Maize Genetics Cooperation Newsletter 80. 2006.

PIRACICABA, SP, BRAZIL

ESALQ – Universidade de São Paulo

PONTA GROSSA, PR, BRAZIL

Universidade Estadual de Ponta Grossa

A seed-by-seed strategy to study the paramutation at r1 locus

--Mondin, M; Gardingo, JR

The paramutation at the r1 locus was largely studied by classical and molecular approaches, and several aspects of its behavior and origin were elucidated. Classical experiments provided information about the locus structure, genetic distance and phenotypical instability, and the molecular genetics revealed the sequence and the elements that comprise the locus, including transposable elements, genes and methylation. Many advances have been made towards the understanding of the control of the paramutation at the r1 locus, however many questions remain unanswered. One question that has interested us is related to the instability of the phenotypes after the paramutagenic allele altered the paramutable one in the heterozygous state. Seven classes of kernel pigmentation are known, varying from colorful with a maximum deposition of anthocyanin to colorless without pigmentation, in the F2 progeny. The literature is categorical in descriptions about reversions to the original state of the allele after several cycles of self-pollination, it being well established that the paramutation is an unstable event. The knowledge about stable paramutant alleles is incipient and the selection of a stable phenotype of each class of pigmentation could represent material important for molecular investigation. In this work, the main aim is the development of a strategy that permits us to understand the instability of the alleles and the selection of possible stable alleles to produce inbred lines that could be used in the molecular investigation.

We had previously described the introgression of a paramutable r allele in traditional varieties of maize from Brazil (Gardingo and Mondin, MNL 77:60-61, 2003). Inbred lines have been derived from the varieties that express the paramutation. The phenotypical classes have been scored from 1 (colorless) to 7 (colorful) and every classes was observed in the S1. To obtain the S2, a bulk of the classes 3, 4 and 5 was selected. In the S3 generation each class was followed seed-by-seed and evaluated as to the seven seed color pigmentation classes. Here, we present some results in one inbred line derived from the Carioca variety (Ca).

The S2 generation was scored considering the ear as a whole. From all patterns of segregation expected, only six were observed (Table 1). In one case, a pattern did not present colorless seeds, and the seeds in the ears segregated from class 2 to class 6. A high number of colorless ears were recovered, which could be a
Table 1. S2 segregation pattern, derived from a bulk of seed of pigmentation color classes 3, 4 and 5.

	Segregation Mode*
Inbred Line	Colorless	Colorless/2-4	Colorless/2-5	Colorless/6	2-6	All Classes (1 to 6)
Ca	81	47	71	13	5	19

* Number of ears scored.

result of the homozygous recessives. A complete imprint was not discarded, but new experiments should be conducted to consider this hypothesis. No colorful seed was recovered in any ear. Class 6 was scored in a low frequency, and the seeds were self-fertilized to observe the pattern of segregation in the S3 generation. Even in the ears segregating to different classes, the highly pigmented seeds were recovered in a low frequency. To exemplify this, the ears segregating colorless/2-5 were scored seed-by-seed. Table 2 presents the absolute numbers, and Figure 1 shows the average of each class from the ears. It is clear that the frequency of the class 5 was significantly lower than the other classes. We have considered this case as a pattern of segregation, since several ears have shown similar frequencies. We expected a higher frequency of reverting seeds, expressing color classes 6 and 7. We have postulated that the effect of the paramutable allele is very strong, and several generations of self-fertilization were needed to recover the colorful class.

Table 2. Total of seeds scored on the colorless/2-5* ears.

	Seed Color Classes
Inbred line	Colorless	2-4	5	Total
Ca	7221	7230	674	15125

*Presented in the Table 1

Figure 1. Average frequency of seeds from different classes of pigmentation scored in colorless/2-5 ears.

In the S3, colorful seeds were not recovered, however classes 5 and 6 were more frequent. The number of non-segregating ears was higher (Table 3). Seed class 1 resulted only in colorless ears. This case has been interpreted to be recessive homozygous, but this class has been analyzed carefully to identify possible reversions to color classes. Every seed class scored segregated colorless. We were expecting a higher frequency of classes 5, 6 and 7, but as they were not observed, the postulate described above seems to be true.

Analyzing seed-by-seed the non-segregating S3 ears derived from class 3 seed, a lower frequency of seeds in class 5 and a
Table 3. S3 segregation pattern derived from selected S2 seed color classes.

Seed Color Classes	Ears Scored	Segregating to colorless	Non-segregating	Colorless
1	40	0	0	40
2	16	14	1	1
3	50	39	7	4
4	30	24	5	1
5	11	7	4	0
6	4	4	0	0

higher frequency in class 4 was observed, while ears derived from class 5 seed presented a lower frequency of seed in the less pigmented classes (Table 4). Class 4 ears showed segregation for classes 2 to 6. Some ears showed a unique class of seed color, mainly when derived from class 5, and these ears have been selected for evaluation. The most important observation was the absence of colorless seeds. The classes observed in a non-segregating ear should be analyzed seed-by-seed, trying to minimize the segregation to different color classes. These results indicate that the selection of some seed color classes in non-segregating ears, followed by seed-by-seed analysis in the next generation, could be a good strategy to stabilize the paramutation. Some crosses between plants of the same class have generated seeds of the same class (data not shown), for example, crosses of class 4 produced ears fully class 4.

Table 4. Frequency of seed color classes on S3 non-segregating ears.

S2 seed color classes	Seed Color Classes
	2	3	4	5	6
3	30	322	264	23	0
4	15	121	497	305	10
5	0	25	75	305	88

We believe that as a stabilized paramutation is obtained, some important aspects of the r1 paramutation will be explained, such as the role of transposable elements in paramutation events. Moreover, the stabilization of a phenotype class could be a result of a chromatin conformation that is transmitted generation to generation without alteration, or a suppression of the recombination among the repeats of r1. All these interesting questions could be investigated utilizing these stable lines with molecular approaches.

_________________________________________________

Please note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors.

Return to MNL 80 on-line index.

Return to MNL index.

Return to MaizeGDB home page

____________________________________________