A single major gene controls most of the difference in susceptibility to streptozotocin-induced diabetes between C57BL/6J and C3H/HeJ mice

Diabetologia, Oct 1989

Summary To assess genetic factors determining sensitivity to streptozotocin-induced diabetes in inbred strains of mice, a genetic analysis of streptozotocin-sensitive C57BL/6J and streptozotocin-resistant C3H/HeJ mice was performed. One week after a single dose of streptozotocin (200 mg/kg body weight), differences in plasma glucose concentration were marked between male mice of the C57BL/6J and C3H/HeJ strains (p<0.001). To determine the number of genes responsible for the difference, F1 male progeny of a cross between parental strains were produced, and found to be streptozotocin resistant like C3H/HeJ parents. F1 mice were, therefore, backcrossed with streptozotocin-sensitive C57BL/6J mice (Backcross: F1♂ ♂ X C57BL/6J ♀ ♀). The plasma glucoses of backcrossed male mice (n=41) following streptozotocin treatment appeared to segregate into two populations, half like the C57BL/6J parent, and half like the F1 parent. Statistical analysis of the data revealed that the data fit a model with two distributions better than one with a single distribution, suggesting a single major gene responsible for the difference in streptozotocin susceptibility. This hypothesis was also supported by the observation that streptozotocin sensitivity in 12 recombinant inbred strains of C57BL/6J and C3H/HeJ mice appeared to segregate into two classes. Resistance to streptozotocin induced diabetes in F1 mice suggested that the expression of this gene is recessive, although X-chromosome linked inheritance could not be excluded. Efforts to map the streptozotocin-sensitivity gene revealed lack of right linkage to several loci including the H-2 locus. If inherited differences in the ability to resist a B-cell toxin play a role in genetic susceptibility to diabetes in man, then mapping the streptozotocin-susceptibility gene in mice may provide a means to evaluate the role of a putative homologous locus in the aetiology of diabetes in man.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://link.springer.com/content/pdf/10.1007%2FBF00274530.pdf

A single major gene controls most of the difference in susceptibility to streptozotocin-induced diabetes between C57BL/6J and C3H/HeJ mice

Diabetologia A single major gene controls most of the difference in susceptibility to streptozotocin-induced diabetes between C5 7 B L / 6 J and C 3 H / H e J mice K. K a k u 1 2 J. McGilP 1 2 M. Province 0 1 2 a n d M.A. Permutt ~ 1 2 0 Divisionof Biostatistics,Department of PreventiveMedicine,WashingtonUniversitySchool of Medicine , St.Louis,Missouri , USA 1 l MetabolismDivision,Department of Internal Medicine 2 Dr. K. Kaku Metabolism Division Department of Internal Medicine Washington University School of Medicine 660 S. Euclid St. Louis, MO 63110 USA Summary. To assess genetic factors determining sensitivity to streptozotocin-induced diabetes in inbred strains of mice, a genetic analysis of streptozotocin-sensitive C57BL/6J and streptozotocin-resistant C3H/HeJ mice was performed. One week after a single dose of streptozotocin (200 mg/kg body weight), differences in plasma glucose concentration were marked between male mice of the C57BL/6J and C3H/HeJ strains (p< 0.001). To determine the number of genes responsible for the difference, F1 male progeny of a cross between parental strains were produced, and found to be streptozotocin resistant like C3H/HeJ parents. Ft mice were, therefore, backcrossed with streptozotocin-sensitive C57BL/6J mice (Backcross: F1 9 9 X C5 7BL/6J c~ d ). The plasma glucoses of backcrossed male mice (n=41) following streptozotocin treatment appeared to segregate into two populations, half like the C5 7BL/6J parent, and half like the F1 parent. Statistical analysis of the data revealed that the data fit a model with two distributions better than one with a single distribution, Streptozotocin; diabetes-susceptibility; genetic analysis; inbred mouse strains; H-2 locus - 9 Springer-Verlag1989 Streptozotocin (SZ)-induced diabetes is mediated b y pancreatic B-cell destruction accompanied by progressively severe hyperglycaemia [ 1-4 ]. The drug, a glucosamine-nitrosourea c o m p o u n d with significant antimicrobial [ 5, 6 ] and antitumor activity [ 7 ] as well as a potent diabetogenic effect [ 8 ], spontaneously decomposes to reactive m e t h y l c a r b o n i u m ions that alkylate D N A and produce interstrand crosslinks [ 9 ]. The diabetogenic effect m a y be related to the B-cell depletion o f nicotinamide adenine dinucleotide, as diabetes can be prevented by concomitant nicotinamide administration [ 10, 11 ]. The diabetogenic effects o f SZ are also related to its glocosamine configuration as the drug is preferentially taken up by B cells relative to methylnitrosourea [121. Inbred strains o f mice showed varying susceptibility to the diabetogenic effects o f SZ, administered either as suggesting a single major gene responsible for the difference in streptozotocin susceptibility. This hypothesis was also supported by the observation that streptozotocin sensitivity in 12 recombinant inbred strains of C5 7BL/6J and C3H/HeJ mice appeared to segregate into two classes. Resistance to streptozotocin induced diabetes in Ft mice suggested that the expression of this gene is recessive, although X-chromosome linked inheritance could not be excluded. Efforts to map the streptozotocin-sensitivity gene revealed lack of right linkage to several loci including the H-2 locus. If inherited differences in the ability to resist a B-cell toxin play a role in genetic susceptibility to diabetes in man, then mapping the streptozotocin-susceptibility gene in mice may provide a means to evaluate the role of a putative homologous locus in the aetiology of diabetes in man. a single dose or as multiple subdiabetogenic doses, indicating the importance o f genetic background in the pathogenesis of SZ-induced diabetes [ 3, 13 ]. Male mice are more sensitive t h a n female mice [ 14-16 ], a n d the explanation for this important sexual dimorphism has yet to be determined. For male mice of different strains, however, there is a prominent difference in SZ-sensitivity, and the genetic loci involved have not been defined. Linkage o f i m m u n e response genes within the major histocompatibility locus (H-2) has been suggested [ 17-19 ], but this finding is still uncertain [ 20-22 ]. Rossini et al. noted that a single high dose injection o f SZ induced a remarkable hyperglycaemia in C 5 7 B L / 6 J mice, while the same dose failed to produce diabetes in the C 3 H / H e J strain [3]. In the present study, the number o f modifying genes for SZ-susceptibility was assessed by measuring the effects of the drug in F1 hybrids . 4 / I 4 m a -6 E E o b 30 20 ~ 22.2 o 3 ~tI/ Weeks after SZinjection F i g . l a and b. Body weight and plasma glucose of C57BL/6J ( 0 ) , C 3 H / H e J (o) and F1 hybrid ( 9 mice at various times after streptozotocin (SZ) administration (200mg/kg) as described in Methods. Values are m e a n  (n=15) for each group. * p<0.05; 9* p < 0.001 compared with values before SZ treatment.t p < 0.001 compared with the value of F1 mice and backcrossed mice, as well as in established recombinant inbred strains o f C 5 7 B L / 6 J X C 3 H / H e J mice (BxH) [ 23 ]. The results o f these studies suggest that an allelic difference at one major locus is responsible for most o f the difference in SZ susceptibility noted between the two strains. Materials and methods Animals Mice were obtained from the Jackson Laboratories, Bar Harbor, Maine, USA. The mice were given free access to tap water and chow (Purina Rodent Chow, Ralston-Purina, St.Louis, Mo., USA) throughout the experiment. To avoid the effect associated with gender differences in pathogenesis of SZ-induced diabetes [ 14-16 ] male C57BL/6J mice were bred with female C 3 H / H e J mice to obtain F1 hybrids, and for backcrossed mice, male C5 7BL/6J mice were bred with female F1 mice. Male mice were used exclusively for analysis. All mice were 5-8 weeks of age at the time of experimentation. Streptozotocin treatment SZ was purchased from Sigma, St. Louis, Missouri, USA. SZ was dissolved in a sterile 0A mol/l sodium citrate, pH 4.5, and injected i.p. within 5 min after preparation. Responses to different doses are presented in Results. All control mice were given injections with citrate buffer alone. Plasma glucose determination and intraperitoneal glucose tolerance test Whole blood (250 lxl)collected from the retro-orbital sinus of all nonfasted animals between 09.00-11.00 hours using heparin-treated capillary tubes was analysed for plasma glucose concentration with a Beckman Glucose Analyzer II (Beckman Instruments, Inc., Arlington Heights, Ill., USA). Intraperitoneal glucose tolerance testing (IPGTI) was performed after fasting overnight (18 h) to characterize glucose tolerance of mice are described previously [ 24 ]. During the IPGTT, the most marked difference in plasma glucose had been noted at 30rain (C57BL/6J, 26.9+1.6mmol/1 vs C3H/HeJ, 1 2 . 5  p<0.001), so blood samples were obtained at 30 min after the injection of glucose (2 g/kg body weight). Determination o f restriction fragment length polymorphisms at the H-2 locus and polymorphic forms o f r-glutamyl myelotransferase To define the H-2 haplotypes of the parental strains of mice, restriction fragment length polymorphisms (RFLPs) were analysed in D N A from C 3 H / H e J (H-2k) and C57BL/6J (H-2b) mice. D N A was digested with Apa I, Barn ]HI, Dra I, Hind III, Kpn I, Pst I, Pvu II, and T a q I, and analysed by Southern blotting with a murine class II genomic probe IA as described [ 25 ]. Polymorphic r-glytamyl myelotransferase enzymic forms characteristic of C5 7BL/6J and C 3 H / H e J were determined by agarose gel electrophoresis of liver extracts according to the method of Tulchin and Taylor [ 27 ]. Statistical analysis All results are expressed as mean  SEM. Student's t-tests (two-tailed) were employed to compare differences between the measures. Correlation coefficients (r) were employed to assess the significance of linear relationship between continuous measurements, p-values are reported as measures of statistical significance. Number of mice used are in parentheses, a p<0.05; b p<0.001 VS C 3 H / H e J ; c p<0.05; d p<0.001 vs F1 Results Strain differences in S Z susceptibility Male mice of the parental strains C57BL/6J and C 3 H / H e J , and Fa hybrids ( C 3 H / H e J 9 9 X C5 7BL/6J d d ) received a single injection of SZ ranging from 0 to 300 m g / k g body weight, and plasma glucose concentrations were determined one week later. Strain differences in the diabetogenic effects of SZ are shown in Table 1. C 3 H / H e J and F1 mice were resistant to SZ at a dose of 150 m g / k g or less, while this dose produced significant hyperglycaemia in C57BL/6J mice (19.3+3.1 mmol/1 vs 11.3_+ 1.2 mmol/1, p<0.05). SZ at 200 m g / k g body weight resulted in the largest difference in plasma glucose between C57BL/6J and Ft hybrids (plasma glucose 26.3 _ 1.5 mmol vs 13.6+ 0.7 mmol/1 for C57BL/6J and F1 respectively, p < 0.001). To further define optimal conditions for observing phenotypic differences in SZ susceptibility, plasma glucose concentrations and body weights were monitored for four weeks after the injection of SZ at 200 m g / k g in each of the parental strains and in F~ hybrids. As shown in Figure 1, a decrease in body weight (p< 0.05) was observed in C57BL/6J mice within 2 weeks after the SZ injection which persisted for the duration of the experiment. The mortality o f C5 7BL/6J mice after 4 weeks was 17% (1 of 6 mice). On the other hand, C 3 H / H e J and F1 mice gained weight and were alive four weeks after SZ injection. Hyperglycaemia in C57BL/6J was apparent within 3 days after the injection (p<0.001) and was persistent. In contrast, C 3 H / H e J and F1 mice appeared to be resistant to SZ, with a significant increase in plasma glucose concentration (p<0.05) observed only 3 weeks or more after SZ injection. There were no differences in plasma glucose concentrations between C 3 H / H e J and F1 mice throughout the experiment. From these findings, the plasma glucose concentration at 1 week after the injection of SZ at 200 m g / k g body weight was chosen to evaluate the genetic differences in SZ susceptibility. Genetic analysis o f phenotypic differences in S Z susceptibility To determine the number of genes involved in the phenotypic difference in SZ susceptibility, F1 mice were backcrossed with C5 7BL/6J mice (Backcross: F1 9 9 X C57BL/6J d d ) . Each male backcrossed mouse received an IPGTT and then within several days an injection of SZ (200 m g / k g body weight). Following SZ, plasma glucose measured 11.7+0.6mmol/1 ( n = 1 5 ) , 26.6 + 1.6 mmol/1 ( n = t5), and 13.6 +_0.7 mmol/1 ( n = 15) for C 3 H / H e J , C5 7BL/6J, and Ft mice, respectively (p<0.001, C57BL/6J vs C 3 H / H e J or Ft). The plasma glucose concentrations after SZ treatment for All backcrossed mice were given an i.p. glucose tolerance test (IPGTI) one week prior to SZ as described in Methods. a For comparison of SZ sensitivity with r-glutamyl cyclotransferase (GgC) alleles or H-2 alleles, individual backcrossed mice were assigned a SZ phenotype as "B" if the plasma glucose after SZ was greater than 2 standard deviations above the mean of the F1 parents ( > 19.0 mmol/l), or "F" if less than 2 standard deviations below the mean of the C57BL/6J parents ( < 14.2 retool/l). GgC alleles were defined as "B" or "F" by electrophoresis of liver protein [271, and H2 alleles were defined as "B" or "F" by Southern blot analysis of D N A hybridized to H-2 cDNA probes after identification of parental restriction length polymorphism fragments as described in Methods ~., Iob ,I] n, , H.I 22,2 33.3 Plasma glucose (mmol/I) Fig.2a and b. Population distribution patterns and chi-square analysis on the expected as a single gene (panel a) and observed (panel b) number of plasma glucose levels after streptozotocin treatment in 41 backcross mice. T and Q2 are SKUMIX parameters (see Table 3). Dotted lines represent mean (panel a) and median (panel b) values of plasma glucose levels in F~ and C5 7BL/6J mice. The distribution of mice with plasma glucose below 13.3mmol/1 (11) between 13.3 mmol/1-28.3 retool/1 [ 18 ] and greater than 28.3 mmol/1 [ 12 ], did not differ from expected (,Z2= 0.66, d f = 2, p = 0.72) individual backcross mice, shown in Table 2, ranged from a low of 10.4 mmol/1 to a high of 36.0 mmol/1. Estimation of the number of genes controlling S Z sensitivity If sensitivity to SZ-induced diabetes was determined by a single genetic locus, then one would expect that half of the backcross progeny would be homozygous like C5 7BL/6J (AA or aa if recessive), and the other half would be heterozygous like the F1 (Aa) mice. This would produce a mixture of two distributions, creating a bimodal distribution of the backcross progeny with one mode near each parental median. If we divide the range of glucose levels into three areas defined as above, between, and below the median values of C5 7BL/6J and F1 parents, respectively, then the plasma glucose following SZ of the backcross mice would be expected to segregate in the ratio 1/4:1/2:1/4 into these three regions. Thus, one would expect the 41 backcross progeny to segregate as 10.25:20.5:10.25, whereas we observe a segregation of 11:18:12 (Fig.2b). This yields a %2of 0.66 with 2 df which is not significant (p= 0.72). Therefore, on the basis of this analysis, the data are compatible with the hypothesis of a single gene control. On the other hand, the simple %2 analysis only counts the number of points in each of the three regions, and is, therefore, ignoring much of the information contained in the continuous variation observed in actual glucose levels. For instance, if glucose levels were really under polygenic control, a unimodal distribution, in which most of the data would be between the two parental median values and fewer in the extreme regions, would be expected. If this polygenic distribution were non-normal, with particularly "fat tails" or skewed distribution, it could also show segregation proportions of nearly 1/4:1/2:1/4. To allow for this possibility, and to attempt to use all of the information available in the backcross mice data, we tested whether the data were significantly better explained as a bimodal mixture of two distributions (as expected if there is a single gene control) or whether a single, unimodal, possibly skewed distribution fitted the data better (as expected if the phenotype were under polygenic control). The program used was S K U M I X [ 26 ], and the results of this analysis are shown in Table 3. For each line of Table 3, six parameters are fitted by the method of maximum likelihood (Figure 2a): the overall mean (U) and variance (V) of the mixed distribution, the distance (T) between the means of the two distributions, the proportion of data in the rightmost distribution (Q2), and two parameters of a power transformation (P and R) applied to each component distribution to more closely transform the data to normality. Finally for each model, -2 times the loglikelihood is given (-2 lnL). For tlhe general two-distribution model (Line 1 of Table 3), Q2 is far from 0 or 1, indicating two distributions; T is large, indicating good separation of them and in fact Q2 is nearly 1/2, which is what is expected if half of the backcross mice resemble Q2 Parameters were fitted by the method of maximum likelihood for each model. V: overall variance; U: overall mean; T: distance between the two means; 02: proportion in right distribution (see Fig. 2); P and R: power transform parameters; - 2 ]nL: - 2 times the log-likelihood Kb 7.55.72.4..a o') ~ U U U U U each of their parents. If there is just one distribution, then T = 0 and Q2= 0, which we have tested in the second line of Table 3. Subtracting the -2 lnLs gives a %2of 9.95, 2 df, p = 0.007. Thus, two distributions fit significantly better than one, even allowing for skewness (P and R). On the other hand, allowing two distributions, but fixing the proportion at Q2 = 1/2 as in the last line, gives a %2 of only 0.06 compared to the general model (p= 0.81). Thus, the fit is not significantly different from the one forcing equal proportions of the two component distributions. The evidence from both tests would seem to indicate a single gene involvement in the expression of phenotype for hyperglycaemia after SZ treatment of C5 7BL/6J mice. S Z susceptibility in B x H recombinant inbred strains The recombinant inbred (RI) strain BxH, originally generated from C57BL/6J and C 3 H / H e J inbred strains and maintained by the Jackson Laboratories for more than 30generations, was utilized for further genetic analysis o f susceptibility to SZ-induced diabetes. The utility of RI strains for purposes of genetic analysis has been described [ 23 ]. RI strains are produced by random sibling mating of the F2 generation, and subsequent separation and breeding of each RI strain until homozygosity is obtained at each locus. Because of inbreeding, each RI strain is homozygous for alleles at a given locus from one parent. If SZ sensitivity were controlled by a single locus, then SZ-sensitivity in RI rats should resemble that o f one of the parent strains. Intermediate values would argue for more than a single locus. The plasma glucose concentrations 1 week after SZ treatment were determined in 5-6 mice of each of the 12 available BxH RI strains (Table 4). Only one strain (BxH-2) resembled the C 3 H / H e J progenitor with resistance to SZ, while the remaining 11 strains showed higher levels of plasma glucose resembling the C5 7BL/6J progenitor. The plasma glucose concentrations of 11 of the RI strains (BxH-3 to 19) did not differ from each other or from the C5 7BL/6J parental strain (unpaired t-test), yet were significantly higher than those o f the C 3 H / H e J parental strain and BxH-2 mice ( p < 0.001). The glucose level of BxH-2 mice was significantly lower than that of the C57BL/6J progenitor ( p < 0.001). These observations provided additional evidence that the difference in SZ susceptibility in the parental strains is mostly the result of a single major gene. Evaluating the major locus controlling S Z susceptibility relative to other loci We previously demonstrated a marked difference in glucose tolerance between these two strains and that this phenotypic difference was determined by multiple genetic loci [ 24 ]. To determine the relationship between these loci and the major locus which controls susceptibility to SZ-induced diabetes, each backcross: F1 X Values indicated are mean + SEM. N u m b e r of mice are in parentheses, a p < 0.001 vs C 3 H / H e J ; b p < 0.001 vs C5 7BL/6J Strains Parental C 3 H / H e J C57BL/6J Recombinant inbred BxH-2 BxH-3 BxH-4 BxH-6 BxH-7 BxH-8 BxH-9 BxH-10 BxH-11 BxH-12 BxH-14 BxH-19 11.9 +0.5 (15) 26.3 _+1.5 (15) 10.8 + 1.1 b (5) 20.0 + 2.8a (5) 21.3 _+2.8a (5) 25.7 _ 2.9a (6) 24.9 _ 3.6a (6) 23.6 + 4.1 a (6) 27.0 + 2.8a (6) 23.4 + 3.5 a (5) 21.3 + 3.1 a (5) 26.2 + 2.8a (5) 21.8 + 3.1 a (5) 27.8 _ 1.3 a (5) H B H B B B B B B B B B B B C57BL/6 mouse was given an IPGTT several days prior to SZ. The plasma glucose levels at 30 min on IPGTT for the parental strains and F1 hybrids were readily distinguishable with plasma glucose concentrations of 12.2+0.9mmol/1 (n=6), 26.3+2.2mmol/1 (n=6), and 15.9+1.0mmol/1 ( n = 6 ) for C3H/HeJ, C5 7BL/6J, and F1 mice, respectively (p< 0.001 for all comparisons), confirming the previous report [ 24 ]. The glucose levels of 41 backcross mice ranged in a continuum of values from a low of 11.3 mmol/1 to a high of 30.7 mmol/1 (Table 2). Analysis of this data by methods used to evaluate SZ sensitivity (i. e. comparing the number of values in backcross mice below, between, and above the median values for C5 7BL/6J and F1 parental strains) allowed rejection of a single gene hypothesis with a high degree of probability (Z2=21.3, d f = 2 , p < 0.001), again confirming the previous report [ 24 ]. Although the phenotypic patterns for SZ sensitivity corresponded to those for glucose tolerance in the parental strains, no correlation was observed between SZ sensitivity and glucose tolerance in individual backcrossed mice ( r = 0.13). The strain distribution pattern of SZ sensitivity in the BxH RI mice was compared to the strain distribution pattern of 95 phenotypic markers previously mapped in the BxH RI lines (provided by Dr. Ben Taylor, The Jackson Laboratories, Bar Harbor, Me, USA). The strain distribution pattern for SZ-sensitivity in the BxH R! lines most nearly resembled the pattern of rglutamyl cyclotransferase alleles (Ggc) on chromosome 6. For this reason the Ggc enzymic forms of the backcross progeny were determined using agarose gel electrophoresis of the liver protein following the method of Tulchin and Taylor [ 27 ], and Ggc alleles compared to SZ-sensitivity. For this analysis, backcross mice were either designated SZ sensitive (C5 7BL/6Jlike or "B") if plasma glucose following SZ was > 2 SD above the mean of the F1 response, or designated SZ resistant (Fl-like or "F") if plasma glucose following SZ was < 2 SD below the mean of the response in C5 7BL/6J mice (see Table 2). Recombination between the Ggc locus and the locus for SZ sensitivity occurred in 9 of 21 backcross progeny examined for the Ggc enzymic forms indicating lack of tight linkage of the two loci. Therefore SZ-sensitivity could not be mapped in the BxH RI strains by strain distribution pattern with known markers. To evaluate possible involvement of the H-2 locus in susceptibility to SZ-induced diabetes, the H-2 genotypes of the parental mouse strains were initially determined by identifying RFLPs in parental DNA. D N A was digested with eight different restriction endonucleases, and RFLPs which distinguished the parental DNAs were observed with an A a genomic probe in Hind III, Pvu II or Taq I digested DNA, and with an E~2 genomic probe in Apa I digested DNA. The H-2 genotype was then defined in each of 15 individual backcross progeny. Results of an analysis of Taq I RFLPs are shown in Table 2 and Figure 3. All backcross mice were either homozygous for the C5 7BL/6J 5.7 kilobase (kb) fragment, or heterozygous for the FI 7.5 kb and 5.7 kb fragments. Recombination between the H-2 locus and SZ sensitivity was present in 8 of 15 backcrossed progeny evaluated, indicating random association of these loci and lack of tight linkage. Discussion The differences in senskivity to SZ-induced diabetes in C57BL/6J and C 3 H / H e J mice noted in this study were in general consistent with those of a previous report [ 3 ]. The plasma glucose concentration one week after the administration of a single dose of SZ in F1 hybrids relative to that in C57BL/6J provided the clearly distinguished parental phenotypes required for the genetic analysis in the current study. Plasma glucose concentrations of backcross (F1 X C57BL/6J) mice appeared to segregate into two populations, half like the C57BL/6J parent, and half like the F1 parent, and it was concluded that in these two strains SZ susceptibility is predominantly determined by different alleles at a single major locus. This hypothesis was also supported by the observation that the response to SZ treatment in the BxH RI strains could b e divided into two classes (Table 4). In addition, the results in the F1 hybrids suggest that the SZ susceptible allele is recessive, although X-chromosome linked inheritance cannot be', excluded, because only male F1 mice were studied. An X-linked factor could be evaluated in crosses of C57BL/6J females and C 3 H / H e J males by defining the S2'-sensitivity of these F1 hybrids. Studies have noted variable susceptibility to SZ-induced diabetes in other inbred strains [ 1-4, 18-20 ], and our findings in C57BL/6J and C 3 H / H e J mice do not exclude the possibility that different loci may be involved in other inbred strains, or that multiple alleles for SZ sensitivity may exist at the same locus. The data on SZ sensitivity in the RI strains provides support for a single major gene, yet the data is not compelling. While the response to SZ in one strain (Table 4, BxH-2) clearly resembled that in the C3 H / H e J parental strain, it is not so clear that the response in all other BxH strains was not different from that in the C57BL/6 parental strain. The variation in responses to SZ in the BxH strains was such that if larger numbers had been evaluated, significant differences from the C57BL/6J parents may have been noted. This would suggest additional minor loci which influence SZ susceptibility. An important test of the single gene hypothesis would be the analysis of SZ-sensitivity in F1 hybrids of BxH-2 and one or more of t]he 11 SZ-sensitive BxH-RI strains, and in backcrosses of F1 hybrids to the sensitive parental LBxH-RI strains. This genetic analysis has not been undertaken, however, and cannot be done in the immediate future due to the recent lack of availability of these mice from the Jackson Laboratories. Only two loci have been tested for linkage to the locus for SZ sensitivity. Using the strain distribution pattern for SZ-sensitive and resistant alleles in the set of BxH RI strains, the locus was estimated to be linked to the GgC locus on chromosome 6. However, direct observation of GgC alleles in the backcross progeny showed a random association between SZ sensitivity and the GgC locus. Silver and Bucker noted that the usual statistical analysis for gene mapping using RI strains may lead to incorrect conclusions about linkage unless unusually stringent criteria are adopted for rejecting the null hypothesis [ 28 ]. Only 12 BxH strains are available, and the present observations clearly demonstrate that gene mapping with a small number of RI strains may lead to an erroneous conclusion, The second locus tested for linkage was the H-2 locus. The H-2 genotypes in the backcross mice were identified using RFLPs at this locus (Figure 3), and the results showed lack of tight linkage. Interestingly, Leiter has recently shown lack of linkage o f susceptibility to multiple low-dose injections of SZ with the H-2 locus in C5 7BL/KsJ and C5 7BL/6J mice [ 22 ]. The metabolic events involved in SZ-induced pancreatic B-cell necrosis and diabetes are not well understood, and one can only speculate as to the nature of the genetic locus defining SZ sensitivity in C 3 H / H e J and C5 7BL/6J mice. Kawada et al. recently reported that a non-metabolizable SZ analogue was a potent as SZ in its diabetogenic effect [ 29 ], suggesting that an important event in SZ diabetes is interaction of the drug with a specific glucose recognition site on pancreatic B cells. In previous studies, a marked difference in glucose-stimulated insulin release was noted to account for at least part of the difference in glucose tolerance between C57BL/6J and C 3 H / H e J mice [24],which suggested that the difference in SZ sensitivity might reflect a difference in pancreatic B-cell glucose recognition. Yet in the current study, there was no correlation between glucose tolerance and SZ sensitivity in the backcross mice (Table 2). Two candidate genetic loci for SZ-susceptibility in mice may be relevant to the genetic susceptibility to diabetes in man. If SZ exerts its cytotoxic action on pancreatic B cells by lowering intracellular levels of N A D through SZ-induced activation of poly (ADP-ribose) synthetase [ 30-33 ], then the SZ-sensitivity locus may code for different alleles for this enzyme. On the other hand, allelic differences in metabolism of nitrosurea compounds may be linked to cytochrome p450 genes in mice [ 34-36 ]. Genetic analysis with cloned gene probes for these loci will define the relationship of these loci to SZ-induced diabetes in mice. If chronic exposure to nitrosurea compounds contributes to diabetes in man, then identification of the major locus responsible for differences in sensitivity in mice may provide a means to assess a putative homologous locus in man. Acknowledgements.We thank Ms. C. Welling for her technical assistance and Ms. J.Wokurka for help in preparation of the manuscript. The mouse H2 genomic DNA probes and strain distribution pattern of the BxH RI lines were kindly provided by Dr. E. Leiter of the Jackson Laboratories. This work was supported by Juvenile Diabetes Foundation Grant 386311 and by NIH Grant AM16746. 1. Junod A , Lambert AE , Orci L , Piltet R , Gonet AE , Renold AE ( 1967 ) Studies of the diabetogenic action of streptozotocin . Proc Soc Exp Biol Med 129 : 201 - 205 2. Like AA , Rossini AA ( 1976 ) Streptozotocin-induced pancreatic insulitis; new model of diabetes mellitus . Science 193 : 415 - 417 3. Rossini AA , Appel MC , Williams RM , Like AA ( 1977 ) Genetic influence of the streptozotocin-induced insnlitisand hyperglycemia . Diabetes 26 : 916 - 920 4. Like AA , Appel MC , Williams RM , Rossini AA ( 1978 ) Streptozotocin-induced pancreatic insulitis in mice. Morphologic and physiologic studies . Lab Invest 38 : 470 - 4186 5. Lewis C , Barbiers AR ( 1959 ) Streptozotocin, a new antibiotic. In vitro and in vivo evaluation . AntibiotAnn pp 247 - 254 6. Varva JJ , DeBoer C , Dietz A , Hanka I.J , Sokolski WJ ( 1959 ) Streptozotocin, a new antibacterial antibiotic . AaatibiotAnn pp 230 - 235 7. White FR ( 1963 ) Streptozotocin . Cancer Chemother Rep 30 : 49 - 53 8. Rakieten N , Rakieten ML , Nadkarni MV ( 1963 ) Studies of the diabetogenic action of streptozotocin (NSC-37917) . Cancer Chemother Rep 29 : 91 - 98 9. Lown JW , McLaughlin LW , Chang Y-M ( 1978 ) Mechanism of action of 2-haloethyl nitrosoureas on DNA and its relation to their antileukemic properties . Bioorg Chem 7 : 97 - 110 10. Dulin WE , Wyse BM ( 1969 ) Studies of the ability of compounds to block the diabetogenic activity of streptozotocin . Diabetes 18 : 459 - 466 11. Shein PS , Cooney DA , Vernon ML ( 1967 ) The use of nicotinamide to modify the toxicity of streptozotocin diabetes without loss of antitumoral activity . Cancer Res 27 : 2324 - 2332 12. Anderson T , Shein PS , McMenamin MG ( 1974 ) Streptozotocin diabetes. Correlation with extent of depression of pancreatic islet nicotinamideadenine dinucleotide . J Clin Invest 54 : 672 - 677 13. Leiter EH ( 1985 ) Differential susceptibility of BALB/c sublines to diabetes induction by multi-dose streptozotocin treatment . Curr Top Microbiol Immunol 122 : 78 - 85 14. Rossini AA , Williams RM , Appel MC , Like AA ( 1978 ) Sex differences in the multiple-dose streptozotocin model of diabetes . Endocrinology 103 : 1518 - 1520 15. Maclaren NK , Neufeld M , McLaughlin JV , Taylor G ( 1980 ) Androgen sensitization of streptozotocin-induced diabetes in mice . Diabetes 29 : 710 - 716 16. Paik S , Michelis MA , Kim YT , Shin S ( 1982 ) Induction of insulindependent diabetes by streptozotocin: inhibitionby estrogens and potentiation by androgens . Diabetes 31 : 724 - 729 17. Kiesel U , Kolb M ( 1982 ) Low-dose streptozotocin-induced autoimmune diabetes is under the genetic control of the major histocompatibility complex in mice . Diabetologia 23 : 69 - 71 18. Kiesel U , Falkenberg FW , Kolb H ( 1983 ) Genetic control of lowdose streptozotocin-induced autoimmune diabetes in mice . J Immunol 130 : 1719 - 1722 19. WolfJ, Lilly F , Shin S ( 1984 ) The influenceof genetic background on the susceptibility of inbred mice to streptozotocin-induced diabetes . Diabetes 33 : 567 - 571 20. Le PH , Leiter EH , Leyendecker JR ( 1985 ) Genetic control of susceptibility to streptozotocin diabetes in inbred mice: effect of testosterone and H-2 haplotype . Endocrinology 116 : 2450 - 2455 21. Kromann H , Christy M , Egeberg J , Lernmark A , Nerup J ( 1982 ) Absence of H-2 genetic influence on streptozotocin-induced diabetes in mice . Diabetologia 23 : 114 - 118 22. Leiter EH , Le PH , ColemanDL ( 1987 ) Susceptibility to db gene and streptozotocin-induced diabetes in C57/BL mice: control by gender-associated, MHC-unlinkedtraits . Immunogenetics26 : 6 - 13 23. Bailey DW ( 1981 ) Recombinant inbred strains and bilineal congenic strains . In: Foster HL , Small JD , Fox JF (eds) The mouse in biomedical research , VolI. Academic Press, New York, pp 223 - 239 24. Kaku K , Fiedorek FT Jr, Province M , Permutt MA ( 1988 ) Genetic analysis of glucose tolerance in inbred strains of mice: evidence for polygenic control . Diabetes 37 : 707 - 713 25. Passmore HC , Kobori JA , Zimmerer EJ , Spinella DG , Hood L ( 1987 ) Molecular characterization within the major histocompatibility complex of the mouse: mapping of crossover sites within the I region . Biochem Genetics 25 : 513 - 523 26. Morton NE , Rao DC , Lalouel JM ( 1983 ) Methods in genetic epidemiology . Karger, Basle New York Tokyo Sydney, pp 21 - 23 27. Tulchin N , Taylor BA ( 1981 ) r-glutamyl cyclotransferase: a new genetic polymorphism in the mouse (mus musculus) linked to LYT-2 . Genetics 99 : 10 % 116 28. Silver J , Buckler CE ( 1986 ) Statistical considerations for linkage analysis using recombinant inbred strains and backcrosses . Proc Natl Acad Sci USA 83 : 1423 - 1427 29. Kawada J , Toide K , Nishida M , Yoshimura Y , Tsujihara K ( 1986 ) New diabetogenic streptozotocin analogue, 3-0-methyl-2- {[(methyl-nitrosamino) carbonyl] amide}-D-glucopyranose . Evidence for a glucose recognition site on pancreatic B-cells . Diabetes 35 : 74 - 77 30. Cooperstein SJ , Watkins D ( 1981 ) Action of toxic drugs on islet cells . In: Cooperstein SJ , Watkins D (eds) The islets of Langerhans . Academic Press, New York, pp 387 - 425 31. Okamoto H ( 1981 ) Regulation of proinsulin synthesis in pancreatic islets and a new aspect to insulin-dependent diabetes . Mol Cell Biochem 37 : 43 - 61 32. Yamamoto H , Uchigata Y , Okamoto H ( 1981 ) Streptozotocin and alloxan induce DNA strand breaks and poly(ADP-ribose) synthetase in pancreatic islets . Nature 294 : 284 -- 286 33. Yamamoto H , Uchigata Y , Okamoto H ( 1981 ) DNA strand breaks in pancreatic islets by in vivo administration of alloxan or streptozotocin . Biochem Biophys Res Comm 103 : 1014 - 1020 34. Rouer E , LeProvost E , Leroux J-P ( 1983 ) ',Studyof benzphetamineN-demethylase in streptozotocin-diabetic mice and rats: evidence of the induction of catalytically and immunologically specific forms of cytochrome p 450 . Biochim 65 : 679 - 683 35. Peng R , Tennant P , Lorr NA , Yang CS ( 1983 ) Alterations of microsomal monooxyganase system and[ carcinogen metabolism by streptozotocin-induced diabetes in rats . Carcinogenesis 4 : 703 - 708 36. Favreau LV , Schenkman JB ( 1988 ) Composition changes in hepatic microsomal cytochrome p-450 during onset of streptozotocin-induced diabetes and during insulin treatment . Diabetes 37 : 577 - 584 Received: 29 December 1988 and in final revised form: 17 July 1989


This is a preview of a remote PDF: https://link.springer.com/content/pdf/10.1007%2FBF00274530.pdf

K. Kaku, J. McGill, M. Province, M. A. Permutt. A single major gene controls most of the difference in susceptibility to streptozotocin-induced diabetes between C57BL/6J and C3H/HeJ mice, Diabetologia, 1989, 716-723, DOI: 10.1007/BF00274530