| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
Department of Biomedical Science, Division of Experimental Pathology and Oncology, University of Sassari, I-07100 Sassari, Italy [M. R. D.M., R. M. P., M. M. S., M. R. M., D. C., F. F.], and Unit of Genetic Cancer Susceptibility, International Agency for Research on Cancer, Lyon F-69372 cedex 08, France [F. C., G. R.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The study of murine models of hepatocarcinogenesis has evidenced a very complex interplay of genetic factors. Thus far, seven susceptibility loci (Hcs1 to Hcs7) and two resistance loci (Hcr1 and Hcr2) have been identified in crosses between susceptible and resistant mice strains (7, 8, 9, 10, 11, 12) . Two more susceptibility genes (Hcf1 and Hcf2) abrogate hepatocarcinogenesis inhibition by ovarian hormones in the mouse (13) . Very little is known about inherited predisposition to HCC in rat. A putative suppressor gene (rcc+), critical for determining the sensitivity of rats to DENA-induced liver carcinogenesis, has been identified in MHC-recombinant rat strains congenic for the MHC and its linked region grc (14) . Partial deletion of WD gene, encoding a copper-transporting protein expressed almost exclusively in liver (15) , results in hepatitis in LEC rats, 60% of which survive and undergo malignant transformation (16) . On the other hand, rat liver preneoplastic and neoplastic lesions, induced in different models, exhibit various commonalties with analogous human liver lesions, with respect to cytological and biochemical changes (17) , molecular alterations [including p53 mutation (18) or deletion (19 , 20) ], c-myc rearrangement and amplification (21 , 22) , and chromosomal aberrations (19 , 20 , 23 , 24) . Thus, the evaluation of the mechanisms of genetic susceptibility to hepatocarcinogenesis in rat and a comparative evaluation of these mechanisms in different rodent species could suggest some possible mechanisms for understanding the genetics of human hepatocarcinogenesis.
We focused our efforts on the genetic control of rat liver tumors in a hybrid BFF1 rat strain, generated in our laboratory by crossing the phylogenetically distant (25) BN (B), resistant, with the F344 (F), susceptible, rat strain (26) . Resistance of BN rats to hepatocarcinogenesis is genetically transmitted as a dominant character to BFF1 hybrids (26) . As previously observed in mice (27) , this genetic trait influences growth ability and progression of initiated cells to HCC, more than the initiation stage (26) . Herein, we performed a linkage analysis in a backcross progeny generated by crossing BFF1 with F344 rat strains. We found significant and suggestive linkage of four unlinked loci, mapped on chromosomes 4, 7, 8, and 10, with the development of neoplastic lesions.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Phenotyping of Parental Strains and Backcross Rats.
Premalignant and malignant lesions were induced in 25 F344, 25 BN, 25 BFF1 rats, and 157 backcross rats by the RH protocol (28)
, which included initiation with a necrogenic dose of DENA (150 mg/kg) and feeding, after repair, of a 0.02% 2-acetylaminofluorene diet, with a partial hepatectomy at the midpoint of this feeding. Seven to twelve rats of each parental strain were killed 32 weeks after initiation, and all of the other surviving rats were killed at 70 weeks. All of the backcross rats were killed at 32 weeks. The livers were resected and rapidly cut into 2–3-mm slices to identify gross neoplastic nodules and HCCs. Small portions of these lesions, as well as of analogous lesions in liver surface, were isolated. Isolated nodules and HCCs, and small pieces of liver from each lobe were fixed in buffered formaldehyde (pH 7) and processed for embedding in paraffin, serially sectioned into 5-µm thick slices, and used for H&E staining and GST-P immunohistochemistry (29)
. Morphometric analysis of microscopic lesions was carried out using a Leica Quantimet 500 image analyzer (Leica Cambridge Ltd, Cambridge, United Kingdom). The number of lesions per cm3, the number of lesions per liver (defined as lesions/cm3 x liver weight, because the density of liver is approximately 1 g/cm3), the mean volume of lesions, and the volume fraction (calculated by the Delesse method) were determined according to Pugh et al. (30)
. Transections with radius >35 µm could be reliably identified and were included in the analysis. At least nine slices per liver were analyzed. Benign lesions were distinguished from carcinoma nodules on the basis of the published criteria (31)
.
Genotyping.
Genomic DNA from spleens of backcross rats was extracted from isolated nuclei, purified and precipitated by the Extragen kit in an automated DNA extractor (Talent, Trieste, Italy). Genotyping was performed using 190 polymorphic microsatellite markers (32
, 33) .4
The markers (Research Genetics, Huntsville, AL) were distributed throughout the entire rat genome (all of the autosomes and the X chromosome) at an average density of one marker/9.78 cM, leaving one gap of 31.8 cM on chromosome 3; 29.9 cM on chromosome 8; and 29.1 cM on chromosome X, whereas all of the other gaps were smaller than 24 cM. The PCR reactions included 50 ng of genomic DNA; 20 mM (NH4)2SO4; 75 mM Tris-HCl (pH 9); 1.5, 2, or 3 mM MgCl2; 3 pmol of each primer; dNTPs (200 µM each), and 0.2 unit of Red Hot DNA Polymerase (Advanced Biotechnologies, Epsom, United Kingdom). Cycling parameters were: 5 min at 96°C; 30 s at 94°C; 30 s at 53, 55, or 57°C; and 30 s at 72°C for 35 cycles, and final extension for 5 min at 72°C in a 9600 GeneAmp PCR system (Perkin-Elmer, Norwalk, CT). The PCR products were run on 3.5% agarose gels to distinguish homozygous from heterozygous backcrosses. When markers polymorphic for only 2–8 bases were used, PCR reaction system was supplemented with 1–4 µM fluorescently labeled dCTP (34)
. The PCR products were appropriately pooled, and aliquots were loaded on a 4.8% polyacrylamide-8 M urea gel and run in a 377 Automated Sequencer (Applied Biosystems, Foster City, CA). The data were automatically collected and analyzed by the GeneScan and Genotyper software (Applied Biosystems, Foster City, CA).
Statistical Analysis.
Linkage maps were constructed using the MAPMAKER/EXP 3.0 program (35)
. The associations of tumor susceptibility (low-responder versus high-responder, as defined on the basis of the phenotypic V and VF parameters in parental strains) with alleles of rat microsatellite markers were evaluated by LOD score, calculated from the recombination frequencies observed between phenotype and genotype. The phenotypic variables were analyzed as such or after rank transformation. Ps in the range of 10-4 were considered a statistically significant threshold for linkage (36)
. Threshold LOD score values at 1.9 and 3.3 were considered for "suggestive" and "significant" linkage, respectively (37)
. The proportion of total backcross variance for phenotypic parameters explained by the association between the market and the genotype class (R2) was taken as an index of the importance of each locus. QTL analysis was carried out using parametric and nonparametric methods with MAPMAKER/QTL 1.9 (38)
. Mean values of phenotypic parameters of backcross rats were determined for each genotype class and analyzed by ANOVA procedure of SAS (SAS Institute Inc., Cary, NC). Potential interactions between genetic loci were analyzed by an ANOVA procedure of SAS, and Ps were corrected for multiple comparisons using a formula proposed by Lander and Schork (36)
. Differences between parental interstrains and between homozygous and heterozygous backcross rats for phenotypic parameters were analyzed by ANOVA, and multiple comparisons were made by Dunnett’s test (39)
.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, the number per liver (N) of GST-P- positive lesions 32 weeks after initiation did not undergo any significant modification in BN and BFF1 rats, with respect to F344 rats, whereas V and VF were 24- and 4-fold higher, in F344 than in BN and BFF1 strains, respectively, without any significant difference between BN and BFF1 rats. Microscopic examination showed that most nodules developing in F344 rats were clear-acidophilic cell lesions, 67% of which presented atypical areas, exhibiting distortion of plate arrangement, thickened plates, and hepatocytes with mild nuclear atypia, suggesting initial malignant transformation. Rare atypical lesions (less than 4%) were found in BN and BFF1 rats. Relatively few gross nodules (
4 mm in diameter), showing atypical pattern, were seen in F344 rats at 32 weeks, but these lesions were present in all of the parental rat strains at 70 weeks. At this time, HCCs were seen in 8 (89%) of 9 F344 rats, in 1 (10%) of 10 BN rats, and in 2 (18%) of 11 BFF1 rats. Lesion multiplicity was 20.6 ± 1.13, 14.8 ± 1.87, and 9.7 ± 0.92 for atypical nodules in F344, BN, and BFF1 rats, respectively (means ± SE; Dunnett’s test: BN and BFF1 versus F344, P < 0.01). These figures were 3.5 ± 0.23, 1.3 ± 0.13, and 1.5 ± 0.18, respectively, for HCC (P < 0.01). Microscopic examination revealed the presence of three well-differentiated, four moderately differentiated, and one poorly differentiated carcinomas in F344 rats. These tumors exhibited large variations in volume (from less than 2 mm to about 2 cm in diameter) and occupied two or more lobules (not shown). Almost all of the tumors in BN and BFF1 rats were well differentiated, whereas one moderately differentiated HCC was found in BFF1 rats. These lesions were generally smaller than in F344 rats. Evaluation of phenotypic parameters (Fig. 1)
showed that almost all of the backcross rats had a number of lesions in the ranges found for F344, BN, and BFF1 rats. The distributions of V and VF exhibited large variations of susceptibility to the development of nodules.
|
|
.
|
; Fig. 2
). We found significant linkage on a telomeric region of chromosome 7, with a LOD score peak value of 3.9 (Z-score = 4.4) and 1-LOD-unit support interval of 24.7 cM between markers D7Mitl0 and D7Ratl6. Another QTL was localized on one telomeric region of chromosome 10, with LOD score 3.4 (Z-score = 3.4), and a 1-LOD-unit support interval of 12.9 cM between D10Ratl13 and D10Rat9, which represent the last known markers for rat chromosome 10. The profiles of LOD score, for chromosomes 7 and 10, for V reproduced those observed for VF but at constantly lower values (Table 2
; Fig. 2
), whereas no significant linkage was found for N. The results in Table 2
show that rats bearing a BN allele at the QTL on chromosome 7 exhibit average V and VF significantly higher than rat homozygous for F344 alleles, and rats bearing a BN allele at the QTL on chromosome 10 have significantly lower VF and V average values than homozygous FF backcross progeny. We tentatively defined these two QTL: rat Hcs1 and rat Hcr1 (for hepatocarcinogenesis sensitivity/resistance), respectively (see "Discussion").
|
, Fig. 2
) with LOD scores suggestive of association with nodule volume. Rat Hcr2 maps on chromosome 4, with a peak LOD score of 3.1 (Z-score = 3.7) and rat Hcr3 is located on chromosome 8 (LOD score = 2.9; Z-score = 3.6). Both localizations also showed suggestive linkage with VF, and weak influence on N (LOD score peaks falling just short of suggestive linkage; Fig. 2). Rats bearing homozygous F344 alleles, showed significantly higher V and VF values, and lower N values than heterozygous backcross rats (Table 2)
.
Finally, a QTL region with LOD score of 3.0 (Z-score = 3.7), suggestive of linkage with N but with very poor association with V and VF, was identified on chromosome 1 (Table 2
; Fig. 2)
. Rats bearing a BN allele at this QTL show a significantly higher average N value than homozygous FF rats. This suggests that the alleles derived from BN, at the above loci, are associated with a higher susceptibility, and this locus was tentatively named rat Hcs2. A few more chromosomal locations showed some linkage with N but with values lower than the limits of suggestive linkage: chromosome 6 (LOD score = 1.5; Z-score = 2.6) and chromosome 13 (LOD score = 1.7; Z-score = 2.7).
Interactions.
Potential interactions between loci in which F and B alleles apparently exert a phenotypic effect on VF and V were examined by ANOVA. Two-by-two ANOVA was performed with all of the markers against all of the other markers. None of the interaction terms in which markers corresponding to the LOD score peaks for VF were tested at Hcr loci were statistically significant, either for two-way interactions [(D10Rat5l x D4Rat40; P = 0.31); (D10Rat5l x D8Ratl8; P = 0.98); (D4Rat40 x D8Ratl8; P = 0.08)] or for three-way interactions: (Dl0Rat5l x D4Rat40 x D8Rat18; P = 0.25). Similar values were obtained for V (data not shown). Among the other two-by-two interactions, calculated on the basis of VF, only the two having P < 10-5 (values not corrected for genome-wide comparisons) were considered. D8Rat36 and D5Mghl show a strong reciprocal interaction (corrected P = 0.0019; see Ref. 36
), although neither of them has a significant individual effect. Homozygosity for the F344 allele at both markers is associated with the highest average value of VF (Table 3)
. The second significant interaction was detected between D5Mgh21 (which has no main effect) and D7Rat6 (corrected P = 0.0023) and D7Mitl0 (corrected P = 0.0028), two adjacent markers on chromosome 7, which are in the region of rat Hcs1 (LOD scores are 2.93 and 2.86, respectively). The average values of VF according to the genotype combinations of D5Mgh2l and D7Rat6 (the stronger interaction) are shown in Table 3
. In this case, rats heterozygous at both markers have the highest VF.
|
shows allelic distribution patterns in backcross rats for markers at chromosomes 10, 4, and 8, in which putative Hcr loci were identified. Fifteen of 17 rats inheriting F344 alleles at markers on these chromosomes were susceptible to hepatocarcinogenesis, as assessed by an evaluation of VF. Similarly, 16 of 22 rats inheriting the BN allele were resistant. Inheritance of B allele at chromosomes 4 and 8 was associated with a susceptible phenotype, whereas a small (insignificant) decrease occurred at chromosome 10. Intermediate phenotypes were seen when two B alleles were present at the three chromosomes. Similar results were obtained when V values were considered (not shown).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We identified four QTL associated with VF and/or V. The loci located on chromosomes 7 and 10, were in significant linkage with VF, and those on chromosomes 4 and 8 were in suggestive linkage with both V and VF. The analysis performed in this study efficiently detects QTL that have a relatively large effect (42) but does not allow us to distinguish whether the phenotypic behavior of backcross progeny depends on (a) the presence of a susceptible gene contributed by the susceptible strain; or (b) the presence of a resistance gene contributed by the other strain. Surprisingly, average phenotypic values in backcross rats were compatible with a dosage-positive effect of the B allele, inherited from the resistant BFF1 strain, for the locus on chromosome 7 which was, therefore, tentatively named rat Hcs1. It may be hypothesized that this locus interacts with F344 genetic factors to support active growth of nodular cells, and that it is partially suppressed in BFF1 and BN rats by some B loci. QTL identified in chromosomes 10, 4, and 8 are apparently associated with a dosage-negative effect of the B allele. However, taking into account that the B allele is dominant on the F allele, we tentatively suggest that lower values of tumor susceptibility parameters, in heterozygous backcross rats, depend on the association of these QTL with resistance to hepatocarcinogenesis. Consequently, we named QTL on chromosomes 10, 4, and 8, rat Hcr1, Hcr2, and Hcr3, respectively. An apparently small dosage-positive effect of B allele on the lesion number in heterozygous backcross rats was found for the QTL on chromosome 1. This Hcs locus (rat Hcs2) suggests the existence in BN and BFF1 parental strains of gene(s) controlling the number of initiated liver cells, with a relatively low effect that could be partially masked by other genetic factors in these strains. Additional studies are under way with (BN x F344)F2 intercross rats to confirm the present results as well as to determine the additive and dominant effects of the segregating QTL, which would allow more accurate evaluation of their effect on susceptibility/resistance to hepatocarcinogenesis.
Additional potential QTL affecting hepatocarcinogenesis were detected by the study of interactions, which brought to evidence two markers, D5Mgh1 and D8Rat36, that lack a main effect as measured by the LOD score but whose interaction affects tumor susceptibility (compare Ref. 43 ). D8Rat36 is positioned in the locus Hcr3, 26.1 cM from D8Rat18, which corresponds to LOD score peak for this QTL. Our results cannot exclude the Hcr3 involvement in the interaction with D5Mgh1. However, it should be noted that all of the other markers at this QTL, with phenotypic effect much stronger than that of D8Rat36, are apparently not involved in this interaction. A possible new QTL at D5Mgh21 was not detected by LOD score analysis but only by an analysis of interactions, presumably because its effect is counteracted by the two markers located on chromosome 7, in the region of rat Hcs1.
Various genes potentially involved in hepatocarcinogenesis map5 on rat chromosomes:
); and where rat Hcs1 and rat Hcr1–3 loci are located at 7q33-q34, 10q11-q24, 4q21-q34, and 8q24-q32 (44) , respectively. These genes are possible candidates for these loci or could be somehow influenced by them. Karyotypic aberrations of chromosome 7 and the rearrangement, amplification, and deregulation of expression of the above genes are involved in the promotion and progression stages of rodent liver carcinogenesis (19, 20, 21, 22, 23, 24 , 45, 46, 47, 48, 49) . Interestingly, genetic polymorphism of cytochromes P-450 2El and 2D6 has been associated with the development of HCC in humans (50) . Furthermore, it is noteworthy the homology of rat chromosomal regions in which rat Hcs1, Hcs2, and Hcr1 are located, with human 8q, 11p13-p15, and 17p13, respectively. Human HCC exhibits duplications at 8q (51) and loss of heterozygosity in the other two regions (24) , which suggests the involvement of these chromosomal regions in human liver carcinogenesis.
Our observations indicate that the genetic susceptibility of rats to hepatocarcinogenesis induced by the RH protocol results from the action of multigenetic factors associated with relatively early events in the evolution of the disease. It is unclear, at this stage of our research, whether the QTL identified would influence the susceptibility to induction of HCCs by other protocols. Susceptibility/resistance genes contributed by the parental rats apparently influence the progression of neoplastic nodules more than they influence the initiation stage (26 , 27) , and it is likely that they affect the evolution of initiated cells to HCC in different hepatocarcinogenesis models. Nevertheless, we cannot exclude the possibility that this effect is more evident in experimental models such as the RH model, characterized by the relatively fast growth of initiated cells (46) .
A difference from the mouse, in which one Hcs locus contributes highly (
85%) to susceptibility to liver cancer, at least in the C3H/He strain (7)
, each rat locus, taken alone, contributes relatively little (R2 between 7.1 and 11.5%) to susceptibility/resistance and seems to influence the phenotype through both epistatic and additive effects. A similar situation may exist in humans, in which epidemiological evidence indicates the existence of rare familiar clusters of HCC, even in high-risk areas (52)
. Future analysis of subphenotypes linked to QTL and the cloning of susceptibility/resistance genes in the rat will cast new light on the pathogenesis of the disease. Because of the commonalties of various biological and molecular mechanisms involved in human and rat hepatocarcinogenesis (17, 18, 19, 20, 21, 22, 23, 24)
, our understanding of the role of susceptibility/resistance genes in the experimental carcinogenesis model should make possible the identification of critical events in the pathogenesis of human HCC.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by grants from Associazione Italiana Ricerca sul Cancro, Ministero dell’Università e Ricerca Scientifica e Tecnologica, and Assessorato Igiene e Sanità RAS. M. R. D. M. is a recipient of a fellowship "Liliana Momigliano Sacerdote-Martha Galle" of the Fondazione Italiana per la Ricerca sul Cancro (FIRC). 
2 To whom requests for reprints should be addressed, at Department of Biomedical Science, Division of Experimental Pathology and Oncology, University of Sassari, Via P. Manzella, 4, I-07100 Sassari, Italy. Phone: 39–079-228307; Fax: 39–079-228305; E-mail: feo{at}ssmain.uniss.it
3 The abbreviations used are: HCC, hepatocellular carcinoma; BN, Brown Norway; DENA, diethylnitrosamine; GST-P, glutathione S-transferase (placental); LOD, logarithm of the odds; QTL, quantitative trait locus/loci; N, nodule number/liver; RH, resistant hepatocyte, V, nodule mean volume, VF nodule volume fraction.
4 Database <http://www.genome.wi.mit.edu/rat/public/>.
5 Database http://ratmap.gen.gu.se.
Received 12/ 7/98. Accepted 7/20/99.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Feo, M. Frau, M. L. Tomasi, S. Brozzetti, and R. M. Pascale Genetic and Epigenetic Control of Molecular Alterations in Hepatocellular Carcinoma Experimental Biology and Medicine, July 1, 2009; 234(7): 726 - 736. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. Dragani 10 Years of Mouse Cancer Modifier Loci: Human Relevance Cancer Res., June 15, 2003; 63(12): 3011 - 3018. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F.S. Falvella, R. M. Pascale, M. Gariboldi, G. Manenti, M. R. De Miglio, M. M. Simile, T. A. Dragani, and F. Feo Stearoyl-CoA desaturase 1 (Scd1) gene overexpression is associated with genetic predisposition to hepatocarcinogenesis in mice and rats Carcinogenesis, November 1, 2002; 23(11): 1933 - 1936. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. De Miglio, R. M. Pascale, M. M. Simile, M. R. Muroni, D. F. Calvisi, P. Virdis, G. M. Bosinco, M. Frau, M. A. Seddaiu, S. Ladu, et al. Chromosome Mapping of Multiple Loci Affecting the Genetic Predisposition to Rat Liver Carcinogenesis Cancer Res., August 1, 2002; 62(15): 4459 - 4463. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Wood, J. E. Korkola, and M. C. Archer Tissue-specific resistance to cancer development in the rat: phenotypes of tumor-modifier genes Carcinogenesis, January 1, 2002; 23(1): 1 - 9. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Yan, Z.-Z. Zeng, S. Higashi, A. Denda, Y. Konishi, S. Onishi, H. Ueno, K. Higashi, and H. Hiai Resistance of DRH strain rats to chemical carcinogenesis of liver: genetic analysis of later progression stage Carcinogenesis, January 1, 2002; 23(1): 189 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Wood, D. S.R. Sarma, and M. C. Archer Inheritance of Resistance to Promotion of Preneoplastic Liver Lesions in Copenhagen Rats Experimental Biology and Medicine, October 1, 2001; 226(9): 831 - 835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z.-Z. Zeng, S. Higashi, W. Kitayama, A. Denda, Y. Yan, K. Matsuo, Y. Konishi, H. Hiai, and K. Higashi Genetic Resistance to Chemical Carcinogen-induced Preneoplastic Hepatic Lesions in DRH Strain Rats Cancer Res., June 1, 2000; 60(11): 2876 - 2881. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, the inheritance by backcross progeny of F allele from F344 rats; 
, the inheritance of B allele from BN. Mean ± SE of VF of each genotype are included in the figure. ANOVA analysis showed significant differences among groups (P < 0.0001). Dunnett’s test: FFF, BFF, FFB, and FBF versus BBB, P < 0.05. BFB, BBF, and FBB versus BBB, not significant.