Published online 18 May 2006
Published in Crop Sci 46:1576-1580 (2006)
© 2006 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
CROP BREEDING & GENETICS
Allelic Frequency at the Rf3 and Rf4 Loci and the Genetics of A3 Cytoplasmic Fertility Restoration in Converted Sorghum Lines
Les C. Kuhlmana,
Daryl R. Pringb,
William L. Rooneya,* and
Hoang V. Tangb
a Dep. of Soil and Crop Sciences, Texas A&M Univ., 2474 TAMU, College Station, TX 77843
b USDA-ARS, Crop Genetics and Environment Research Unit, Dep. of Plant Pathology, Univ. of Florida, Gainesville, FL 32611
* Corresponding author (wlr{at}tamu.edu)
 |
ABSTRACT
|
|---|
Cytoplasmic-nuclear male sterility (CMS) in sorghum [Sorghum bicolor (L.) Moench] is integral to hybrid seed production. One such CMS, A3, is controlled by a two gene gametophytic fertility restoration system involving two restoration alleles, Rf3 and Rf4, which are required for individual male gamete viability. A survey of the allelic frequencies of Rf3 and Rf4 in the sorghum germplasm collection was undertaken using 140 lines from the Sorghum Conversion Program. These lines were hybridized to two testers, A3FL3 (Rf3Rf3rf4rf4) and A3FL4 (rf3rf3Rf4Rf4). The resulting hybrids were evaluated for I2–KI pollen stainability and seed set to determine their fertility restoration reaction. Transcript analyses were performed on lines that contained restoration alleles to determine whether they fit the expected genetic model of fertility restoration. Three lines (SC306, SC512, and SC215) were found to restore fertility with both testers and thus contain both restoration alleles. The surveyed allelic frequency of Rf3 and Rf4 was 5.1 and 4.8%, respectively. Using transcript analyses, two exceptional lines were identified (SC306 and SC215) that carried Rf3 but did not confer enhanced transcript processing activity (TPA) of mitochondrial orf107. Enhanced TPA was thought to be tightly linked to or represent gene action of Rf3, but identification of these lines demonstrates that enhanced TPA is not required for fertility restoration and does not represent gene action of Rf3.
Abbreviations: CMS, cytoplasmic-nuclear male sterility • nt, nucleotide
|
INTRODUCTION
|
|---|
COMMERCIAL PRODUCTION of hybrid sorghum would not be feasible without the CMS system (Stephens and Holland, 1954). The system is based on the interaction of both nuclear and cytoplasmically inherited genetic components, allowing the production of uniformly male sterile lines that can produce progeny in which fertility is restored through the action of nuclear restorer (Rf) genes (Kempken and Pring, 1999).
Since the characterization of the initial CMS in sorghum, there have been several other different CMS reported as well. The original CMS, designated A1, is the system used in production of the vast majority of sorghum hybrids. The A2 system was identified and developed so that hybrid seed production would be feasible if needed but it is not commonly used (Schertz and Ritchey, 1978). The A3 CMS was also characterized, but use of this system has been limited due to complexity in restoration (Worstell et al., 1984; Tang and Pring, 2003). In addition to these CMS, at least four other CMS have been characterized (Pring et al., 1995; Xu et al., 1995).
Within these CMS, two mechanisms of sterility induction have been identified based on when the sterility is manifested in relation to meiosis. Sterility induced before meiosis (sporophytic) results in pollen abortion and is restored by the nuclear genotype of the plant. Manifestation of sterility after meiosis (gametophytic) also results in pollen abortion but fertility is restored by the genotype of the individual haploid male gamete. In the A1 and A2 system, fertility is restored in a sporophytic manner by one or two dominant fertility restoration alleles (Schertz et al., 1989). In the A3 CMS, a unique two-gene gametophytic fertility restoration system that requires complementary action of two nuclear restoration genes was identified (Tang et al., 1998). In this system, both restoration alleles, designated Rf3 and Rf4, must be present in an individual male gamete for its viability to be restored. Thus, in an A3 CMS hybrid, all of the pollen cannot be viable. The frequency of viable pollen depends on the alleles present in the hybrid. A restored A3 hybrid that is heterozygous at one or both loci would produce only 50 or 25% viable pollen, respectively (Pring et al., 1999; Tang et al., 1998).
Molecular analysis revealed that the A3 CMS is characterized by a chimeric mitochondrial open reading frame, orf107, which is associated with expression of CMS (Tang et al., 1996a, 1998). The basis for loss of pollen viability in this CMS system is unknown. Restoration of fertility is associated with aberrations in transcriptional characteristics of orf107; an enhanced transcript processing activity (TPA), detected in leaves and pollen of lines restored to fertility, results in cleavage of 70% of whole length transcripts within orf107, thus precluding abundant transcripts for potential translation. Genetic analyses indicated that enhanced TPA is tightly linked to, or represents, a gene required for fertility restoration, and the restoring allele was designated Rf3 (Tang et al., 1998, Pring et al., 1999).
The usefulness of a CMS is partially dependent on the frequency of restoration alleles in elite breeding populations. Restorers for the A1 CMS are common; they are estimated to be in approximately 68% of the sorghum lines from the USDA-Texas Agricultural Experiment Station Sorghum Conversion Program (Torres-Cardona et al., 1990; Bosques-Vega et al., 1987). Fertility restorers of A3 CMS are much rarer; only three (0.7%) were found (SC426, SC835, and SC273) in sorghum lines screened by Torres-Cardona et al. (1990) and Bosques-Vega et al. (1987).
The frequencies previously reported for A3 fertility restoration were based on evaluating seed set in crosses with A3Tx398. In subsequent research, the genotype of A3Tx398 was identified as rf3rf3rf4rf4 (Tang et al., 1998). Thus, the reports by Torres-Cardona et al. (1990) and Bosques-Vega et al. (1987) determined whether the entry being tested contained both restoration alleles, Rf3 and Rf4, but did not take into account the expected lower seed set due to gametophytic restoration. Tang and Pring (2003) developed two sorghum lines, A3FL3 (Rf3Rf3rf4rf4) and A3FL4 (rf3rf3Rf4Rf4), to permit screening of germplasm for individual restoration alleles based on expected pollen staining and seed set.
The objective of this research was to use the newly available testers, A3FL3 and A3FL4, to conduct an allelic frequency screening of lines from the Sorghum Conversion Program. Lines identified as restorers (at either allele) based on pollen screening and seed set were screened using transcript analyses to discern whether they fit the currently accepted genetic model.
|
MATERIALS AND METHODS
|
|---|
Plant Materials
One hundred and forty sorghum lines were used to estimate the allelic frequency of the restoration alleles Rf3 and Rf4. They represent a random selection from the approximately 650 converted lines from the Sorghum Conversion Program (Stephens et al., 1967). Hybrids were created using two female testers, A3FL3 (Rf3Rf3rf4rf4) and A3FL4 (rf3rf3Rf4Rf4), during the summer of 2003 and/or 2004 in College Station, TX. For each SC line, two hybrids were produced. The fertility reaction of both hybrids was used to identify the allelic composition of the Rf3 and Rf4 loci in the SC lines (Table 1). Hybrids were made using IS1112C (Rf3Rf3Rf4Rf4) with both testers for use as controls.
View this table:
[in this window]
[in a new window]
|
Table 1. Expected allelic composition of the SC lines based on their phenotypic reaction for fertility when testcrossed to A3FL3 and A3FL4.
|
|
Fertility Evaluation
Hybrids were grown in a greenhouse during the winter of 2003 and 2004 in College Station using practices standard for the production of sorghum in a greenhouse. At anthesis, anthers were removed from mature florets within 1 d of flowering. Three anthers were macerated with tweezers in a drop of 1% iodine-potassium iodide stain on a glass slide and covered with a cover slip. This process was completed on two random plants per testcross with one slide made from each plant. The respective panicles were then bagged and at maturity scored for seed set on a percentage scale. The slides were viewed with a Zeiss Universal II microscope (Carl Zeiss Inc., Gottingen, Germany). A graduation of stained pollen was observed; using three locations on each slide, pollen was classified as either fully stained, greater than 50% stained, less than 50% stained, and unstained. The term pollen stainability represents only the number of fully stained pollen grains in a sample. A minimum of 120 pollen grains were classified per slide.
Hybrids having the expected 25% pollen stainability and at least 25% seed set were classified as carrying the restoration allele. Hybrids that failed both parameters were classified as not carrying the restoration allele. The designation of the restoring alleles as dominant (Rf3 and Rf4) is tentative as, since this is a gametophytic system, restoration is based on the haploid genome of the male gamete.
Transcript Analyses
Total leaf RNA from 3 to 5 greenhouse or field-grown plants from each line was prepared as described (Tang et al., 1998), and RNA from the equivalent of 2.5 g fresh weight was electrophoresed in agarose gels, blotted to membranes, and hybridized as described (Tang et al., 1996a). The membranes were probed with clone pHC104, which spans orf107 and carries sequences of atp9 (Tang et al., 1996b).
|
RESULTS
|
|---|
Allelic Frequency of Rf3 and Rf4
A total of 128 A3FL3 testcross hybrids and 83 A3FL4 testcross hybrids were evaluated. These represented a total of 140 SC lines of which 71 were evaluated with both testcross, 57 were evaluated with only A3FL3, and 12 were evaluated with only A3FL4. The control hybrids, A3FL3/IS1112C and A3FL4/IS1112C showed pollen stainability (Fig. 1) close to 50% with high seed set (Table 2). Most of the SC lines did not restore fertility nor produce stainable pollen. These lines are all classified as rf3rf3rf4rf4 (Table 3).
View larger version (62K):
[in this window]
[in a new window]
|
Fig. 1. Iodine staining of pollen from hybrids between A3FL3 (Rf3Rf3rf4rf4) and (A) IS1112C (Rf3Rf3Rf4Rf4), (B) SC512 (Rf3Rf3Rf4Rf4), and (C) SC214 (rf3rf3Rf4Rf4) and between A3FL4 (rf3rf3Rf4Rf4) and (D) IS1112C, (E) SC512, and (F) SC214. All show approximately 50% fully stainable pollen except C (25%) and F (0%).
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Mean pollen stainability and seed set (and standard errors) of testcross hybrids with sorghum lines identified herein to possess at least one restoration allele at the Rf3 and Rf4 loci.
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Listing of lines (SC number and PI or IS number) tested in the current study for allelic composition at the Rf3 and Rf4 loci found to have no restoration function at either or both loci.
|
|
Three SC lines, SC306, SC512, and SC215, restored fertility to both hybrids and had pollen stainability ranging from 29 to 58% in both testcross hybrids, indicating each contain both restoration alleles. One of six testcross plants from A3FL3/SC306 and two of six hybrid plants from A3FL4/SC306 showed no fertility restoration activity, indicating that SC306 is probably heterozygous or heterogeneous at both loci.
Two lines, SC315 and SC214, restored fertility and showed 13 to 24% pollen stainability in testcross hybrids with A3FL3 but testcross hybrids with A3FL4 were sterile and had no pollen stainability, indicating that their allelic composition is rf3rf3Rf4Rf4. SC192 and SC235 contain the restoring allele at the Rf3 locus while SC221 and SC1108 contain the restoring allele at the Rf4 locus. However, the allelic composition at the complementary locus is unknown because the appropriate testcross was not available.
Seven entries were identified that contain restoring alleles at the Rf4 locus, and five entries that contain restoring alleles at the Rf3 locus. In the germplasm surveyed, the allelic frequency of Rf4 was 4.8% and Rf3 was 5.1%. SC512 and three others were not included in the frequency calculation since they were included nonrandomly in the sample based on previous screening information.
A number of sorghum lines showed incomplete fertility restoration with low seed set and less than expected pollen stainability with at least one tester (Table 4). These lines do not show the expected frequency of pollen stainability nor the ability to restore more than minimal fertility. This restoration activity does not mimic the complementary effect associated with a genotype carrying either Rf3 or Rf4. These lines may be carrying a similar less functional restoration allele at the Rf3 or Rf4 locus, or environmental effects precluded their identification as restorers.
View this table:
[in this window]
[in a new window]
|
Table 4. Mean pollen stainability and seed set (and standard errors) of testcrosses hybrids with sorghum lines showing fertility restoration inconsistent with the reported activity of the Rf3 and Rf4 restoration alleles.
|
|
Transcript Analyses
Transcriptional patterns of orf107 were examined for 12 hybrids with restored fertility and two sterile hybrids (Table 2), as well as for the parental lines SC306, SC512, SC192, SC235, and SC215. The five SC lines each carry Rf3. When probed with the orf107 clone pHC104, SC192 (Fig. 2A) and SC306 (Fig. 2D) displayed only a 650-nucleotide (nt) atp9 transcript, whereas three of the lines surprisingly displayed orf107 transcripts, indicating that the open reading frame was intact and transcribed. SC215 (Fig. 2B), SC235 (Fig. 2C), and SC512 (Fig. 2E) displayed 1110-, 870-, and 810-nt whole-length orf107 transcripts and the 650-nt atp9 transcript. Additionally, SC235 and SC512 exhibited enhanced orf107 TPA, as shown by the abundant 380-nt processed transcript, similar to the pattern of IS1112C (Fig. 2N). In contrast, SC215 showed no evidence of enhanced TPA, thus indicating an exception to the association of Rf3 with enhanced TPA. In hybrid combination with A3FL4, SC235 (Fig. 2G) and SC512 (Fig. 2J) conferred enhanced TPA, while SC215 (Fig. 2F) did not confer TPA, consistent with observations of the male parents. SC192, which does not carry orf107, conferred enhanced TPA in the hybrid (Fig. 2H). In other A3FL4 hybrids, SC214 (Fig. 2I) and SC315 (Fig. 2L), which were scored as rf3rf3, did not show enhanced TPA, similar to the pattern of A3Tx398 (Fig. 2M). SC306, scored as Rf3Rf3, did not confer enhanced TPA (Fig. 2K), indicating a second exception to the association of Rf3 and enhanced TPA. Seven A3FL3 hybrids each exhibited enhanced TPA (not shown), as expected since the female parent conferred enhanced processing.
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 2. Northern transcript analyses of sorghum lines probed with orf107 clone pHC104 (Tang et al. 1996b). A, SC192; B, SC215; C, SC235; D, SC306; E, SC512; F, A3FL4/SC215; G, A3FL4/SC235; H, A3FL4/SC192; I, A3FL4/SC214; J, A3FL4/SC512; K, A3FL4/SC306; L, A3FL4/SC315; M, A3Tx398; N, IS1112C. Transcripts of 1110, 870, and 810 nt represent intact orf107 transcripts, while the 380 nt transcript represents product of enhanced orf107 transcript processing. The 650 nt transcript is derived from atp9, detected by virtue of atp9 sequences in the orf107 chimeric configuration.
|
|
|
DISCUSSION
|
|---|
A3 CMS is controlled by a two-gene gametophytic restoration system and restoring alleles at both Rf3 and Rf4 are required for individual male gamete viability. Enhanced orf107 TPA was hypothesized to be tightly linked to, or to represent action of, the restoring Rf3 allele in IS1112C (Tang et al., 1998; Pring et al., 1999). Among the 83 lines examined for action of Rf3 in the A3FL4 tester line, five were shown to carry the restoring allele. Analysis of these five Rf3-carrying lines leads to the conclusion that mitochondrial orf107 and enhanced orf107 TPA are represented independently among the five lines. Three (SC215, SC235, SC512) of the five lines were characterized by the presence of mitochondrial orf107; of these, two (SC235, SC512) exhibited enhanced TPA. Of the two lines that did not carry orf107, SC306 did not confer enhanced TPA in the tester line, in contrast to enhanced TPA conferred by SC192. Since SC215 and SC306 do not confer enhanced TPA in testcross hybrids, these exceptional lines indicate that enhanced orf107 TPA is not a manifestation of gene action of the restoring allele, and thus enhanced TPA is not mandatory for fertility restoration.
Two of the five Rf3 lines, SC235 and SC512, appear to be very similar to IS1112C in regard to characteristics examined here. Both lines carry Rf3, mitochondrial orf107, and enhanced TPA; SC235 was not tested for presence of Rf4, but SC512 was shown to carry the allele. Emasculation of these lines and hybridization with a fertility maintainer, like B3Tx398 (rf3rf3rf4rf4), would reveal whether their cytoplasms are also causal of A3 CMS.
We have recovered A3 fertility restoration capability from sudangrass [S. bicolor (L.) Moench] that similarly does not involve enhanced orf107 TPA (Tang et al., unpublished data, 2005). Progeny analyses indicate that the sudangrass-derived restoration is through a sporophytic, and not gametophytic, mechanism. Although pollen stainability of A3FL4 hybrids resulting from pollination with SC215 and SC306 parallels a gametophytic pattern, analyses of backcross lines and F2 progeny will be required to establish if restoration in these cases is indeed gametophytic.
Frequency of the restoration allele at the Rf3 and Rf4 loci in the surveyed material was approximately 5% each and 2% of lines tested at both loci possessed both restoring alleles. If the loci were in linkage equilibrium, the expected frequency of individuals homozygous for both restoration alleles would be 0.000625%, which is well below the observed frequency. The cause of the observed linkage disequilibrium between the two loci is likely selection, as any genotype carrying mitochondrial orf107 will only transmit functional male gametes which contain both restoration alleles.
The restoration frequency is higher than previously reported likely because of the ability to score allelic composition at each loci involved in fertility restoration. Herein, three lines that fully restore fertility to A3 CMS were identified (SC512, SC306, and SC215) and three other lines likely contain both Rf3 and Rf4 restoration alleles based on their relatively high pollen stainability (SC192, SC221, SC1108).
Torres-Cardona et al. (1990) reported that 7.1% of evaluated lines were capable of partial fertility restoration in A3 cytoplasm with testcross hybrid seed set of between 5 and 80%. With the current knowledge that A3 CMS fertility restoration is gametophytic, seed set in that range is expected, thus at least some of those identified as partial restorers may contain both restoration alleles. In fact, SC512 (identified herein as Rf3Rf3Rf4Rf4) was indeed identified as a partial fertility restorer in that study. Other partial fertility restorers, however, do not contain restoration alleles based on our evaluation (SC603 and SC690).
Research into the genetics of A3 CMS should utilize multiple identified restoration sources to elucidate the cytoplasmic and nuclear gene action of sterility and fertility restoration. Assuming restoration is by the same two-gene gametophytic system, gene action of sterility and fertility restoration should be similar across all identified lines containing restoration alleles.
Received for publication October 18, 2005.
|
REFERENCES
|
|---|
- Bosques-Vega, A., A. Sotomayor-Rios, S. Torres-Cardona, H.R. Perrerly, and K.F. Schertz. 1987. Maintainer and restorer reaction with A1, A2, and A3 cytoplasms of lines from the Sorghum Conversion Program. Texas Agric. Exp. Stn. Misc. Pub. 1976. Texas A&M Univ., College Station.
- Kempken, F., and D. Pring. 1999. Male sterility in higher plants—Fundamentals and applications. Prog. Bot. 60:139–166.
- Pring, D.R., H.V. Tang, W. Howad, and F. Kempken. 1999. A unique two-gene gametophytic male sterility system in sorghum involving a possible role of RNA editing in fertility restoration. J. Hered. 90:386–393.[Abstract/Free Full Text]
- Pring, D.R., H.V. Tang, and K.F. Schertz. 1995. Cytoplasmic male sterility and organelle DNAs of sorghum. p. 461–495. In C.S. Levings III and I.K. Vasil (ed.) The molecular biology of plant mitochondria. Kluwer Academic Publ., Dordrecht, the Netherlands.
- Schertz, K.F., and J.M. Ritchey. 1978. Cytoplasmic-genic male-sterility systems in sorghum. Crop Sci. 18:890–893.[Abstract/Free Full Text]
- Schertz, K.F., A. Sotomayor-Rios, and S. Torres-Cardona. 1989. Cytoplasmic-nuclear male sterility—Opportunities in breeding and genetics. Proc. Grain Sorg. Res. Util. Conf. 16:175–186.
- Stephens, J.C., and R.F. Holland. 1954. Cytoplasmic male-sterility for hybrid sorghum seed production. Agron. J. 46:20–23.[Free Full Text]
- Stephens, J.C., F.R. Miller, and D.T. Rosenow. 1967. Conversion of alien sorghums to early combine genotypes. Crop Sci. 7:396.[Abstract/Free Full Text]
- Tang, H.V., R. Chang, and D.R. Pring. 1998. Cosegregation of single genes associated with fertility restoration and transcript processing of sorghum mitochondrial orf107 and urf209. For. Genet. 150:383–391.
- Tang, H.V., and D.R. Pring. 2003. Conversion of fertility restoration of the sorghum IS1112C (A3) male-sterile cytoplasm from two genes to one gene. Crop Sci. 43:1747–1753.[Abstract/Free Full Text]
- Tang, H.V., D.R. Pring, F.R. Muza, and B. Yan. 1996a. Sorghum mitochondrial orf25 and a related chimeric configuration of a male-sterile cytoplasm. Curr. Genet. 29:265–274.[Web of Science][Medline]
- Tang, H.V., D.R. Pring, L.C. Shaw, F.A. Salazar, F.R. Muza, B. Yan, and K.F. Schertz. 1996b. Transcript processing internal to a mitochondrial open reading frame is correlated with fertility restoration in male-sterile sorghum. Plant J. 10:123–133.[CrossRef][Web of Science][Medline]
- Torres-Cardona, S., A. Sotomayor-Rios, A. Quiles Belén, and K.F. Schertz. 1990. Fertility restoration to A1, A2, and A3 cytoplasm systems of converted sorghum lines. Texas Agric. Exp. Stn. Misc. Pub. 1721. Texas A&M Univ., College Station.
- Worstell, J.V., H.J. Kidd, and K.F. Schertz. 1984. Relationships among male-sterility inducing cytoplasms of sorghum. Crop Sci. 24:186–189.[Abstract/Free Full Text]
- Xu, G.-W., Y.-X. Cui, K.F. Schertz, and G.E. Hart. 1995. Isolation of mitochondrial DNA sequences that distinguish male-sterility-inducing cytoplasms in Sorghum bicolor (L.) Moench. Theor. Appl. Genet. 90:1180–1187.
This article has been cited by other articles:
|
|
|
|
 
H. V. Tang, J. F. Pedersen, C. D. Chase, and D. R. Pring
Fertility Restoration of the Sorghum A3 Male-Sterile Cytoplasm through a Sporophytic Mechanism Derived from Sudangrass
Crop Sci.,
May 31, 2007;
47(3):
943 - 950.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
