112 (map position: 1-32.8). As assayed by in situ hybridization on polytene chromosomes, each X-chromosome had no more than four P element insertions. 4 of the 8 mutations recovered in this study proved to have P element insertions at or very close to sites to which MMS sensitivity has been mapped. Hybrid dysgenesis-induced reversion of 2 mutants, mei-9RT1 and mei-41RT2, is associated with the loss of the P element from regions 4B and 14C respectively.     

Analysis of ALS5 and ALS6 allelic variability in a geographically diverse collection of Candida albicans isolates

The Candida albicans ALS (agglutinin-like sequence) gene family encodes eight cell-surface glycoproteins, some of which function in adhesion to host surfaces. ALS genes have a central tandem repeat-encoding domain comprised entirely of head-to-tail copies of a conserved 108-bp sequence. The number of copies of the tandemly repeated sequence varies between C. albicans strains and often between alleles within the same strain. Because ALS alleles can encode different-sized proteins that may have different functional characteristics, defining the range of allelic variability is important. Genomic DNA from C. albicans strains representing the major genetic clades was PCR amplified to determine the number of tandemly repeated sequence copies within the ALS5 and ALS6 central domain. ALS5 alleles had 2-10 tandem repeat sequence copies (mean=4.82 copies) while ALS6 alleles had 2-8 copies (mean=4.00 copies). Despite this variability, tandem repeat copy number was stable in C. albicans strains passaged for 3000 generations. Prevalent alleles and allelic distributions varied among the clades for ALS5 and ALS6. Overall, ALS6 exhibited less variability than ALS5. ALS5 deletions can occur naturally in C. albicans via direct repeats flanking the ALS5 locus. Deletion of both ALS5 alleles was associated particularly with clades III and SA. ALS5 exhibited allelic polymorphisms in the coding region 5' of the tandem repeats; some alleles resembled ALS1, suggesting recombination between these contiguous loci. Natural deletion of ALS5 and the sequence variation within its coding region suggest relaxed selective pressure on this locus, and that Als5p function may be dispensable in C. albicans or redundant within the Als family.     

Analysis of the rat major histocompatibility system by Southern blot hybridization

The organization of the rat major histocompatibility complex, RT1, was studied at the DNA level by Southern blot hybridization. Genomic DNA from eight different RT1 congenic rat strains was digested by various restriction enzymes and was hybridized under stringent conditions with probes of mouse class I and class II H-2 genes. Few cross-hybridizing DNA fragments, showing no polymorphism, were seen with class II A alpha and A beta probes. The class I probes allowed for the distinction of about 8 to 19 cross-hybridizing bands, which exhibited extensive polymorphism. With the use of five RT1 recombinants, about 20% of the DNA fragments could be mapped to the RT1.A region, which codes for the ubiquitously expressed class I antigens, and about 80% to the RT1.C region-determining class I-like antigens, which are different from RT1.A antigens with respect to tissue distribution, restriction function in immune responses, and allograft rejection. The number of class I genes present in the rat genome and the possible relationship of RT1.C to H-2Qa, Tla of the mouse are discussed.     

Quantitative Trait Locus Analysis of Plasma Cholesterol and Triglyceride Levels in C57BL/6J × RR F2 Mice

A highly significant cholesterol quantitative trait locus (QTL) (Cq6) was identified on chromosome 1 in C57BL/6J x RR F2 mice. The Cq6 was located over the gene for apolipoprotein A-Il (Apoa2), and the RR allele was associated with increased plasma cholesterol. C57BL/6J has Apoa2a alleles and RR has Apoa2b alleles. Three different Apoa2 alleles are known on the basis of amino acid substitutions at four residues. Analysis with partial Apoa2 congenic strains possessing Apoa2a, Apoa2b, and Apoa2C alleles revealed that the Apoa2b allele is unique in the ability to increase cholesterol among the three Apoa2 alleles, and that the Ala-to-Val substitution at residue 61 may be crucial as far as cholesterol metabolism is concerned. We also investigated the question of whether the Apoa1 gene is responsible for the cholesterol QTLs (Cq4 and Cq5) that had been identified previously on chromosome 9 in C57BL/6J x KK-Ay/a F2 and in KK x RR F2, but not in C57BL/6J x RR F2 mice. Similar to Apoa2 alleles, three different Apoal alleles with two successive amino acid substitutions were revealed among the strains. However, we could not correlate Apoal polymorphisms with the occurrence of QTLs in these three sets of F2 mice.     

Polymorphic class II sequences linked to the rat major histocompatibility complex (RT1) homologous to human DR and DQ sequences

Until recently, the analysis of Class II genes linked to the rat major histocompatibility complex, RT1, has been confined to serologic and electrophoretic analysis of their gene products. To obtain a more definitive estimate of the number and relative polymorphism of RT1 Class II sequences, we performed Southern blot analysis of rat genomic DNA employing human cDNA probes specific for Class II heavy and light chain genes. Southern blots of EcoRI and BamHI digests of genomic DNA from ten inbred strains, expressing eight RT1 haplotypes, were hybridized with the human DQ beta or DR beta cDNA that are homologous to Class II light chain sequences. Four to eight bands were observed to hybridize with the light chain cDNA: band sizes ranged from 2.5 to 28 kb. Restriction fragment patterns were polymorphic; the only identical patterns observed were those associated with RT1 haplotypes with identical RT1.B regions. The number and size of bands hybridizing with DQ beta and DR beta suggested a minimum of four light chain sequences in each haplotype. Southern blots of BamHI and EcoRI digests of genomic DNA from the same strains were hybridized with a DR alpha cDNA that is homologous to Class II heavy chain sequences. All RT1 haplotypes expressed either a 10.0-kb or 13.0-kb band when digested with BamHI, and either a 17-kb or 3.7-kb band when digested with EcoRI. Considerably less polymorphism was detected with the DR alpha probe; this observation is consistent with previously reported limited protein polymorphism of the rat equivalent of the I-E alpha subunit. The size and number of bands hybridizing with the DR alpha probe suggests a minimum of two heavy chain sequences. These observations suggest that the RT1 complex includes more Class II sequences than have been observed in serologic and electrophoretic analyses of Class II gene products. Furthermore, the level of polymorphism of RT1 Class II sequences appears to be comparable with mouse and human Class II sequences.     

Genetic determinants of acute hypoxic ventilation: patterns of inheritance in mice.

Acutely lowering ambient O(2) tension increases ventilation in many mammalian species, including humans and mice. Inheritance patterns among kinships and between mouse strains suggest that a robust genetic influence determines individual hypoxic ventilatory responses (HVR). Here, we tested specific genetic hypotheses to describe the inheritance patterns of HVR phenotypes among two inbred mouse strains and their segregant and nonsegregant progeny. Using whole body plethysmography, we assessed the magnitude and pattern of ventilation in C3H/HeJ (C3) and C57BL/6J (B6) progenitor strains at baseline and during acute (3-5 min) hypoxic [mild hypercapnic hypoxia, inspired O(2) fraction (FI(O(2))) = 0.10] and normoxic (mild hypercapnic normoxia, FI(O(2)) = 0.21) inspirate challenges in mild hypercapnia (inspired CO(2) fraction = 0.03). First- and second-filial generations and two backcross progeny were also studied to assess response distributions of HVR phenotypes relative to the parental strains. Although the minute ventilation (VE) during hypoxia was comparable between the parental strains, breathing frequency (f) and tidal volume were significantly different; C3 mice demonstrated a slow, deep HVR relative to a rapid, shallow phenotype of B6 mice. The HVR profile in B6C3F(1)/J mice suggested that this offspring class represented a third phenotype, distinguishable from the parental strains. The distribution of HVR among backcross and intercross offspring suggested that the inheritance patterns for f and VE during mild hypercapnic hypoxia are consistent with models that incorporate two genetic determinants. These results further suggest that the quantitative genetic expression of alleles derived from C3 and B6 parental strains interact to significantly attenuate individual HVR in the first- and second-filial generations. In conclusion, the genetic control of HVR in this model was shown to exhibit a relatively simple genetic basis in terms of respiratory timing characteristics.     

Genetics and bone. Using the mouse to understand man

The rationale for use of inbred strains of mice in bone research is well recognized and includes: a) practical factors (economics of scale, rapid development of adult status, pre-existing knowledge, down-sized technologies) and b) proven methodologies for genetic studies (polygenic trait analyses, mapping tools, genomic sequencing, methods for gene manipulation). Initial investigations of inbred strains of mice showed that femoral and lumbar vertebral volumetric bone mineral density (BMD, mg/mm(3)) by pQCT varied in excess of 50% for femurs and 9% in vertebral BMD. Two strains - low BMD C57BL/6J (B6) mice and high BMD C3H/HeJ (C3H) - were investigated for insights to their BMD diversity. B6C3F2 females derived from intercrossing B6C3F1s were raised to adult skeletal status at 4 months, then necropsied for phenotyping of bone and genotyping of genomic DNA. 1000 F2 females were genotyped for PCR product polymorphisms on all 19 autosomes at approximately 15 cM. Genome wide analyses for genotype-phenotype correlations showed 10 chromosomes (Chrs) carried genes for femoral and 7 Chrs for vertebral BMD. LOD scores ranged from 2.90 to 24.4, and percent of F2 variance accounted for ranged from 1 to 10%. Analyses of main effects revealed both dominant-recessive and additive inheritance patterns. Both progenitor strains carried alleles with positive and negative effects on BMD of each bone sites. A remarkable array of additonal skeletal phenotypes (femur and vertebral geometry, strength measures, serum markers) also proved polygenic in nature, with complex segregation patterns. Verification of BMD quantitative trait loci (QTLs) was undertaken by creating congenic B6 strains carrying individual QTL regions from C3H. Following 6 cycles of backcrossing a QTL-containing region from C3H to the B6 strain, N6F2 congenic strain mice were aged to 4 months, then genotyped for the QTL region and phenotyped for skeletal traits. Comparison of mice homozygous for C3H alleles versus homozygous for B6 alleles in the QTL regions showed that femoral BMD increased or decreased significantly in congenic strains, as was predicted from F2 data. Gender differences specific to BMD QTLs have been revealed, as have more than 30 additional phenotypes associated with cortical and trabecular structural parameters and biomechanical properties.     

The genetics of glutamic-pyruvic transaminase in mice: Inheritance,electrophoretic phenotypes,and postnatal changes

Glutamic-pyruvic transaminase (GPT, E.C. 2.6.1.2) from 18 inbred strains of mice was subjected to starch gel electrophoresis. Two electrophoretic phenotypes were observed: a fast-migrating pattern in 16 strains and a slower-migrating pattern in two strains. A comparison of electrophoretic patterns of F₁ and backcross progeny of two strains of mice showed that the inheritance of GPT is autosomal with two codominant alleles. The genetic locus for GPT is designated Gpt-1, and its two alleles are designated Gpt-1 ^a and Gpt-1 ^b to represent the fast-migrating (A) and slow-migrating (B) patterns. The GPT was expressed in 11 tissues with different amounts of enzyme activity. Developmental studies of GPT activity in liver showed that between 5 and 12 days after birth the mean activity was 10 units/g protein. Between 12 and 19 days, a dramatic rise in activity occurred and adult values of 300 units/g protein were reached by 26 days.This research was supported by The National Foundation (CRBS-258) and the National Institutes of Health (GM15253).Preliminary results were reported at the Annual Meeting of the American Society of Human Genetics, October 11–14, 1972, in Philadelphia.R. P. D. is an investigator of the Howard Hughes Medical Institute.     

Restriction pattern of the major outer-membrane protein gene provides evidence for a homogeneous invasive group among ruminant isolates of Chlamydia psittaci.

Thirty-six ruminant isolates of Chlamydia psittaci, previously classified as invasive or non-invasive in a mouse model of virulence, were compared by analysing AluI restriction patterns of the major outer-membrane protein (MOMP) gene after DNA amplification by the polymerase chain reaction. The 24 invasive isolates, although from various origins, all belonged to serotype 1 and represented a strictly homogeneous group sharing a specific MOMP-gene restriction pattern that was not observed in the non-invasive strains. On the other hand, the 12 non-invasive strains, although all belonging to serotype 2, constituted a heterogeneous group with eight distinct MOMP-gene restriction patterns. However, all eight patterns shared a 180 bp fragment or the corresponding restricted fragments of 110 and 70 bp. MOMP-gene restriction patterns also clearly distinguished the ruminant strains from an avian C. psittaci isolate, a C. pneumoniae isolate and two C. trachomatis isolates which were studied for comparison. The homogeneous character of the invasive C. psittaci strains argues strongly for their genetic relatedness. Our results illustrate the usefulness of the MOMP-gene restriction mapping in typing chlamydiae.     

Origins of mouse inbred strains deduced from whole-genome scanning by polymorphic microsatellite loci

Microsatellite loci are uniformly distributed at approximately 100-kbp intervals on all chromosomes except the chromosome Y, and genetic information about more than 9000 loci and high-throughput polymorphism analysis are now available. Taking advantage of these properties, we carried out whole-genome scanning using eight common inbred strains (CIS) of laboratory mice, including A/J, C57BL/6J, CBA/J, DBA/2J, SM/J, SWR/J, NC/Nga, and 129/SvJ, and eight wild-derived inbred strains (WIS), BGL2/Ms, CAST/Ei, JF1/Ms, MSM/Ms, NJL/Ms, PGN2/Ms, SK/CamEi, and SWN/Ms. We selected and located 1226 informative loci at 1.2-cM average intervals on all of the chromosomes of the 16 strains and compared the polymorphisms of the eight CIS with those from the eight WIS as subspecies representatives. More than 50% of the loci can be identified as WIS (therefore, subspecies-specific) alleles in the CIS genomes. We also discovered that the CIS chromosomes form a mosaic structure with an average ratio of domesticus to non-domesticus alleles of 3:1. Furthermore, the domesticus alleles were present much more frequently on the CIS chromosome X than on their autosomes, suggesting that successive backcrossing of non-domesticus stocks to domesticus stocks had been undergone at the beginning of CIS history.     

Synergistic gene interactions control the induction of the mitochondrial uncoupling protein (Ucp1) gene in white fat tissue

Among a selected group of mouse strains susceptible to dietary obesity, those with an enhanced capacity for Ucp1 and brown adipocyte induction in white fat preferentially lost body weight following adrenergic stimulation. Based on the generality of this mechanism for reducing obesity, a genetic analysis was initiated to identify genes that control brown adipocyte induction in white fat depots in mice. Quantitative trait locus (QTL) analysis was performed using the variations of retroperitoneal fat Ucp1 mRNA expression in progeny of genetic crosses between the A/J and C57BL/6J parental strains and selected AXB recombinant inbred strains. Three A/J-derived loci on chromosomes 2, 3, and 8 and one C57BL/6J locus on chromosome 19 were linked to Ucp1 induction in retroperitoneal fat. Although A/J-derived alleles seemed to contribute to elevated Ucp1 expression, the C57BL/6J allele on chromosome 19 increased Ucp1 mRNA to levels higher than parental values. Thus, novel patterns of C57BL/6J and A/J recombinant genotypes among the four mapped loci resulted in a transgressive variation of Ucp1 phenotypes. Although the extent of the interchromosomal interactions have not been fully explored, strong synergistic interactions occur between a C57BL/6J allele on chromosome 19 and an A/J allele on chromosome 8. In addition to selective synergistic interactions between loci, variations in recessive and dominant effects also contribute to the final levels of Ucp1 expression.     

Esterases in the house fly. Polymorphisms and inheritance patterns.

Polyacrylamide gel electrophoresis was used to examine the variability and inheritance of esterases in five strains of the house fly, Musca domestica L. Individual zymograms exhibited 8 to 15 bands that could be assigned to one of five zones designated as A through E from anode to cathode. Correlations of P1-F1 banding patterns indicated the existence of at least 3 different loci in zone A. 2 each in zones B and C, and 4 in zone D; no clear inheritance patterns were discernable for the bands of zone E. Only the Es-5 locus of zone C was monomorphic in all of the strains studied. Eight loci possessed null alleles and codominant alleles were detected at six loci. The results suggest that esterases should prove useful for measuring relationships among fly populations or for various studies of population dynamics.     

Evidence for extensive polymorphism of RT1 Class II molecules in the rat

RT1.B Class II molecules, comparable to I-A and I-E molecules in mice, have been characterized by two-dimensional (2D) gel electrophoresis for seven rat strains expressing different RT1 haplotypes. RT1.B molecules were immunoprecipitated from radiolabeled rat lymphocyte preparations with mouse monoclonal antibodies reactive with mouse I-Ab antigens and I-Ek antigens. For each RT1 haplotype, two I-A-like subunits, alpha and beta, and two I-E-like subunits, gamma and delta, have been identified. 2D gel analysis of RT1.B molecules from different strains in paired combinations reveals that, with a few notable exceptions, the alpha, beta, and delta subunits associated with different RT1 haplotypes are characterized by unique 2D gel spot patterns. In contrast, the gamma subunits of all RT1.B haplotypes appear identical. This evidence confirms and extends our previous analysis of RT1.B molecules by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and suggests a high level of polymorphism for genes in the RT1.B region, with the exception of the RT1.Bc gene encoding gamma. Interestingly, three rat strains appear exceptional in the identity of their alpha, gamma, and delta subunits, which suggests that these strains express closely related RT1 haplotypes derived from a common progenitor.     

γ-Glutamyl Cyclotransferase: A New Genetic Polymorphism in the Mouse (MUS MUSCULUS) Linked to LYT–2

Electrophoretically variant forms of gamma-glutamyl cyclotransferase have been identified in red cells of inbred mouse strains. Each inbred strain exhibited a major band of activity and a minor band that migrated more anodally. The polymorphism affects the migration of both the major and minor bands in a similar way. F1 hybrids between strains with fast forms (A/J) and strains with the slow forms (C57BL/6J) exhibited a four-banded pattern consistent with co-dominant inheritance. The patterns observed in backcross and F2 mice were consistent with the segregation of a pair of autosomal co-dominant alleles. Recombinant inbred strains and a congenic strain were used to show that the locus controlling gamma-glutamyl cyclotransferase (Ggc) is linked to Lyt-2, a lymphocyte alloantigen locus on chromosome 6, with an estimated map distance of 5.0 +/- 2.5 centimorgans.     

Crystal structures of two rat MHC class Ia (RT1-A) molecules that are associated differentially with peptide transporter alleles TAP-A and TAP-B

Antigenic peptides are loaded onto class I MHC molecules in the endoplasmic reticulum (ER) by a complex consisting of the MHC class I heavy chain, beta(2)-microglobulin, calreticulin, tapasin, Erp57 (ER60) and the transporter associated with antigen processing (TAP). While most mammalian species transport these peptides into the ER via a single allele of TAP, rats have evolved different TAPs, TAP-A and TAP-B, that are present in different inbred strains. Each TAP delivers a different spectrum of peptides and is associated genetically with distinct subsets of MHC class Ia alleles, but the molecular basis for the conservation (or co-evolution) of the two transporter alleles is unknown. We have determined the crystal structures of a representative of each MHC subset, viz RT1-A(a) and RT1-A1(c), in association with high-affinity nonamer peptides. The structures reveal how the chemical properties of the two different rat MHC F-pockets match those of the corresponding C termini of the peptides, corroborating biochemical data on the rates of peptide-MHC complex assembly. An unusual sequence in RT1-A1(c) leads to a major deviation from the highly conserved beta(3)/alpha(1) loop (residues 40-59) conformation in mouse and human MHC class I structures. This loop change contributes to profound changes in the shape of the A-pocket in the peptide-binding groove and may explain the function of RT1-A1(c) as an inhibitory natural killer cell ligand.     

Allotypes of mouse complement component C6 in inbred strains and some wild populations

This collaborative work was undertaken to resolve discrepancies in reports of the number of forms of complement component C6 present in the circulation of mice from various inbred strains. Plasma C6 was analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate and by isoelectric focusing (IEF), and C6 band patterns were developed by electroblotting and immunoprobing. Results of C6 allotyping of mice from 36 strains confirmed that while 20 strains (prototype strain BALB/c) possessed only one relative mass (M _r) form which typed C6A1 on IEF, the other 16 strains all possessed more than one C6 M _r form. Moreover, IEF analysis demonstrated additional polymorphic differences; among these 16 strains, 11 typed C6A l B l like the prototype strain CBA, the AKR and RF/J strains typed C6A2B2, and the Japanese MOM strain as well as the C57BR/cdJ and C57L/J strains possessed two forms with IEF mobilities intermediate between C6A1B1 and C6A2B2. These will now be referred to as C6A3B3. Thus, a total of four different mouse C6 haplotypes have been identified.Testing C6 allotypes in a limited number of wild mice revealed that haplotypes found in inbred strains of Western or Eastern origin tend to reflect haplotypes of the wild mice from Europe or Japan, respectively.     

The effects of genetic variation in N-acetyltransferases on 4-aminobiphenyl genotoxicity in mouse liver

Inbred, congenic and transgenic strains of mice were characterized for acetylation of p-aminobenzoic (PABA) and the carcinogen 4-aminobiphenyl (4ABP). C57Bl/6 mice have the NAT2*8 allele, A/J mice have NAT2*9 and congenic B6.A mice have NAT2*9 on the C57Bl/6 background. The first transgenic strain with human NAT1, the functional equivalent of murine NAT2, was also tested. The murine NAT2*9 allele correlated with a slow phenotype measured with the murine NAT2 selective substrate PABA. The two strains having this allele also had a lower capacity to acetylate 4ABP. A line with five copies of the human NAT1 transgene was bred for at least five generations with either C57Bl/6 or A/J mice. There was no significant change in PABA NAT activity on the C57Bl/6 background but a 2.5-fold increase was seen in hNAT1:A/J compared with A/J. The effect of variation in NATs on 4ABP genotoxicity was assessed in these strains. Twenty-four hours after exposure to a single oral dose of 120 mg 4ABP/kg, hepatic 4ABP-DNA adducts were detected by immunofluoresence in all strains. Nuclear fluorescence intensities (mean+/-S.D.) were 41.1+/-3.6 for C57Bl/6, 37.9+/-1.11 for A/J and 36.3+/-2.44 for B6.A. There was no correlation between murine NAT2 alleles and 4ABP-DNA adduct levels. Similar results were seen with the transgenic strains. The data indicate that the range of variation present in these strains of mice was insufficient to alter susceptibility to 4ABP genotoxicity. The impact of these relatively modest differences in the acetylation of the activation of 4ABP may be masked by other competing biotransformation reactions since 4ABP is a substrate for both NAT1 and NAT2. Mouse models with variation in both isoforms are needed to adequately assess the role of variation in NATs in susceptibility to 4ABP genotoxicity.