Chromosome Instabilities and Programmed Cell Death in Tapetal Cells of Maize With B Chromosomes and Effects on Pollen Viability

Corresponding author: María J. Puertas, Facultad de Biología, Universidad Complutense, José Antonio Novais 2, 28040 Madrid, Spain., majetas{at}bio.ucm.es (E-mail)

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

B chromosomes (B's), knobbed chromosomes, and chromosome 6 (NOR) of maize undergo nondisjunction and micronucleus formation in binucleate tapetal cells. These chromosome instabilities are regular events in the program of tapetal cell death, but the B's strongly increase A chromosome instability. We studied 1B and 0B plants belonging to selected lines for high or low B transmission rate and their F₁ hybrids. These lines are characterized by meiotic conservation or loss of B chromosomes, respectively. The female B transmission (fBt^l) allele(s) for low B transmission is dominant, inducing micronucleus formation and B nondisjunction. We hypothesize that the fBt^l allele(s) induces knob instability. This instability would be sufficient to produce B loss in both meiocytes and binucleate tapetal cells. B instability could, in turn, produce instabilities in all chromosomes of maize complement. To establish whether the chromosomal instabilities are related to the tapetal programmed cell death (PCD) process, we applied the TUNEL technique. PCD, estimated as the frequency of binucleate tapetal cells with TUNEL label, was significantly correlated with the formation of micronuclei and the frequency of pollen abortion. It can be concluded that the observed chromosome instabilities are important to the PCD process and to the development of microspores to form viable pollen grains.

THE tapetum is the innermost cell layer that lines the anther locules. It is in direct contact with the pollen mother cells (PMCs). The tapetal tissue has a secretory role providing essential nutrients required for microspore and pollen grain development. The peculiar physiological and cytological features of tapetum development are well documented in a number of taxa (PACINI 1997 , PACINI 2000 ; RAGHAVAN 1997 ). In maize, after at least one mitosis with frequent abnormal chromatid segregation, cytokinesis is not completed and binucleate tapetal cells are formed when the PMCs are at meiotic prophase. The tapetal tissue disintegrates at the dehiscent pollen stage when the anther breaks down to allow pollination. Aberrant divisions of tapetal nuclei are an intrinsic characteristic of tapetal development, in such a way that disturbances of tapetal degeneration result in male sterility (LEE et al. 1979 ; RAGHAVAN 1997 ).

The identification and functional characterization of genes expressed in the tapetum are critically important due to the diverse degrees of male sterility of their mutant phenotypes. For example, CIGAN et al. 2001 found that in maize the MS45 protein is localized to the tapetum and expressed maximally during the early vacuolate microspore stage of development. WILSON et al. 2001 determined that the Arabidopsis MS1 gene is a critical sporophytic controlling factor for anther and pollen development, with expression in the tapetum at the stage of microspore release from tetrads. KAPOOR et al. 2002 reported a functional characterization of the tapetum-specific zinc-finger gene TAZ1, which causes premature degeneration and pollen abortion in petunia. MURRAY et al. 2003 studied the expression of the transcription factor HvGAMYB during early development of barley anthers. SORENSEN et al. 2003 found that the AMS gene in Arabidopsis encodes a transcription factor playing a crucial role in tapetal cell and microspore development. Some of these mutations have been cytologically characterized. For example, FEI and SAWHNEY 2001 described the effects of MS33 mutation on pollen development at the ultrastructural level. It interferes with intine formation and tryphine deposition in Arabidopsis.

Genes related to the deposition of pollen wall components from the tapetum are also being characterized. Some of these products play a role in the highly specific interactions at the pollen stigma interface during sexual reproduction. In maize pollen coats, the predominant protein is an endoxylanase whose mRNA is located in the tapetum. WU et al. 2002 localized prexylanase in the tapetum and studied how it is transferred to the pollen coat. KARIMI et al. 2002 found that genes coding for pollenins are expressed in the anthers and more specifically in the tapetum of Arabidopsis. LAI et al. 2002 identified the AhSLF-S-2 gene related to the incompatibility multiallelic S locus in Antirrhinum. WANG et al. 2002 found that three TAA1 homeologous genes in wheat express specifically within the tapetum cells. Their temporal expression matched the assembly of wall-impregnated sporopollenin.

PAPINI et al. 1999 and WANG et al. 1999 proposed that tapetal degeneration is a process of programmed cell death (PCD). Evidence that leads to the conclusion that anther development in angiosperms culminates in the PCD of specific tissues to facilitate the release of pollen has accumulated (reviewed in WU and CHEUNG 2000 ).

A number of workers find morphological and biochemical similarities between PCD processes in plants and animals. For example, JONES 2000 states that the mitochondrion is a common factor that can serve in plant and animal cell death as a stress sensor and as a dispatcher of PCD. Similarly, BEERS and MCDOWELL 2001 indicate that the mitochondrion is a mediator of at least some forms of PCD. Moreover, SOLOMON et al. 1999 found cysteine proteases in soybean as the key enzymes in animal PCD, whereas DE JONG et al. 2000 stated that caspase-like proteases are involved in plant PCD. LINCOLN et al. 2002 indicate that plants possess a protease with substrate-site specificity that is functionally equivalent to certain animal caspases. However, plants display a unique PCD mechanism that depends on vacuolar lytic function (KURIYAMA and FUKUDA 2002 ).

In this and a previous article (CHIAVARINO et al. 2001 ), we use fluorescence in situ hybridization (FISH) to study binucleate tapetal cells in maize carrying B chromosomes. The FISH technique allowed us to determine that B chromosomes, knobbed chromosomes, and chromosome 6 (NOR) undergo nondisjunction and micronucleus formation. We concluded that these instabilities are regular events in tapetal cell death development, but B's strongly increase A chromosome instability.

In this work we extend that study to 1B and 0B plants belonging to selected lines for high (H) or low (L) B transmission rate and both HL and LH F₁ hybrids, to better determine the genetic control and significance of chromosome instabilities during anther development. In addition, to establish whether the chromosomal instabilities are related to the PCD process, we applied the TdT-mediated dUTP nick end labeling (TUNEL) technique to H, L, HL, LH, 0B, and 1B plants. We also compared pollen viability in the mature pollen grains of the same plants to determine how these alterations influence pollen development.

As far as we know, this is the first work in which the behavior of specific chromosomes is related to the PCD process.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The materials used were 0B and 1B plants from selected lines of maize, Zea mays ssp. mays. The original material from which the selection was initiated belongs to the native race Pisingallo from northwest Argentina naturally carrying B chromosomes (ROSATO et al. 1998 ). The selection process was carried out following the method described in ROSATO et al. 1996 , which consists of selecting plants with the highest and the lowest B transmission rate in female 1B x male 0B crosses. As the B's were always transmitted on the female side we called these lines H^f and L^f in previous works, but in this study we call them H and L for simplicity. F₁ hybrids between the lines were also obtained (GONZALEZ-SANCHEZ et al. 2003 ). HL F₁ hybrids were obtained by crossing 1B H x 0B L plants, whereas 1B L x 0B H plants were crossed to obtain LH F₁ hybrids.

Root tips were scored for B number, and then the plants were grown in an experimental field. Male inflorescences at meiosis were fixed in 3:1 ethanol:acetic acid and refrigerated at 4° until analyzed. Anther squashes for determining the convenient meiotic or pollen grain stage were made in 1% acetocarmine. Binucleate tapetal cells were scored in anthers from diplotene-metaphase I for chromosome instability studies using either Feulgen staining or FISH. For the TUNEL detection, male inflorescences were fixed in 0.1% paraformaldehyde. Anthers from pachytene tetrads were used. The last tapetal mitosis was found in anthers at zygotene.

The FISH procedure was the same as in CHIAVARINO et al. 2001 , using the following repetitive DNA sequences as probes:

1. pZmBs, a clone containing the maize B chromosome-specific sequence (ALFENITO and BIRCHLER 1993 ), kindly provided by J. A. Birchler (Columbia, Missouri). This probe labels the B centromeric regions and, in especially good slides, a small B telomeric region is also labeled.

2. pZm4-21, a clone containing the maize 180-bp knob repeat (PEACOCK et al. 1981 ), kindly provided by J. A. Birchler (Columbia, Missouri).

3. pTa71, a clone containing the rDNA gene unit; the 5.8S, 18S, and 28S genes; and the intergenic spacer from Triticum aestivum (GERLACH and BEDBROOK 1979 ).

The TUNEL technique was used to determine the relative stage of PCD. Binucleate tapetal cells were scored in anthers at three PMC stages: pachytene, metaphase I, and tetrads. In situ detection of DNA fragmentation was carried out with a modification of the TUNEL method (TILLY and HSUEH 1993 ; SURH and SPRENT 1994 ). Squashed preparations were incubated in pepsin 0.1% for 4 min at 37°, fixed in 4% paraformaldehyde (Sigma, St. Louis) in 2x SSC for 30 min at room temperature, and then washed in 2x SSC. For labeling, slides were incubated for 60 min at 37° in the presence of a terminal deoxynucleotidyl transferase (TdT) with the in situ Cell Death Detection kit (Boehringer Mannheim, Mannheim, Germany), according to manufacturer's instructions. Controls were made without TdT. Slides were counterstained with 4',6-diamidino-2-phenylindole (Boehringer Mannheim) and mounted in Vectashield (Vector Laboratories, Burlingame, CA).

Alexander staining was used as a conventional and simple technique to distinguish normal (fully stained) vs. aborted (empty) pollen grains. Anthers containing either bicellular or tricellular pollen grains were scored.

Photographs were made in an epifluorescence Olympus microscope equipped with a CCD camera. The images were optimized for best contrast and brightness using Adobe Photoshop 7.0.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Binucleate tapetal cells were scored for the presence of micronuclei with standard Feulgen staining in 0B and 1B plants of the H, HL, LH, and L genotypes, using at least three individuals per genotype and at least three anthers per individual. Table 1 shows that H plants have fewer micronuclei than HL, LH, and L plants and that, in these three genotypes, 1B plants have more micronuclei than 0B plants (Fig 1A). A two-way ANOVA showed that there are significant differences among genotypes (F = 5.73; P = 0.004) and between 0B and 1B plants (F = 10.63; P = 0.003), the interaction being nonsignificant (F = 0.93; P = 0.44). The least significant difference test (LSD) post hoc test shows that there are nonsignificant differences among the HL, LH, and L genotypes, but the H genotypes are significantly different from the other three.

View larger version (51K): In this window In a new window Download PPT slide   Figure 1. (A–F) Mean frequency of micronuclei, B nondisjunction, unequal knob distribution, chromosome 6 nondisjunction, PCD, and aborted pollen grains in 0B () and 1B () plants of the H, HL, LH, and L genotypes.

The pZmBs probe, specific to the maize B chromosomes, was used in FISH experiments to study the B behavior in binucleate tapetal cells in 1B plants of the four genotypes. The results are shown in Table 2. Unequal B distribution was observed, as one of the nuclei of the binucleate cell showed two labels and the other showed no label (Fig 2A). This indicates that B nondisjunction occurred in the mitotic anaphase preceding binucleate tapetal cell formation. One-way ANOVA showed that there are significant differences among the genotypes (F = 6.75; P = 0.0025), but the LSD post hoc test shows that there are nonsignificant differences among HL, LH, and L, whereas significantly less B nondisjunction occurs in the H line (Fig 1B).

View larger version (65K): In this window In a new window Download PPT slide   Figure 2. (A–C) Localization of the maize B-specific probe (red) in binucleate tapetal cells. (A) B nondisjunction. The labels corresponding to the B centromere and telomere are side by side. (B) B in the micronucleus. (C) Normal disjunction of the B and unlabeled micronucleus. (D) Localization of the B-specific probe (red) and the knob-specific probe (green). B nondisjunction and unequal knob distribution is shown. (E) Localization of the B-specific probe (red) and the chromosome 6-specific probe (green). Normal distribution of the B and nondisjunction of chromosome 6 is shown. (F) Localization of the chromosome 6-specific probe (green). Chromosome 6 in the micronucleus is shown. (G and H) TUNEL labeling in binucleate tapetal cells. (G) Positive TUNEL, tapetal cell strongly labeled. (H) Negative TUNEL, cell slightly labeled on the surface. (I) Alexander staining in pollen grains of 1B H plants. Aborted pollen grains are not fully stained.

From the total of 129 cells observed with micronuclei, 19 (14.73%) were labeled with the pZmBs probe (Table 2; Fig 2B and Fig C). This indicates not only that the B formed micronuclei in the tapetal cells, but also that other chromosomes of the normal complement suffered instabilities leading to their loss in the preceding anaphase. It is evident that the B is more unstable than the A's, because if all chromosomes had the same probability of forming micronuclei, the expected frequency of B micronuclei would be 1/21 (4.76%) and the expected frequency for any A chromosome would be 20/21 (95.24%). However, the observed relative proportion of B micronuclei is much higher (14.73%).

Double FISH was carried out with the pZmBs probe, specific to the maize B's, and the pZm4-21 probe, specific to the maize heterochromatic knobs, simultaneously. Previous data (CHIAVARINO et al. 2001 ) indicated that this maize race is polymorphic for the heterochromatic knobs, with five large and at least three small knobs. Due to this large number of knobs it is not possible to determine the exact number of knobs in each nucleus of the binucleate cells, but it is possible to distinguish between nuclei with apparently equal or unequal knob distribution (Fig 2D). Table 3 shows the frequency of binucleate tapetal cells with equal or unequal knob distribution in 0B and 1B plants of the four genotypes. A two-way ANOVA showed that there are significant differences among genotypes (F = 5.31; P = 0.004) and between 0B and 1B plants (F = 7.49; P = 0.009), the interaction being nonsignificant (F = 0.39; P = 0.76). In all cases 1B plants show more knob unequal distribution than 0B plants (Fig 1C). LSD post hoc test shows that in 1B plants the L genotype has significantly more unequal knob distribution than the other three. In 0B plants the H and L genotypes significantly differ, whereas the HL and LH genotypes are intermediate.

1B plants were studied simultaneously for the distribution of B's and knobs. In Table 4 the observed cells of the four genotypes are classified in four classes according to the normal disjunction or nondisjunction of the B's and the knobs (Fig 2D). Chi-square contingency tests were made to test if B disjunction vs. nondisjunction and equal vs. unequal knob distribution are independent events. In the H and LH lines the differences were significant (² = 15.72, P = 0.0001 and ² = 2.76, P = 0.097, respectively), whereas in the HL and L lines the differences were nonsignificant (² = 2.76, P = 0.09 and ² = 0.16, P = 0.69, respectively). Contingency tables show that in the H, HL, and LH genotypes the observed number of normal distribution of both the B and the knobs and nondisjunction of the B and the knobs was higher than expected. Conversely, the expected number of cells showing unequal distribution of only one of the chromosomes was lower than expected.

The pTa71 probe was used in FISH experiments to study the behavior of chromosome 6 in binucleate tapetal cells in 0B and 1B plants of the four genotypes. This probe is specific to chromosome 6, which is the only maize chromosome carrying the nucleolar organizing region, located on the short arm. The results are shown in Table 5.

Nondisjunction of one or both chromosome 6's was observed (Fig 2E). The number of observed cells with nondisjunction of both chromosome 6's is small, but it seems that it is lower than that expected for randomness, which is the squared frequency of nondisjunction of one chromosome 6.

A two-way ANOVA showed that there are nonsignificant differences of chromosome 6 nondisjunction among genotypes (F = 0.74; P = 0.53), but significant differences between 0B and 1B plants (F = 47.08; P = 0.0000), the interaction being nonsignificant (F = 1.96; P = 0.13). In all cases 1B plants show a higher level of nondisjunction of chromosome 6 than do 0B plants (Fig 1D).

This probe also allowed observation of labeled micronuclei in the tapetal cells, demonstrating that chromosome 6 may be lost during the mitosis preceding binucleate tapetal cell formation (Fig 2F). The number of cells with labeled micronuclei is small, but labeled micronuclei were not observed in the H line, indicating again that the H line forms fewer micronuclei than the other three. It can also be observed that the frequency of labeled micronuclei is 18/83 (21.69%), which corresponds to 10.84 per chromosome 6. Comparing this frequency with the frequency of micronuclei with B label (19/129; 14.73%; Table 2), it can be concluded that the B is lost with a higher frequency. However, the frequency of chromosome 6 forming micronuclei is higher than the random probability for any chromosome to be lost (1/21; 4.76%).

The pTa71 and the pZmBs probes were simultaneously used in FISH experiments in 1B plants to study the distribution of both the B and the chromosome 6's (Table 6). Contingency tables show that in all cases the observed number of normal distribution of both the B and the 6 and nondisjunction of the B and the 6 was higher than expected. Conversely, the observed number of cells showing nondisjunction of only one of the chromosomes was lower than expected, but the differences were significant in three of the four cases. Chi-square contingency tests were made to test whether disjunction vs. nondisjunction of the B and chromosome 6 are independent events. In the H, HL, and L lines the differences were significant (² = 6.41, P = 0.0; ² = 11.94, P = 0.000; and ² = 14.96, P = 0.0001, respectively), whereas in the LH line the differences were nonsignificant (² = 0.55, P = 0.46).

The TUNEL technique was used to study the possible effect of these chromosome instabilities on the PCD process. The anthers were analyzed at three PMC stages: pachytene, metaphase I, and tetrads. The chromosomes in the PMCs are always labeled along their entire length, probably due to the DNA breakage produced at early meiotic stages related to crossing over. The binucleate tapetal cells were found either strongly labeled on the whole nuclei or slightly labeled on the surface (Fig 2G and Fig H) and the number of both types was scored.

Considering all genotypes as a whole, the frequency of binucleate tapetal cells showing strong TUNEL label increased from pachytene to tetrads, indicating that the normal PCD process progresses as the PMCs develop, although it never reached 100% at these stages. However, remarkable differences were found in some cases. Table 7 shows the frequency of binucleate tapetal cells showing strong TUNEL label in 1B and 0B plants of the four genotypes when the PMCs are at metaphase I. A two-way ANOVA showed that there are nonsignificant differences of TUNEL label between 0B and 1B plants (F = 4.48; P = 0.06), but significant differences among genotypes (F = 15.04; P = 0.0005), the interaction being nonsignificant (F = 2.95; P = 0.08). However, an LSD post hoc test showed that there are significant differences between 1B and 0B plants of the H line and nonsignificant differences between 1B and 0B plants of HL, LH, and L lines. On the other hand, HL, LH, and L lines showed a higher number of cells with TUNEL label than did H lines (Fig 1E).

The frequency of normal and aborted pollen was estimated using the Alexander staining in 1B and 0B plants of the four genotypes (Table 8; Fig 2I). A two-way ANOVA showed that there are significant differences of pollen grain abortion among genotypes (F = 7.88; P = 0.0001) and between 0B and 1B plants (F = 4.43; P = 0.038), the interaction being also significant (F = 4.52; P = 0.006). The LSD post hoc test shows that the H line has more aborted pollen, and 1B plants of the H line have more aborted pollen than any other genotype (Fig 1F).

To determine whether the chromosome instabilities and micronuclei observed in the tapetal cells were related to the PCD process and pollen abortion, we studied the correlations between the following variables: micronucleus frequency, B nondisjunction frequency, knob unequal distribution, chromosome 6 nondisjunction, PCD (as frequency of TUNEL labeled cells), and pollen abortion. Since the variables related to B chromosomes can be studied only in 1B plants, we calculated the correlation coefficients in 1B and 0B plants separately. Table 9 shows the significant correlation coefficients, indicating that B nonsdisjunction, micronucleus frequency, and PCD are negatively correlated with pollen abortion.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

B chromosomes are extra chromosomes that are not fully integrated into the normal behavior of the standard complement. In many plant species they are unstable at meiosis and undergo nondisjunction at pollen mitosis determining particular B transmission mechanisms. In maize, B nondisjunction regularly occurs at the second pollen mitosis (RANDOLPH 1941 ; CARLSON 1986 ), in the endosperm and in the tapetum. In these latter cases the B's may induce instabilities in the A's (RHOADES et al. 1967 ; RHOADES and DEMPSEY 1972 , RHOADES and DEMPSEY 1973 ; ALFENITO and BIRCHLER 1990 ; CHIAVARINO et al. 2001 ). These alterations might produce deleterious quantitative phenotypic effects, particularly related to fertility, because chromosome instabilities occur at various stages of sexual reproduction, whereas they are somatically stable. Remarkably, there are very few works on B maize effects on fitness variables in spite of the large literature on maize B chromosomes (JONES and REES 1982 ).

Previous works of our laboratory have characterized the genetic control of maize B transmission rate (GONZALEZ-SANCHEZ et al. 2003 ). One gene located on the A chromosomes, which we call mBt (male B transmission), controls B preferential fertilization on the male side. Female B transmission is controlled by a gene(s) called fBt (female B transmission) also located on the A's. The H and L lines used in this work are the fBt^h fBt^h and the fBt^l fBt^l homozygous, respectively. In the L line and in the LH and HL F₁ hybrids (fBt^l fBt^h heterozygous) a significant loss of the B chromosome occurs both at meiosis of 1B plants and in the progeny of 1B x 0B crosses.

Chromosome instability is easily observed as micronucleus formation, which is a regular event in binucleate tapetal cells, because micronuclei appear in plants of all constitutions. The B chromosome is particularly unstable because it forms 15% of the micronuclei, whereas 1/21 is expected. However, L, HL, and LH plants form more micronuclei than the H line does, showing, first, a genetic component in this process and, second, that the fBt^l allele(s) is dominant at this level. We think that micronucleus formation is also affected by environmental conditions because we found a significantly higher frequency of micronuclei in a previous study (CHIAVARINO et al. 2001 ).

The frequency of B nondisjunction in the binucleate tapetal cells is similarly related to the studied genotypes. Also in this case the fBt^l allele(s) behaves as a dominant, because H plants show significantly less B nondisjunction than do L, HL, and LH plants.

A parallel behavior of the B chromosome occurs in PMCs and in the tapetum. B univalents are conserved at male meiosis in the H line, whereas they are lost in a significant proportion of the microspores of the L line and the HL and LH hybrids (GONZALEZ-SANCHEZ et al. 2003 ). This shows that the fBt^l allele(s) is dominant and that there is not a maternal effect, because in both hybrids the B is lost both at meiosis and in its transmission with the same frequency as in the L line. We have no data on the B meiotic behavior on the female side because a quantitative study of maize female meiosis is unattainable. However, it is reasonable to accept that B behavior at female meiosis is similar to that of male meiosis because B loss or conservation at male meiosis and female B transmission are strongly correlated. Therefore, it can be suggested that the fBt^l allele(s) produces B instability in both the PMCs and the binucleate tapetal cells. Since the fBt^l allele(s) is dominant, it can also be suggested that its expression is sporophytic and occurring at the diploid level, before the reduction of chromosome number.

The similarity between PMCs and tapetal cells is remarkable. Both the sporogenous and tapetal cells have a common origin from the archesporial cells (ECHLIN 1973 ; CANALES et al. 2002 ) and remain in close contact during their development. YANG et al. 1999 , analyzing the SPL mutation in Arabidopsis, suggested that the development of the anther walls, the tapetum, and the microsporocytes is tightly coupled. Interestingly, ARAGON-ALCAIDE et al. 1997 found association of homologous chromosomes simultaneously both in premeiotic PMCs and in the surrounding tapetal cells of wheat. It may be hypothesized that sporophytic to gametophytic transition is initiated with a signal terminating in tapetal PCD on one side and pollen grain maturation on the other.

Knobbed chromosomes also show instabilities in binucleate tapetal cells. In all cases 1B plants show more knob unequal distribution than 0B plants, and L plants show more unequal distribution than H plants, but HL and LH behave as intermediates. On the other hand, in H, HL, and LH the observed number of normal distribution of both the B and the knobs and nondisjunction of both the B and the knobs was higher than expected. This suggests that instability is extended to all chromosomes when either the B's or the knobbed chromosomes become unstable.

Our hypothesis is that the fBt^l allele is directly related to a differential knob constitution present in the L line and absent in the H line. If the fBt^l allele(s) induces knob instability, the B, which also carries a knob, is also affected. Probably, the B would be affected more than any other chromosome because of its own special constitution. This instability would be sufficient to induce B loss in many cells, both PMCs and binucleate tapetal cells. B instability could, in turn, produce instabilities in all of the chromosomes of the maize complement. This hypothesis would also explain the intermediate behavior of the HL and LH F₁ hybrids for the unequal knob distribution, because the L line would have two of these unstable knobs, whereas the hybrids would have only one.

This hypothesis is in agreement with the instabilities observed in chromosome 6. In this case nondisjunction of chromosome 6 is not related to the H, L, HL, and LH genotypes, but 1B plants show more instabilities than 0B plants. In all cases, the observed number of normal distribution of both the B and the 6 and nondisjunction of both the B and the 6 was higher than the nondisjunction of only one of the chromosomes. This is again in agreement with the hypothesis that B instability induces instability in the remaining chromosomes, particularly in knobbed chromosomes like chromosome 6. In this case, chromosome 6 is not affected by the fBt^l allele.

It is interesting to study whether these chromosome instabilities are related to the PCD process, which is an essential feature of the binucleate tapetal cells. GRANELL 1999 , BUCKNER et al. 1998 , BUCKNER et al. 2000 , WU and CHEUNG 2000 , and GIULIANI (2002) reported reviews of the PCD program in plants where the TUNEL technique is mainly used (JONES 2001), but FISH is not generally used in PCD studies to reveal the fate of specific chromosomes during cell death. In our laboratory both TUNEL and FISH have been used, FISH being particularly useful to show chromosome instabilities in binucleate tapetal cells when the PMCs are at meiotic stages.

A higher frequency of cells showing TUNEL label was found in L, HL, and LH plants than in the H line at metaphase I. This result is not surprising because one would expect that higher levels of chromosome instabilities would result in acceleration of the PCD process. It is, however, unexpected that 1B plants of the H line suffer a significant delay with respect to 0B plants because 1B plants are always more unstable.

Interestingly, the PCD process, estimated as the frequency of binucleate tapetal cells with TUNEL label, is significantly correlated with the frequency of pollen abortion. L, HL, and LH plants show very low frequency of pollen abortion, whereas the plants of the H line show significantly higher pollen abortion, particularly 1B ones. It can be concluded that the chromosome instabilities observed in the L, HL, and LH plants are important to the PCD process and to the development of microspores to form viable pollen grains. We do not have sufficient information to determine whether or not the fBt gene(s) is directly related to the PCD process. However, the death signal transduction occurs via pleiotropic signaling pathways (KURIYAMA and FUKUDA 2002 ) and PCD seems to be affected by the expression of fBt.

The fBt^l dominant allele(s), inducing knobbed chromosome instabilities, increases plant fitness because plants with this allele(s) have nearly 100% of viable pollen grains and tend to lose the B's. Conversely, the fBt^h recessive allele produces chromosome stability and, consequently, lack of B meiotic loss in 1B plants. It is possible that B presence during microspore development produces deleterious effects, the frequency of pollen aborted reaching 22.5%. Unfortunately, we have no data showing a possible reduction of fertility on the female side in 1B plants.

As instabilities of knobbed chromosomes and chromosome 6 are not correlated with TUNEL label frequency or pollen abortion, it can be deduced that A chromosome instabilities induced by B chromosomes do not affect normal pollen development. This indicates the slight parasitic nature of maize B chromosomes and reinforces the idea that the fBt^l allele(s) provides the defense of the A complement against the B chromosome attack (GONZALEZ-SANCHEZ et al. 2003 ).

	ACKNOWLEDGMENTS

This work was supported by grant BOS 2002-3572 of the Dirección General de Investigación of Spain.

Manuscript received October 3, 2003; Accepted for publication November 10, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALFENITO, M. R. and J. A. BIRCHLER, 1990  Studies on B chromosome stability during development. Maydica 35:359-366.

ALFENITO, M. R. and J. A. BIRCHLER, 1993  Molecular characterization of a maize B chromosome centric sequence. Genetics 135:589-597.[Abstract]

ARAGÓN-ALCAIDE, L., S. READER, A. BEVEN, P. SHAW, and T. MILLER et al., 1997  Association of homologous chromosomes during floral development. Curr. Biol. 7:905-908.[CrossRef][Medline]

BEERS, E. P. and J. M. MCDOWELL, 2001  Regulation and execution of programmed cell death in response to pathogens, stress and developmental cues. Curr. Opin. Plant Biol. 4:561-567.[CrossRef][Medline]

BUCKNER, B., D. JANIK-BUCKNER, J. GRAY, and G. S. JOHAL, 1998  Cell-death mechanisms in maize. Trends Plant Sci. 3:218-223.

BUCKNER, B., G. S. JOHAL, and D. JANIK-BUCKNER, 2000  Cell death in maize. Physiol. Plant 108:231-239.[CrossRef]

CANALES, C., A. M. BHATT, R. SCOTT, and H. DICKINSON, 2002  EXS, a putative LRR receptor kinase, regulates male germline cell number and tapetal identity and promotes seed development in Arabidopsis. Curr. Biol. 12:1718-1727.[CrossRef][Medline]

CARLSON, W. R., 1986  The B chromosome of maize. Crit. Rev. Plant Sci. 3:201-226.

CHIAVARINO, A. M., M. ROSATO, M. MANZANERO, G. JIMÉNEZ, and M. GONZÁLEZ-SÁNCHEZ et al., 2001  Chromosome nondisjunction and instabilities in tapetal cells are affected by B chromosomes in maize. Genetics 155:889-897.

CIGAN, A. M., E. UNGER, R. J. XU, T. KENDALL, and T. W. FOX, 2001  Phenotypic complementation of ms45 maize requires tapetal expression of MS45. Sex. Plant Reprod. 14:135-142.

DE JONG, A. J., F. A. HOEBERICHTS, E. T. YAKIMOVA, E. MAXIMOVA, and E. J. WOLTERING, 2000  Chemical-induced apoptotic cell death in tomato cells: involvement of caspase-like proteases. Planta 211:656-662.[CrossRef][Medline]

ECHLIN, P., 1973 The role of the tapetum during microsporogenesis of Angiosperms, pp. 41–61 in Pollen, Development and Physiology, edited by J. HESLOP-HARRISON. Butterworths, London.

FEI, H. M. and V. K. SAWHNEY, 2001  Ultrastructural characterization of male sterile33 (ms33) mutant in Arabidopsis affected in pollen desiccation and maturation. Can. J. Bot. 79:118-129.[CrossRef]

GERLACH, W. L. and J. R. BEDBROOK, 1979  Cloning and characterization of ribosomal RNA genes from wheat and barley. Nucleic Acid Res. 7:1869-1885.[Abstract/Free Full Text]

GIULIANI, C., G. CONSONNI, G. GAVAZZI, M. COLOMBO, and S. DOLFINI, 2002  Programmed cell death during embryogenesis in maize. Ann. Bot. 90:287-292.[Abstract/Free Full Text]

GONZÁLEZ-SÁNCHEZ, M., E. GONZÁLEZ-GONZÁLEZ, F. MOLINA, A. M. CHIAVARINO, and M. ROSATO et al., 2003  One gene determines maize B chromosome accumulation by preferential fertilisation; another gene(s) determines their meiotic loss. Heredity 90:122-129.[CrossRef][Medline]

GRANELL, A., 1999 Dying according to programme: occurrence in plant developmental processes and in response to environmental cues, pp. 159–177 in Enviromental Stress and Gene Regulation, edited by K. B. STOREY. BIOS Scientific, Oxford.

JONES, A., 2000  Does the plant mitochondrion integrate cellular stress and regulate programmed cell death? Trends Plant Sci. 5:225-230.[CrossRef][Medline]

JONES, A. M., S. COIMBRA, A. FATH, H. SOTTOMAYOR, and H. THOMAS, 2001  Programmed cell death assays for plants. Methods Cell Biol. 66:437-451.[Medline]

JONES, R. N., and H. REES, 1982 B Chromosomes. Academic Press, London.

KAPOOR, S., A. KOBAYASHI, and H. TAKATSUJI, 2002  Silencing of the tapetum-specific zinc finger gene TAZ1 causes premature degeneration of tapetum and pollen abortion in petunia. Plant Cell 14:2353-2367.[Abstract/Free Full Text]

KARIMI, M., C. L. D. MANES, M. VAN-MONTAGU, and G. GHEYSEN, 2002  Activation of a pollenin promoter upon nematode infection. J. Nematol. 34:75-79.

KURIYAMA, H. and H. FUKUDA, 2002  Developmental programmed cell death in plants. Curr. Opin. Plant Biol. 5:568-573.[CrossRef][Medline]

LAI, Z., W. S. MA, B. HAN, L. Z. LIANG, and Y. S. ZHANG et al., 2002  An F-box gene linked to the self-incompatibility (S) locus of Antirrhinum is expressed specifically in pollen and tapetum. Plant Mol. Biol. 50:29-42.[CrossRef][Medline]

LEE, S. L. J., V. E. GRACEN, and E. D. EARLE, 1979  The cytology of pollen abortion in C-cytoplasmic male-sterility corn anthers. Am. J. Bot. 66:656-667.[CrossRef]

LINCOLN, J. E., C. RICHAEL, B. OVERDUIN, K. SMITH, and R. BOSTOCK et al., 2002  Expression of the antiapoptotic baculovirus p35 gene in tomato blocks programmed cell death and provides broad-spectrum resistance to disease. Proc. Natl. Acad. Sci. USA 99:15217-15221.[Abstract/Free Full Text]

MURRAY, F., R. KALLA, J. JACOBSEN, and F. GUBLER, 2003  A role for HvGAMYB in anther development. Plant J. 33:481-491.[CrossRef][Medline]

PACINI, E., 1997  Tapetum character states: analytical keys for tapetum types and activities. Can. J. Bot. 75:1448-1459.

PACINI, E., 2000  From anther and pollen ripening to pollen presentation. Plant Syst. Evol. 222:19-43.[CrossRef]

PAPINI, A., S. MOSTI, and L. BRIGHIGNA, 1999  Programmed-cell-death events during tapetum development of angiosperms. Protoplasma 207:213-221.[CrossRef]

PEACOCK, W. J., E. S. DENNIS, M. M. RHOADES, and A. PRYOR, 1981  Highly repeated DNA sequence limited to knob heterochromatin in maize. Proc. Natl. Acad. Sci. USA 78:4490-4494.[Abstract/Free Full Text]

RAGHAVAN, V., 1997 Molecular Embriology of Flowering Plants. Cambridge University Press, Cambridge, UK.

RANDOLPH, L. F., 1941  Genetic characteristics of the B chromosomes in maize. Genetics 26:608-631.[Free Full Text]

RHOADES, M. M. and E. DEMPSEY, 1972  On the mechanism of chromatin loss induced by the B chromosome of maize. Genetics 71:73-96.[Abstract/Free Full Text]

RHOADES, M. M. and E. DEMPSEY, 1973  Chromatin elimination induced by the B chromosome of maize. I. Mechanism of loss and the pattern of endosperm variegation. J. Hered. 64:13-18.[Free Full Text]

RHOADES, M. M., E. DEMPSEY, and A. GHIDONI, 1967  Chromosome elimination in maize induced by supernumerary B chromosomes. Proc. Natl. Acad. Sci. USA 57:1626-1632.[Free Full Text]

ROSATO, M., A. M. CHIAVARINO, C. A. NARANJO, M. J. PUERTAS, and L. POGGIO, 1996  Genetic control of B chromosome transmission rate in Zea mays ssp. mays (Poaceae). Am. J. Bot. 83:1107-1112.[CrossRef]

ROSATO, M., A. M. CHIAVARINO, C. A. NARANJO, J. CÁMARA HERNÁNDEZ, and L. POGGIO, 1998  Genome size and numerical polymorphism for the B chromosome in races of maize (Zea mays ssp. mays, Poaceae). Am. J. Bot. 85:168-174.[Abstract]

SOLOMON, M., B. BELENGHI, M. DELLEDONNE, E. MANACHEM, and A. LEVINE, 1999  The involvement of cysteine proteases and protease inhibitor genes in the regulation of programmed cell death in plants. Plant Cell 11(3):431-443.[Abstract/Free Full Text]

SORENSEN, A. M., S. KROBER, U. S. UNTE, P. HUIJSER, and K. DEKKER et al., 2003  The Arabidopsis aborted microspores (AMS) gene encodes a MYC class transcription factor. Plant J. 33:413-423.[CrossRef][Medline]

SURH, C. D. and J. SPRENT, 1994  T-cell apoptosis detected in situ during positive and negative selection of the thymus. Nature 372:100-103.[CrossRef][Medline]

TILLY, J. L. and A. J. W. HSUEH, 1993  Microscale autoradiographic method for the qualitative and quantitative analysis of apoptotic DNA fragmentation. J. Cell. Physiol. 154:519-526.[CrossRef][Medline]

WANG, A. M., Q. XIA, W. S. XIE, T. DUMONCEAUX, and J. T. ZOU et al., 2002  Male gametophyte development in bread wheat (Triticum aestivum L.): molecular, cellular, and biochemical analyses of a sporophytic contribution to pollen wall ontogeny. Plant J. 30:613-623.[CrossRef][Medline]

WANG, M., S. HOEKSTRA, S. VAN BERGEN, G. E. M. LAMERS, and B. J. OPPEDIJK et al., 1999  Apoptosis in developing anthers and the role of ABA in this process during androgenesis in Hordeum vulgare L. Plant Mol. Biol. 39:489-501.[CrossRef][Medline]

WILSON, Z. A., S. M. MORROLL, J. DAWSON, R. SWARUP, and P. J. TIGHE, 2001  The Arabidopsis MALE STERILITY1 (MS1) gene is a transcriptional regulator of male gametogenesis, with homology to the PHD-finger family of transcription factors. Plant J. 28:27-39.[CrossRef][Medline]

WU, H. M. and A. Y. CHEUNG, 2000  Programmed cell death in plant reproduction. Plant Mol. Biol. 44:267-281.[CrossRef][Medline]

WU, S. S. H., D. F. SUEN, H. C. CHANG, and A. H. C. HUANG, 2002  Maize tapetum xylanase is synthesized as a precursor, processed and activated by a serine protease, and deposited on the pollen. J. Biol. Chem. 277:49055-49064.[Abstract/Free Full Text]

YANG, W. C., D. YE, J. XU, and V. SUNDARESAN, 1999  The sporocyteless gene of Arabidopsis is required for initiation of sporogenosis and encodes a novel nuclear protein. Genes Dev. 13:2108-2117.[Abstract/Free Full Text]

This article has been cited by other articles:

				V. V. Lia, V. A. Confalonieri, and L. Poggio B Chromosome Polymorphism in Maize Landraces: Adaptive vs. Demographic Hypothesis of Clinal Variation Genetics, October 1, 2007; 177(2): 895 - 904. [Abstract] [Full Text] [PDF]

				F. Han, J. C. Lamb, W. Yu, Z. Gao, and J. A. Birchler Centromere Function and Nondisjunction Are Independent Components of the Maize B Chromosome Accumulation Mechanism PLANT CELL, February 1, 2007; 19(2): 524 - 533. [Abstract] [Full Text] [PDF]