PASCANI, MOLDOVA
Maize and Sorghum Research Institute

A modifier of the Bg element

— Koterniak, VV

A genetic factor contained in inbred 346 enhances the ability of the Bg-lf element to cause reversion of the o2-lf allele. Genetic instability of the o2-m(r) alleles (the alleles which arise as a result of insertion of the non-autonomous (receptor) rbg element in the opaque2 locus, characterized by opaque endosperm phenotype) is conditioned by their ability to revert to the wild-type O2 allele due to excision of rbg in the presence of the autonomous (regulatory) Bg element. If such reversion occurs before the first mitotic division of the primary endosperm nucleus, it leads to phenotypically normal kernels (whole endosperm revertants, WER). Earlier, under disruption selection for WER, closely related lines (designated as LFWER and HFWER of homozygous o2-lf; Bg-lf and o2-hf; Bg-hf genotypes, respectively) differing in their mutable o2-m(r) alleles and autonomous Bg elements (Maydica 44: 195–203, 1999; MNL 73: 76–79, 1999) were obtained. Further, specificity of interaction between these mutable alleles and autonomous Bg elements was determined. Thus, in the presence of the Bg-hf element, both the o2-lf and o2-hf alleles are characterized by high frequency of reversion; whereas Bg-lf determines high reversion frequency of the o2-hf allele but low reversion frequency of the o2-lf allele (Maydica 48: 275–281, 2003).

Significantly lower reversion frequency of the o2-lf allele in comparison with o2-hf was observed in crosses of LFWER and HFWER strains with different inbreds lacking an active Bg element and carrying different o2 alleles (both dominant and recessive) unrelated to o2-lf and o2-hf alleles’ origin (Maydica 44: 195–203, 1999; MNL 73: 76–79, 1999 and unpublished data). Thus, the frequency of WER formation in F2 kernels obtained in the crosses of the LFWER strain with the inbred 502 O2; +Bg (carrying wild-type O2 allele and not containing regulatory Bg element) was approximately 4.5 times lower than the frequency of revertants observed in the crosses of 502 O2; +Bg with the HFWER strain (MNL 73: 76–79, 1999).

However, in the crosses with the inbred line 346 O2; +Bg (carrying a wild type O2 allele and lacking an active Bg element), the reversion frequency of the o2-lf allele in the presence of Bg-lf was significantly enhanced (Table 1). (Inbreds 346 and 502 are the parental forms of the hybrid Pioneer 3978 and were bought in 1977 by the Ministry of Agriculture of the USSR from Pioneer Hi-Bred International Inc.)

Table 1. Segregation of F2 kernels in families of selfed progenies of 4 ears obtained in the crosses of LFWER and HFWER strains (homozygous o2-lf; Bg-lf and o2-hf; Bg-hf genotypes, respectively) with the inbred 346 O2, +Bg.

* Only ears with the ratio of the sum of phenotypically wild-type and variegated kernels to opaque kernels not significantly different from 15:1 (according to the χ2 test) were used for the analysis. This was the case for all obtained ears except 1 out of 16 and 2 out of 13 for A2 and B2 families, respectively.

† Here and in Table 2, n, v, o: phenotypically wild-type (normal), variegated and opaque kernels, respectively.

‡ Reversion frequency (RF) was calculated by the formula RF=100·(n-12/15·(t-o))/(9/15·(t-o)), where t is the total number of kernels, n and o are numbers of phenotypically wild-type and opaque kernels, respectively.

§ A common letter following mean values indicates insignificance of differences between them (P≤0.05).

Data obtained indicate that the reversion frequency of the o2-lf allele in the presence of Bg-lf in the F2 of the selfed progenies of the o2-lf; Bg-lf × 346 O2; +Bg genotype (Table 1A, families A1 and A2) is significantly enhanced (in comparison, for example, with selfed LFWER strains, see Table 2E) and is approximately at the same level with reversion frequency of the o2-hf allele in combination with Bg-hf observed in the progeny of one of the ears of the selfed o2-hf; Bg-hf × 346 O2; +Bg genotype (Table 1B, family B1).

The original 346 inbred does not contain an active regulatory Bg element that is indicated by kernel segregation ratios on studied ears (Table 1 and unpublished data); therefore, the observed enhancement of the reversion frequency of o2-lf cannot be connected with the presence of another (different from Bg-lf) autonomous element. Hence, this enhancement may indicate the presence of a modifier factor (or factors) in inbred 346. In the cross with the o2-lf; Bg-lf genotype, the 346 O2; +Bg line (Table 1) was used as male parent; therefore, the modifier factor is not connected with DNA-containing elements of the cytoplasm. In the studied genotypes (a part of which is presented in Table 2), high WER content was also observed in subsequent F3 and F4 progenies of selfing (see Table 2A and D), in which ear selection was carried out without preliminary testing for WER content. This indicates that the modifier factor is not a complex of several unlinked genes but is a monogenic modifier, the action of which can enhance the ability of Bg-lf to cause reversion of the o2-lf allele. This modifier will be further referred to as Mbg (Modifier of Bg).

As observed previously (Genetika (Moscow) 39: 709–712; 769–774, 2003), the studied genotypes showed a significantly higher variability of reversion frequency of the o2-hf allele in comparison with the o2-lf allele. Thus, in the progeny of one of the selfed ears of the (o2-hf; Bg-hf × 346 O2; +Bg) genotype (family B2 in Table 1B), the frequency of reversion of the o2-hf was higher than both the reversion frequency of o2-lf in selfed progenies of o2-lf; Bg-lf × 346 O2; +Bg ears (Table 1A, families A1 and A2) and the reversion frequency of o2-hf in the progeny of another selfed ear of the o2-hf; Bg-hf × 346 O2; +Bg genotype (Table 1B, family B1). The cause of higher variability of reversion frequency of the o2-hf allele can be connected with its high rate of change in state, leading to appearance of new, derived alleles differing from the initial allele not only by their reversion frequency, but also by their ability to change reversion frequency in the presence of Mbg.

Modifier action of Mbg needs joint presence of this modifier with the Bg-lf element in the same genotype before meiosis. This conclusion follows from the analysis of the crosses of the plants obtained in the F3 progenies of selfed ear 01-4788-2 (obtained in progeny of the family A1, see Table 1A). Notwithstanding that this ear contained the o2-lf allele in combination with the Bg-lf element, it was characterized by high frequency of reversion of the o2-lf allele (Table 2A), which indicates the presence of Mbg in its genotype. High frequency of reversion of the o2-lf allele was also observed in all studied selfed ears obtained from variegated kernels of this ear (Table 2D), which also showed the presence of Mbg in their genotypes. This indicates that the Mbg modifier also should be present in the plants obtained from opaque kernels of the same 01-4788-2 ear. However, in the crosses of the plants originating from opaque kernels (used as male parent) with the LFWER line (of homozygous o2-lf; Bg-lf genotypes), the obtained ears contained only variegated kernels and did not contain WER kernels (Table 2B). In contrast, in analogous crosses of plants originated from variegated kernels (Table 2D); WER content was 7–8 times higher than the same parameter observed on selfed ears of the LFWER tester (Table 2E). In the first type of these crosses (Table 2B), in contrast to the second type (Table 2C), the Mbg modifier and the Bg-lf element were present together only after fertilization. The indicated types of crosses contained somewhat different doses of Bg-lf: 2 doses in the first type (Table 2B) or 3 (Table 2C, the cross with the plant 02-4625p61) and 2 or 3 doses (Table 2C, the cross with the plant 02-4626p61), heterozygous for Bg-lf. However, differences only in number of doses of the autonomous element could not influence significantly WER content: 2 or 3 Bg-lf doses determine low reversion of o2-lf (Genetika (Moscow) 39: 769–774, 2003).

Table 2. Kernel segregation on the selfed 01-4788-2 ear (containing F3 kernels of the cross o2-lf; Bg-lf × 346 O2; +Bg, obtained in the family A1, see Table 1A), on the progenies of variegated and opaque kernels of this ear obtained after selfing and crossing with homozygous o2-lf; Bg-lf tester, and on selfed ears of the tester.

*Reversion frequency was calculated as a percentage of the ratio of phenotypically normal (WER) kernels to the sum of normal and variegated kernels per one dose of the o2-lf allele (100·n/(3·(n+v))).

†All selfed ears obtained from opaque kernels of the ear 01-4788-2 contained only opaque kernels.

The need for common presence in one genotype for Mbg and Bg-lf for expression of modifier action of Mbg resembles the necessity of similar association for the o2-hf allele with Bg-lf or Bg-hf elements for high reversion frequency of o2-lf (Genetika (Moscow) 39: 769–774, 2003; Maydica 48: 275–281; 2003). One explanation for this phenomenon could be an epigenetic modification of the rbg element in the absence of an active Bg element (e.g. higher degree of its methylation) (Genetika (Moscow) 39: 769–774, 2003; Maydica 48: 275–281; 2003). By analogy, it is possible to assume the existence of the same phenomenon on epigenetic modification for the Mbg modifier in absence of the Bg element.

Dosage effects of Mbg on reversion frequency of o2-lf in the presence of Bg-lf. High WER content in the studied progenies of inbred 346 indicates that they contain the Mbg modifier. From these crosses it is not clear, however, whether this modifier is present in a heterozygous or homozygous state. Nevertheless, the data in Table 2 allow some conclusions about the dosage effect of Mbg on reversion frequency of o2-lf in presence of Bg-lf.

Thus, the number of doses of Mbg on selfed ears of the plants 02-4625p61 and 02-4626p61 (Table 2D) should be 3 times higher than the number of doses of this modifier contained in the crosses of these plants used as the male parent (Table 2C) with the LFWER tester (homozygous o2-lf; Bg-lf tester lacking Mbg). Comparing WER content on selfed and crossed ears of the indicated plants, it is possible to conclude that a 3-fold increase of Mbg leads to a 2- to 2.3-fold increase in reversion frequency of the o2-lf allele (Table 2C and D). Close values of increase could be noted here, notwithstanding that one plant (02-4625p61) was homozygous for Bg-lf while the other was heterozygous for this element. Significant variability in WER content (42.92–70.63%) in heterozygous Bg-lf plants also could be noted (see Table 2D).

The nature of the Mbg modifier and possible mechanism of its action. According to one hypothesis, transposon modifiers are defective derivatives of regulatory elements and the interaction between modifiers and autonomous elements can be explained by genetic complementation (Fedoroff NV, in “Mobile Genetic Elements” pp. 1–63; 1983; Seo B-S and Peterson PA, Maydica. 42: 379–384, 1997). Assuming that Mbg is a defective Bg element, the increase in reversion frequency of o2-lf can be explained by genetic complementation between Mbg and Bg-lf. Such complementation suggests an interaction between the Bg-lf encoded transposase with the product of the Mbg modifier and/or their common participation in forming a transposition complex which determines rbg excision.

Earlier, based on the behavior of o2-lf and o2-hf alleles in the presence of Bg-lf and Bg-hf elements, a conclusion was made about participation of the rbg product together with the Bgs encoded transposases in the transposition complex responsible for rbg excision (Maydica 48: 275–281, 2003). This conclusion had found further confirmation in the study of the behavior of the same o2-m(r) alleles with the Bg-Ref element (unpublished data: Bg-Ref element is a regulatory element independent of Bg-hf and Bg-lf origin; this line was a kind gift from Istituto Sperimentale per la Cerealicoltura (Bergamo, Italy)). In favor of the participation of the products of defective autonomous elements (i.e. modifiers and non-autonomous elements) in the process of excision of rbg is the fact that nonautonomous elements often are defective derivatives of the autonomous ones (see, for example, Fedoroff NV, 1983, in “Mobile Genetic Elements” pp. 1–63), and therefore should have a high degree of homology with the autonomous elements. Thus, the rbg element shares more than 75% homology with the Bg element (Hartings H et al., Molecular and General Genetics 227: 91–96, 1991). In this context, it is not excluded that Mbg is an rbg element present not in the o2 locus, the modifier action of which is conditioned by the ability of its product to increase excision frequency of the rbg element contained in the o2-lf allele. In this respect, the Mbg product shows similarity with the putative product of the rbg element present in the o2-hf allele in combination with the latter of one of the Bg-hf, Bg-lf or Bg-Ref elements (Maydica 48: 275–281, 2003; unpublished data).

It was also suggested that transposon modifiers could be cellular genes (encoding host factors) which are able to participate in the transposition process of mobile elements (Seo B-S and Peterson PA, Maydica 42: 379–384, 1997). Thus, the increase of expression of a cellular cofactor (HMGB1) in wild-type mouse cells enhances transposition of the Sleeping Beauty transposon in these cells (Zayed H et al., Nucleic Acids Research 31: 2313–2322, 2003). By this hypothesis, however, it is more difficult to explain the above-demonstrated low frequency of reversion of o2-lf in the crosses, in which Mbg and Bg-lf were present together only after fertilization (see Table 2C and D). Nonetheless, the possibility that Mbg encodes a cellular factor is not excluded. Thus, if the expression of this factor is connected with certain stages of development of the gametophyte or endosperm (occurring after the first mitotic division of its primary nucleus), then it can explain the absence of WER kernels in the indicated type of cross.

Return to the MNL Volume 79 Index
Return to the index of Maize Newsletters
Return to the Maize Genome Database Page