Logo of plntcellLink to Publisher's site
Plant Cell. 2002 Oct; 14(10): 2353–2367.
PMCID: PMC151222

Silencing of the Tapetum-Specific Zinc Finger Gene TAZ1 Causes Premature Degeneration of Tapetum and Pollen Abortion in Petunia


TAZ1 (TAPETUM DEVELOPMENT ZINC FINGER PROTEIN1; renamed from PEThy; ZPT3-2) cDNA was first isolated as an anther-specific cDNA from petunia. Here, we report a functional characterization that includes analysis of spatial and temporal expression profiles and examination of anther phenotypes in TAZ1-silenced plants. TAZ1 showed a biphasic expression pattern. In the premeiotic phase, TAZ1 transcripts were found to accumulate in all cell types of the anther except the tapetum and gametophytic tissues, whereas the postmeiotic phase of anther development was characterized by expression exclusively in the tapetum. Silencing of TAZ1 by cosuppression resulted in aberrant development and precocious degeneration of the tapetum, followed by extensive microspore abortion that started soon after their release from pollen tetrads. A few pollen grains that survived showed reduced flavonol accumulation, defects in pollen wall formation, and poor germination rates. This study demonstrates an essential role for TAZ1 in the postmeiotic phase of tapetum development.


Development of the male gametophyte (pollen) occurs within the anther compartment of the stamen and requires cooperative functional interactions between gametophytic and sporophytic tissues (Raghavan, 1997a). The tapetum is the innermost of the four sporophytic layers of the anther wall that comes in direct contact with the developing gametophyte; therefore, it has long been considered to play a crucial role in the maturation of microspores (Shivanna et al., 1997). Early studies using microscopic and cytological techniques revealed that tapetal cells contain dense cytoplasm with abundant nuclei and other cell organelles, indicating the metabolically dynamic state of these cells (Stevens and Murray, 1981). The plasma membrane facing the developing microspores in these cells was shown to develop membrane-lined lipid bodies (orbicules), which were thought to contribute to the synthesis of the pollen wall (Steer, 1977). Additionally, a number of male-sterile mutants were found to have defects in the structure of the tapetum (Raghavan, 1997b). The indispensability of the tapetum during microspore development was demonstrated further by the specific ablation of tapetal cells by targeted expression of cytotoxin genes, which resulted in the premature abortion of microspores (Goldberg et al., 1993).

The understanding of the function of the tapetum prompted the identification of tapetum-specific genes such as those coding for callase (Hird et al., 1993) and the enzymes involved in the biosynthesis of lipids (Foster et al., 1992) and flavonoids (Koes et al., 1990; van Tunen et al., 1990). Genes coding for oleosins, which are specialized proteins that enclose naked oil droplets of storage triglycerides, and those for small Gly- or Cys-rich secretory proteins that expressed specifically or preferentially in the tapetum, also were characterized (Aguirre and Smith, 1993; Chen and Smith, 1993; Robert et al., 1994; Ross and Murphy, 1996; Ruiter et al., 1997). Regarding the hierarchy of gene function during flower development, the examples cited above belong to the category of downstream genes that are required for the function of the tapetum. On the other end of this regulatory hierarchy, a number of “master” regulatory genes that are involved in the establishment of the floral meristem and the determination of floral organ identity also have been characterized (reviewed by Theiben, 2001). The most neglected aspect of flower development, however, has been the delineation of the middle-level regulatory mechanisms that commence after floral organ primordia have been established and direct the completion of organ differentiation (Smyth, 2001).

We previously identified a number of genes for floral organ–specific zinc finger transcription factors in petunia by PCR-based screening of organ-specific cDNA libraries (Takatsuji, 1999). Seven such anther-specific genes were found to express sequentially but transiently during the course of anther development, suggesting that they might form a regulatory cascade to control anther development (Kobayashi et al., 1998). Of these, the expression peak of TAZ1 (TAPETUM DEVELOPMENT ZINC FINGER PROTEIN1; renamed from PEThy-ZPT3-2), which encodes a protein with three zinc finger motifs, preceded those of six other genes during the early stages of anther development (Kobayashi et al., 1998). Here, we report detailed expression profiles and the functional characterization of TAZ1. The data show a distinctive biphasic expression pattern of TAZ1 during the early stages of anther development in petunia. Analysis of transgenic plants, in which the TAZ1 gene was silenced by cosuppression, demonstrated that TAZ1 activity is essential for the postmeiotic phase of tapetum development, which in turn plays an essential role in microspore maturation. Based on these data, a possible role for TAZ1 in the transcriptional regulation associated with the development and function of the tapetum as well as the maturation of pollen is discussed.


Temporal and Spatial Expression Pattern of TAZ1 in Sporophytic Anther Tissues

To determine the temporal expression of TAZ1, RNA gel blot analysis was performed using anthers from 12 early (bud lengths [Bl] 3 to14 mm at 1-mm intervals) and 5 late (Bl 15 to 20, 20 to 30, 30 to 40, 40 to 50, and 50 to 60 mm) stages of flower development. The early stages up to Bl 14 mm displayed TAZ1 expression ranging from strong (Bl 3 to 4 mm) to medium (Bl 5 to 8 mm) to low (Bl 9 to 14 mm) (Figure 1). Bl 3 to 4 mm corresponds to the pollen mother cell stage of microspore development, which is followed by meiosis in Bl 4 to 5 mm. The uninucleate stage of microspore development extends from Bl 6 to 20 mm. Meanwhile, the tapetum reaches the peak of its development by Bl 6 to 8 mm; thereafter, it gradually starts to degenerate. By Bl 35 mm, the tapetum has disappeared completely from the anthers. Notably, the disappearance of TAZ1 mRNA coincided well with the degeneration of the tapetum.

Figure 1.
Temporal Expression Pattern of TAZ1.

To define the cell type–specific expression pattern of TAZ1, in situ hybridization analysis was performed on transverse sections of 3-, 4-, 5-, 6-, 7-, 9-, and 13-mm flower buds. The entire TAZ1 cDNA was used to prepare sense and antisense RNA probes, as described in Methods. This analysis revealed two distinct phases of TAZ1 expression (Figure 2). In the early developmental stages up to Bl 5 mm (i.e., phase I), TAZ1 expression was detected in all of the cell types of anther except tapetum and the sporogenous tissue (Figures 2A to 2I). In sharp contrast to this, during phase II (Bl 6 to 13 mm), the TAZ1 mRNAs were detected exclusively in the tapetum (Figures 2J to 2U).

Figure 2.
Analysis of the Spatial and Temporal Expression Patterns of TAZ1 by in Situ Hybridization.

In the early stages of phase I, TAZ1 expression was detected uniformly in the connective, vascular tissue, wall layers, and stomium cells (Figures 2B and 2C). However, with the progression of development, it gradually started to concentrate in the stomium and the connective cells adjoining the inner tapetum (Figures 2H and 2I). Most notably, soon after the separation of individual microspores from tetrads in Bl 6 mm, there was a dramatic shift in the TAZ1 expression pattern that marked the beginning of phase II (Figures 2J to 2U). In this phase, expression was localized exclusively in the tapetum, whereas in the surrounding tissues, it was reduced to almost undetectable levels. In tapetal cells, the high levels of TAZ1 expression continued at least until Bl 8 mm (Figures 2Q and 2R), and as the anther matured, there was an overall decrease in the expression levels that coincided with the degeneration of the tapetum. In agreement with the RNA gel blot data (Figure 1), transverse sections of anthers from Bl 13 mm showed significant reductions in TAZ1 expression (Figures 2T and 2U). Collectively, these data demonstrate two distinct phases in which TAZ1 expression is regulated in a spatial and temporal manner.

A 2-kb Upstream Region Is Sufficient for the Phase II–Specific Expression of TAZ1

To delineate the promoter region of TAZ1, we constructed transgenic petunia plants with a chimeric gene containing a β-glucuronidase (GUS) coding sequence fused to an ~2-kb upstream region of TAZ1. Five independently transformed lines were selected for the characterization of the promoter activity. GUS expression was localized mainly in the anthers of young flower buds in these lines (Figure 3A). Further histochemical analysis localized the GUS expression mostly in the tapetal cells; however, a low level of GUS activity was detected in the adjoining cells of the middle layer and in the connective (Figures 3B and 3C). In contrast to the RNA gel blot and in situ hybridization analyses, which displayed relatively higher TAZ1 expression during phase I, GUS activity was undetectable in phase I flower buds (Bl < 6 mm; data not shown). The absence of GUS activity in young buds may suggest that (1) the cis element responsible for TAZ1 expression in the premeiotic phase existed either farther upstream of the 2-kb promoter region or downstream of the translation initiation site, or (2) the 271-bp TAZ1 5′ untranslated region (Figure 3D) contained some regulatory sequence(s) that suppressed the translation of the transcripts during phase I.

Figure 3.
TAZ1 Promoter-Driven GUS Expression in Transgenic Plants.

TAZ1-GFP Is Targeted to the Nucleus

The presence of three transcription factor IIIA (TFIIIA)-type zinc finger motifs strongly suggested that TAZ1 is a transcription factor. To qualify as a transcription factor, however, a protein must be translocated into the nucleus to interact with its target DNA. TAZ1 contains two basic motifs, 5′-KKCKKLNPFGSRYYKKR-3′ (amino acids 81 to 99) and 5′-KKKKKK-3′ (amino acids 263 to 270), resembling the bipartite and the monopartite nuclear localization signals, respectively. Based on the presence of these basic motifs, it was predicted to be a nucleus-localized protein with a certainty of 0.96 by PSORT (www.psort.nibb.ac.jp). To determine whether TAZ1 is imported into the nucleus in vivo, onion peel cells were bombarded with 35S-sGFP-TAZ1 and a control plasmid (35S-sGFP) (Figure 4B). In the case of the control plasmid, the green fluorescent protein (GFP) was distributed throughout the cytoplasm and the nucleus (Figure 4A, left). By contrast, the recombinant protein sGFP-TAZ1 was localized exclusively in the nucleus (Figure 4A, right). These data indicate that TAZ1 is a nuclear protein that probably interacts with the DNA and functions as a transcription factor.

Figure 4.
TAZ1 Is a Nucleus-Localized Protein.

Functional Analysis of TAZ1 Using TAZ1-Silenced Plants

The complete TAZ1 cDNA (cTAZ1) under the control of the 35S promoter of Cauliflower mosaic virus (CaMV 35S promoter) was introduced into petunia by Agrobacterium tumefaciens–mediated transformation (Figure 5A). The accumulation of TAZ1 transcripts in transgenic plants was analyzed by gel blot analysis of the total RNA extracted from leaves of 15 transformants. Six transgenic lines (1, 10, 12, 16, 17, and 19) showed moderate to high accumulation of TAZ1 transcripts in leaves (Figure 5B). However, they did not display any apparent physiological or morphological aberrancy under normal greenhouse conditions; therefore, they were not analyzed further. The transgenic lines that did not express TAZ1 RNA in leaves were analyzed for its expression in anthers. The RNA samples were prepared from pooled anthers from Bl 3 to 6 mm (two buds at each 1-mm interval), representing both phase I and phase II of anther development. Of the nine transgenic lines that did not show TAZ1 expression in leaves, five (2, 8, 13, 14, and 20) accumulated TAZ1 RNA in anthers at levels similar to that in the wild type and were morphologically and developmentally indistinguishable from the wild type. The rest of the transgenic lines (5, 9, 15, and 18), however, did not show any TAZ1 expression in anthers, suggesting cosuppression of TAZ1 gene expression (Figure 5C). Because the RNA preparations contained representative anthers from both phase I and phase II (described above), it is possible that TAZ1 expression was silenced in all tissues of the anther, including the tapetum, in these four transgenic lines.

Figure 5.
Analysis of TAZ1 Transcripts in CaMV35S::TAZ1 Transgenic Petunia Plants.

The CaMV 35S promoter has been shown to be expressed in the vascular tissue, connective, and wall layers but not in the tapetum (Plegt and Bino, 1989; van der Meer et al., 1992). Therefore, it is unlikely that TAZ1 expression under the control of the CaMV 35S promoter could have triggered cosuppression directly in the tapetal cells. Because the silencing signal is known to propagate through the plant (Vaucheret et al., 1998), one can postulate that a TAZ1 silencing signal translocated from the site of its initiation, presumably from the cells of the connective and/or anther wall, to the adjoining cells of the tapetum. In these lines, TAZ1 transgene expression was not detected even in leaves.

To verify the specificity of TAZ1 silencing, we analyzed the expression of another zinc finger–encoding gene, ZPT2-5, in TAZ1-silenced lines (Figure 5D). Of the seven anther-specific zinc finger genes described previously, the expression profile of only ZPT2-5 overlapped with that of TAZ1, whereas others were expressed later during anther development (Kobayashi et al., 1998). Furthermore, the conserved zinc finger–encoding regions in TAZ1 and ZPT2-5 displayed the highest level (~68%) of sequence similarity among all of the anther-specific zinc finger genes. The level of ZPT2-5 transcripts in TAZ1-silenced lines was the same as that in the wild type, confirming that the silencing was specific to TAZ1. These plants had extremely low pollen counts at the time of dehiscence. Apparently, however, there was no effect on female fertility, because cross-pollination of TAZ1-silenced flowers by wild-type pollen yielded normal seed set.

Development of the Tapetum in TAZ1-Silenced Plants

Transverse sections of anthers from wild-type and TAZ1-silenced plants were analyzed for the effect of TAZ1 silencing on various cell types of the anther (Figure 6). Early stages of anther development (phase I) showed no noticeable effect on the development of the wall layers, connective, vascular tissue, and stomium cells. In wild-type plants, tapetum in the anthers at Bl 4 mm consists of a single layer of columnar cells (Figure 6A). Thereafter, the tapetum undergoes a phase of active cell proliferation to become multilayered by the time individual microspores are released from pollen tetrads (Bl 6 mm; Figure 6C). At this stage, the tapetum cells are characterized by a dense cytoplasm and a few small vacuoles. By Bl 9 mm, the tapetum is reduced to two layers, but the size of individual tapetal cells increases significantly (Figure 6E). The gain in cell size continues until the Bl 13-mm stage, concomitant with the degeneration of inner tapetum layers. At the Bl 13-mm stage, the tapetum is left with only a single layer of cells with less dense, vacuolated cytoplasm (Figure 6G). The degeneration process continues and the tapetum disappears completely by the time the flower buds are 35 mm in size (data not shown).

Figure 6.
Silencing of TAZ1 Results in Poorly Developed Tapetum.

In TAZ1-silenced lines, the development of tapetum at the Bl 4-mm stage was comparable to that in the wild type in having a monoseriate tapetum at the completion of meiosis (Figure 6B). However, soon after the separation of individual haploid microspores from tetrads, the tapetal cells started to degenerate. Noticeably, at the Bl 6-mm stage, the tapetum in TAZ1-silenced lines consisted of a single layer of vacuolated cells, instead of the multilayered tapetum that was observed in wild-type anthers at this stage. The extent of tapetum degeneration at this stage in TAZ1-silenced anthers was comparable to that of the Bl 13-mm stage in wild-type plants (Figures 6D and 6G). The degeneration of tapetal cells continued during the Bl 9-mm stage, and most of the cells had collapsed by the Bl 13-mm stage (Figures 6F and 6H).

Effects of TAZ1 Silencing on Pollen Development

Premature Pollen Abortion

The TAZ1-silenced plants had drastically reduced numbers of pollen (~1000/anther) compared with wild-type plants (~30,000/anther) (Figure 7). Moreover, the surviving pollen grains in these lines were found to be fragile, because most of them burst within 10 min of incubation in 10% glycerol solution (data not shown). In this solution, wild-type pollen grains remained intact for >24 h.

Figure 7.
Pollen Degeneration as a Result of TAZ1 Silencing.

Poor Viability

To examine the viability of the surviving pollen in TAZ1-silenced lines, they were germinated under in vitro and semi-in vivo conditions, as described in Methods. As shown in Figure 8, wild-type pollen demonstrated nearly 100% germination under both conditions. Most of the pollen tubes grew ~400 μm under in vitro conditions and rarely showed bursting (Figure 6A). Under semi-in vivo conditions, the pollen tubes grew longer than 1 cm in 14 h (Figure 8C). By contrast, pollen grains from TAZ1-silenced lines rarely germinated. In the small fraction that germinated, the pollen tubes did not grow >80 μm in length (Figure 8B). Furthermore, most of these pollen tubes displayed burst tips (Figure 8B, inset). Under semi-in vivo germination conditions, the pollen grains from TAZ1-silenced lines showed poor germination, because few pollen tubes were visible at the cut end of the style (Figure 8D).

Figure 8.
Effects of TAZ1 Silencing on Pollen Germination.

Reduced Flavonol Accumulation

The flavonols are synthesized in tapetal cells and are released subsequently into the anther locule to be taken up by the developing microspores (Mo et al., 1992; Ylstra et al., 1994). To determine whether precocious degeneration of the tapetum had any effect on the supply of flavonols to the developing microspores, both young and mature pollen grains from wild-type and TAZ1-silenced plants were stained for flavonols. In microspore tetrads (Figures 9A and 9B), there was no apparent difference in the accumulation of flavonols between the wild-type and TAZ1-silenced lines (Figures 9A and 9B). However, mature pollen grains in TAZ1-silenced lines displayed markedly reduced (~30-fold less) fluorescence of flavonols compared with that in the wild type (Figures 9C to 9F). Furthermore, the collapsed pollen grains in TAZ1-silenced lines fluoresced blue under UV light, suggesting that the walls of these microspores did not accumulate any flavonols. To determine whether flavonols generally were absent in the pollen wall or that this absence resulted from the silencing of TAZ1 expression, intact pollen grains from wild-type and TAZ1-silenced lines were ruptured mechanically, washed, and stained for flavonols. Like the naturally collapsed pollen, intentionally ruptured pollen showed very little flavonol accumulation in TAZ1-silenced lines. By contrast, the walls of wild-type pollen displayed significant accumulation of flavonols (Figures 9G and 9H). Together, these data indicate that precocious degeneration of the tapetum resulting from the silencing of TAZ1 has adverse effects on the synthesis of flavonols and/or their transport into the developing microspores and that these effects manifest themselves only after the completion of male meiosis.

Figure 9.
TAZ1 Silencing Results in Flavonol-Deficient Pollen.

Anomalies in Pollen Wall Development

To search for defects in pollen wall formation, mature pollen grains were analyzed using a scanning electron microscope. Wild-type pollen grains were spherical and had sculptured exine walls. There were three germination pores through which future pollen tubes could emerge (Figure 10A). The surviving pollen grains in TAZ1-silenced lines had normal-appearing exine sculpturing, yet the wall formation, especially around the germination pores, was incomplete (Figure 10B). The diameter of these pollen grains was up to 30% larger than that of the wild type, and they displayed characteristically bulging germination pores.

Figure 10.
Characterization of the Pollen Wall.

All plant cells, including pollen grains, undergo plasmolysis when placed in hypertonic solutions. Wild-type mature pollen grains showed characteristic separation of the plasma membrane from the pollen wall (consisting of exine and intine) when incubated in 20% glycerol solution (Figures 10C and 10E). At this concentration of glycerol, however, there was no effect on the pollen grains of TAZ1-silenced lines (data not shown). Even when the glycerol concentration was increased to 50%, these pollen grains remained intact (Figure 10D). However, after 2 h of incubation in this solution, the intine started to separate from the exine near the germination pores and appeared to be pulled inside, albeit slightly, along with the plasma membrane (Figures 10D and 10F). The inability of the intine to separate from the exine under strong osmotic pressure indicates abnormalities during the synthesis of the intine and/or the plasma membrane. Because precursors of pollen wall are provided mainly by the tapetum, this phenotype probably is attributable to premature degeneration of the tapetal cells.


TAZ1 belongs to the 5-enolpyruvylshikimate-3-phosphate synthase promoter binding factor family of TFIIIA-type zinc finger transcription factors, in which two Cys and two His residues tetrahedrally coordinate a zinc atom to form a compact structure that interacts with the major groove of DNA in a sequence-specific manner (Takatsuji, 1996). The 5-enolpyruvylshikimate-3-phosphate synthase promoter binding factor family differs from the animal TFIIIA-type zinc finger protein in having a conserved QALGGH motif in the putative DNA-contacting domain and in the presence of long spacers of varied lengths between zinc fingers (Takatsuji, 1999). TAZ1 codes for a 444–amino acid polypeptide that contains three zinc finger domains. Other members of this class, such as ZPT2-1 and ZPT2-2, have been shown to interact with two tandem but separated core DNA sequences in a manner that is dependent on the length of the spacer between the two core sites (Takatsuji and Matsumoto, 1996). Existing homologies between these two zinc finger proteins and TAZ1 suggest that TAZ1 is a DNA binding protein. This notion is supported further by the subcellular localization of TAZ1 in the nucleus.

The expression pattern of TAZ1 is summarized along with the stages of anther development in Figure 11A. Based on the spatial and temporal distribution of its transcripts, the expression of TAZ1 consists of two distinct phases. In phase I, it is expressed in all of the cell types of the anther except for the tapetum and sporogenous cells; in phase II, TAZ1 expression is localized exclusively in the tapetum.

Figure 11.
Summary of TAZ1 Expression and Proposed Model of Its Role.

During phase I of anther development, which is represented by flower buds up to Bl 5 to 6 mm in petunia, anther primordia differentiate into specific cells and tissues, the morphology of the anther is established, and sporogenous cells undergo meiosis to produce haploid microspores (Goldberg et al., 1993). After their differentiation, tapetal cells divide mostly anticlinally in petunia, forming a single layer of columnar cells that lines the anther locule at the end of meiosis. During meiosis, tapetal cells also undergo nuclear divisions, resulting in an overall increase in DNA content from 2 to 4C (during meiotic prophase) to 7 to 8C by the end of meiosis (Liu et al., 1987). Because they are intensely involved in the replication of DNA, tapetal cells have been shown to lack poly(A) RNA. For this reason, the tapetum has been suggested to play no active role in the nutrition of microsporocytes before and during meiosis (Raghavan, 1989). In the sequence of developmental events, the first known contribution of the tapetum to the developing microspores is the release of callase, which frees individual microspores by digesting the callose wall around tetrads (Frankel et al., 1969; Stieglitz and Stern, 1973). In TAZ1-silenced lines, we found no apparent anomaly in the development of tissues (e.g., connective, vascular tissue, anther wall, and stomium) that expressed TAZ1 during the premeiotic phase of anther development. Moreover, none of the developmental events mentioned above were affected by TAZ1 silencing. The microsporocytes underwent normal meiosis and produced pollen tetrads. A single-layered tapetum also was observed at the tetrad stage, and individual microspores were separated, suggesting adequate and timely production of callase from the tapetum. The absence of TAZ1 transcripts in both phase I and phase II anthers of TAZ1-silenced lines was confirmed by RNA gel blot analysis. Therefore, the apparently wild-type-like phenotype of premeiotic anthers in TAZ1-silenced lines suggests either that the function of TAZ1 is redundant with those of other genes or that this gene has no function in cell types other than the tapetum. A possible translational block as a result of the 5′ untranslated region of TAZ1, which was suggested by promoter-GUS experiments, could account for the absence of a phenotype during phase I.

From the time of completion of meiosis in the microsporocytes to the stage at which the tapetum begins to disintegrate, tapetal cells are characterized by the highest rates of transcriptional activity of all of the cell types in the anther (Raghavan, 1981, 1989). During this phase (phase II), TAZ1 was expressed exclusively in the tapetum. The beginning of phase II also was characterized by the rapid multiplication of tapetal cells. As a result of both periclinal and anticlinal divisions, the number of tapetal layers increased from one (monoseriate) at the tetrad stage to at least three (triseriate) by the Bl 6-mm stage. After Bl 6 mm and at least until Bl 13 mm, the size of individual tapetal cells increased nearly twofold to fivefold; however, because of the gradual degeneration of the inner layers, the tapetum was left with only a single layer of cells by the Bl 13-mm stage. By contrast, the phase of tapetum proliferation was completely absent in TAZ1-silenced lines. Soon after the release of individual microspores from tetrads, tapetal cells began to shrink and collapse, and by the Bl 9-mm stage, the tapetum was represented by a single layer of darkly stained collapsed cells in the anthers of TAZ1-silenced lines.

The importance of the tapetum in pollen development has been demonstrated using the “genetic laser” technique (Koltunow et al., 1990; Mariani et al., 1990). In these experiments, selective destruction of the tapetum caused by the expression of cytotoxin genes under the control of a tapetum-specific TA-29 promoter resulted in the complete degeneration of pollen grains in tobacco. Recently, mutation in a tapetum-specific Arabidopsis gene, MALE STERILE1 (MS1), was demonstrated to cause precocious degeneration of the tapetum and complete male sterility (Wilson et al., 2001). MS1 is a plant homeodomain protein finger protein that lacks any apparent similarity to TAZ1. The reported expression pattern of the MS1 gene was similar to that of TAZ1 during phase II. As with TAZ1, the accumulation of MS1 transcripts begins only after male meiosis is complete and individual microspores have separated from the tetrads. Furthermore, both TAZ1 and MS1 code for putative transcription factors. The similarity in the expression pattern and the phenotypes resulting from the lack of their respective expression suggest that both TAZ1 and MS1 could serve as components of the regulatory mechanism that controls the postmeiotic phase of tapetum development.

Based on the data presented here, we propose a hypothesis for the possible role of TAZ1 (Figure 11B). According to this model, both constructive and degenerative forces act simultaneously during phase II of tapetum development. The degenerative forces are governed by the programmed cell death mechanism, whereas TAZ1, along with MS1, forms an essential component of the constructive forces that cause tapetal cells to proliferate, allowing them to play their role in sustaining pollen development. In the absence of TAZ1 (in silenced lines), however, the programmed cell death mechanism proceeds unchecked and causes precocious degeneration of the tapetum. The absence of the cell proliferation and synthesis phase in the TAZ1-minus development deprives the developing microspores of nutrients, resulting in their premature abortion.


Cloning of the TAZ1 Gene and Its Promoter

The TAZ1 gene was isolated by screening a genomic DNA library of Petunia hybrida var Mitchell in EMBL3 vector. Subsequently, a 2.2-kb EcoRI fragment containing a 1991-bp promoter region upstream of the ATG initiation codon and 209 bp of the coding region was subcloned in pBluescript SK+ (Stratagene, La Jolla, CA) and sequenced. The sequence was submitted to DDBJ.

Plasmid Construction


The TAZ1::GUS reporter gene was constructed by fusing a 2.0-kb upstream fragment (from the 2.2-kb EcoRI fragment mentioned above) of the TAZ1 gene to the β-glucuronidase (GUS) coding sequence. A 20-base oligonucleotide (5′-GATCTATATGTCGACATA-TA-3′; the SalI site is underlined) was self-hybridized and introduced as a linker into the BglI site that was located 13 bp downstream of the TAZ1 translation initiation codon. Thereafter, the 2012-bp HindIII-SalI fragment (HindIII is of pBluescript SK+; SalI is a restriction site introduced as part of the linker) was spliced into the same sites of pUCAP (van Engelen et al., 1995), placing it 5′ to the GUS gene that was cloned already in SmaI and SacI sites of this vector. This synthetic gene containing TAZ1(promoter)::GUS::Tnos was excised from pUCAP using AscI and PacI and was inserted into pBINPLUS (van Engelen et al., 1995) using the same set of enzymes.


A SalI-NotI fragment containing the complete TAZ1 cDNA (cTAZ1) sequence (Kobayashi et al., 1998) was excised from pSPORT vector using KpnI and SacI. This fragment was inserted between the 35S promoter of Cauliflower mosaic virus (CaMV35S) and the nopaline synthase (Nos) terminator, which were cloned already in the SalI and EcoRI sites, respectively, of pUCAP (van Engelen et al., 1995). This synthetic gene containing CaMV35S::cTAZ1::Tnos was excised from pUCAP using AscI and PacI and was inserted into pBINPLUS (van Engelen et al., 1995) using the same set of enzymes.


The complete TAZ1 coding sequence was amplified by PCR using two oligonucleotide primers, 5′-ACTAGGGCCCATGGTTGATTATCA-AGATCTTCAAGTTGGG-3′ and 5′-ACTAGGGCCCTTAAATTGGAA-AAAATGTAAAATACTGATGATCACGG-3′. The underlined regions denote XmaI sites that were added to both primers to facilitate the cloning of amplified DNA. After digestion with XmaI, the coding region of TAZ1 was ligated to the XmaI-linearized psGFPcs vector (Jiang et al., 2001), placing it between the green fluorescent protein (GFP) coding sequence and the Nos terminator.

Plant Transformation

The TAZ1::GUS and 35S::TAZ1::NosT chimeric genes were introduced into petunia plants by Agrobacterium tumefaciens (GV3101)–mediated transformation (Jorgensen et al., 1996). After regeneration on selective medium, transformed petunia lines were checked for the presence of the transgene by PCR of the neomycin phosphotransferase II sequence and were transferred to a greenhouse.

Transient expression of the 35S::GFP::TAZ1 gene in onion peel cells was performed as described by Jiang et al. (2001). The fluorescence of sGFP fusion proteins was observed with a confocal microscope (Bio-Rad Laboratories, Hercules, CA), and the resulting digital micrographs were assembled using Photoshop software (Adobe Systems, San Jose, CA).

RNA Gel Blot Analysis

Total cellular RNA from petunia leaves and anthers was isolated using the procedure developed by Logemann et al. (1987) and stored at −70°C in 95% ethanol. Aliquots (10 μg) of RNA were separated on 1.2% agarose gels containing 0.4 M formaldehyde and transferred to Gene Screen membranes (DuPont–New England Nuclear Life Science Products, Boston, MA). A digoxigenin-labeled antisense RNA corresponding to the 1172-bp 3′ terminal region (3′ of the EcoRI site at position 717) of cTAZ1, prepared according to the manufacturer's instructions (Boehringer Mannheim), was used as a probe. Hybridization of blots and detection of chemiluminescence also were performed according to the Boehringer Mannheim protocol.

In Situ Hybridization

In situ hybridization was essentially carried out by using Ribomap kit and Discovery automatic staining module (Ventana Medical Systems, Tucson, AZ according to manufacturer's instructions). The excised anthers were fixed in formaldehyde 10%, acetic acid 5%, and ethanol 50% for four days at 4°C. The cTAZ1 was digested with EcoRI and SpeI. The 491-, 243-, 263-, 167-, 154-, 434-, and 172-bp fragments generated were cloned in pBluescript SK+. Both antisense and sense probes from all of the cTAZ1 fragments were synthesized using the SP6/T7 digoxigenin RNA labeling kit (Roche Diagnostics, Indianapolis, IN), according to the manufacturer's instructions. The sense and antisense digoxigenin-labeled RNAs were pooled and used for hybridization with the fixed anther sections at 60°C for 6 h in Ribohyb hybridization solution (Ventana Medical Systems). After hybridization, sections were washed in 0.1 × SSC (three times for 6 min) at 70°C. Detection of hybrids was performed using the digoxigenin nucleic acid detection kit (Boehringer Mannheim), according to the manufacturer's instructions. Detection of hybrids with nitroblue tetrazolium salt and 5-bromo-4-chloro-3-indolyl phosphate 4-toluidinium salt was performed at 20°C for 16 h in the dark. Sections were dehydrated through an ethanol series (30, 50, 70, 90, and 100% ethanol for 2 min each) and washed twice for 5 min in xylene before mounting in Malinol mounting medium (Muto Pure Chemicals, Tokyo, Japan).

Cultivation and Observation of Pollen Tubes

For in vitro culture, pollen grains from a single anther were dusted onto a petri dish containing solid pollen germination medium [0.05% Ca(NO3)2·4H2O, 0.01% H3BO3, 5% Suc, and 0.15% gellan gum] according to the protocol described by Higashiyama (2000). Plates were incubated for 14 h at 25°C and then analyzed and photographed using the phase-contrast option in a DMR microscope (Leica Microscopy Systems, Wetzlar, Germany).

For semi-in vivo germination, pollen grains of either wild-type or TAZ1-silenced plants were smeared onto the stigmas of wild-type pistils. The styles were cut (1 cm from the top) at 15 min after pollination and placed onto the solid pollen germination medium. To compensate for the low pollen count in TAZ1-silenced lines, three anthers were used to pollinate a single stigma, instead of one in the case of the wild type. This experiment was performed in triplicate. After 14 h, the plates were examined using the dark-field settings of the DMR microscope.

Detection of Flavonols

Flavonol staining was performed essentially as described (Ylstra et al., 1994). Pollen from freshly dissected anthers was placed for 2 h in a saturated solution (<0.5%, w/v) of diphenylborinic acid ethanolamine ester (Aldrich) with 0.01% Triton X-100 and 10% glycerol. After washing with a solution of 10% glycerol, the pollen was visualized and photographed under UV light using the DMR microscope.

Pollen Plasmolysis

The anthers were pressed open (by gently pressing the cover glass to release the pollen grains) in 20-μL hypertonic solutions (13 to 50% glycerol) on glass slides. After 5 min of incubation at room temperature, the solutions were covered with glass cover slips and photographed using the Nomarski optics (interference microscopy) option of the DMR microscope.

Upon request, all novel materials described in this article will be made available in a timely manner for noncommercial research purposes. No restrictions or conditions will be placed on the use of any materials described in this article that would limit their use for noncommercial research purposes.


Hiroshi Kouchi and James Doughty are gratefully acknowledged for their valuable suggestions regarding in situ hybridization experiments. We are also grateful to Akiko Baba-Kasai for providing us with psGFPcs vector, to Osamu Ueno and A.B.M. Siddique for their help during experiments involving microtomy and light microscopy, and to Renu Wadhwa for her suggestions during the preparation of the manuscript. This work was supported by Center of Excellence, Special Coordination Funds for the Promotion of Science and Technology from the Science and Technology Agency of Japan and a PROBRAIN grant from the Bio-Oriented Technology Research Advancement Institution of Japan.


Article, publication date, and citation information can be found at www.plantcell.org/cgi/doi/10.1105/tpc.003061.


  • Aguirre, P.J., and Smith, A.G. (1993). Molecular characterization of a gene encoding a cysteine-rich protein preferentially expressed in anthers of Lycopersicon esculentum. Plant Mol. Biol. 23, 477–487. [PubMed]
  • Chen, R., and Smith, A.G. (1993). Nucleotide sequence of a stamen- and tapetum-specific gene from Lycopersicon esculentum. Plant Physiol. 101, 1413. [PMC free article] [PubMed]
  • Foster, G.D., Robinson, S.W., Blundell, R.P., Robert, M.R., Hodge, R., Draper, J., and Scott, R.J. (1992). A Brassica napus mRNA encoding a protein homologous to phospholipid transfer proteins is expressed specifically in the tapetum and developing microspores. Plant Sci. 84, 187–192.
  • Frankel, R., Izhar, S., and Nitsan, J. (1969). Timing of callase activity and cytoplasmic male sterility in Petunia. Biochem. Genet. 3, 451–455. [PubMed]
  • Goldberg, R.B., Beals, T.P., and Sanders, P.M. (1993). Anther development: Basic principles and practical applications. Plant Cell 5, 1217–1229. [PMC free article] [PubMed]
  • Higashiyama, T. (2000). Dark-field microscopy and its application to pollen tube culture. In Methods in Plant Electron Microscopy and Cytochemistry, W.V. Dashek, ed (Totowa, NJ: Humana Press), pp. 113–119.
  • Hird, D.L., Worrall, D., Hodge, R., Smartt, S., Paul, W., and Scott, R. (1993). The anther-specific protein encoded by the Brassica napus and Arabidopsis thaliana A6 gene displays similarity to beta-1,3-glucanases. Plant J. 4, 1023–1033. [PubMed]
  • Jiang, C.J., Shoji, K., Matsuki, R., Baba, A., Inagaki, N., Ban, H., Iwasaki, T., Imamoto, N., Yoneda, Y., Deng, X.W., and Yamamoto, N. (2001). Molecular cloning of a novel importin alpha homologue from rice, by which constitutive photomorphogenic 1 (COP1) nuclear localization signal (NLS)-protein is preferentially nuclear imported. J. Biol. Chem. 276, 9322–9329. [PubMed]
  • Jorgensen, R.A., Cluster, P.D., English, J., Que, Q., and Napoli, C.A. (1996). Chalcone synthase cosuppression phenotypes in petunia flowers: Comparison of sense vs. antisense constructs and single-copy vs. complex T-DNA sequences. Plant Mol. Biol. 31, 957–973. [PubMed]
  • Kobayashi, A., Sakamoto, A., Kubo, K., Rybka, Z., Kanno, Y., and Takatsuji, H. (1998). Seven zinc-finger transcription factors are expressed sequentially during the development of anthers in petunia. Plant J. 13, 571–576. [PubMed]
  • Koes, R.E., van Blokland, R., Quattrocchio, F., van Tunen, A.J., and Mol, J.N.M. (1990). Chalcone synthase promoters in Petunia are active in pigmented and unpigmented cell types. Plant Cell 2, 379–392. [PMC free article] [PubMed]
  • Koltunow, A.M., Truettner, J., Cox, K.H., Wallroth, M., and Goldberg, R.B. (1990). Different temporal and spatial gene expression patterns occur during anther development. Plant Cell 2, 1201–1224. [PMC free article] [PubMed]
  • Liu, X.C., Jones, K., and Dickinson, H.G. (1987). DNA synthesis and cytoplasmic differentiation in tapetal cells of normal and cytoplasmic male sterile lines of Petunia hybrida. Theor. Appl. Genet. 74, 846–851. [PubMed]
  • Logemann, J., Schell, J., and Willmitzer, L. (1987). Improved method for the isolation of RNA from plant tissues. Anal. Biochem. 163, 16–20. [PubMed]
  • Mariani, C., De Beuckeleer, M., Truettner, J., Leemans, J., and Goldberg, R.B. (1990). Induction of male sterility in plants by a chimeric ribonuclease gene. Nature 347, 737–741.
  • Mo, Y.-Y., Nagel, C., and Taylor, L.P. (1992). Biochemical complementation of chalcone synthase mutants defines a role for flavonols in functional pollen. Proc. Natl. Acad. Sci. USA 89, 7213–7217. [PMC free article] [PubMed]
  • Plegt, L., and Bino, R.J. (1989). β-Glucuronidase activity during development of the male gametophyte from transgenic and non-transgenic plants. Mol. Gen. Genet. 216, 321–327.
  • Raghavan, V. (1981). A transient accumulation of poly(A) containing RNA in the tapetum of Hyoscyamus niger during microsporogenesis. Dev. Biol. 81, 342–348. [PubMed]
  • Raghavan, V. (1989). mRNAs and a cloned histone gene are differentially expressed during anther and pollen development in rice (Oryza sativa L.). J. Cell Sci. 92, 217–229. [PubMed]
  • Raghavan, V. (1997a). Anther developmental biology. In Molecular Embryology of Flowering Plants, V. Raghavan, ed (Cambridge, UK: Cambridge University Press), pp. 17–60.
  • Raghavan, V. (1997b). Pollen abortion and male sterility. In Molecular Embryology of Flowering Plants, V. Raghavan, ed (Cambridge, UK: Cambridge University Press), pp. 120–151.
  • Robert, L.S., Gerster, J., Allard, S., Cass, L., and Simmonds, J. (1994). Molecular characterization of two Brassica napus genes related to oleosins which are highly expressed in the tapetum. Plant J. 6, 927–933. [PubMed]
  • Ross, J.H., and Murphy, D.J. (1996). Characterization of anther-expressed genes encoding a major class of extracellular oleosin-like proteins in the pollen coat of Brassicaceae. Plant J. 9, 625–637. [PubMed]
  • Ruiter, R.K., Van Eldik, G.J., Van Herpen, R.M., Schrauwen, J.A., and Wullems, G.J. (1997). Characterization of oleosins in the pollen coat of Brassica oleracea. Plant Cell 9, 1621–1631. [PMC free article] [PubMed]
  • Shivanna, K.R., Cresti, M., and Ciampolini, F. (1997). Pollen development and pollen-pistil interaction. In Pollen Biotechnology for Crop Production and Improvement, K.R. Shivanna and V.K. Sawhney, eds (Cambridge, UK: Cambridge University Press), pp. 15–39.
  • Smyth, D.R. (2001). Flower development. Curr. Biol. 11, R82–R84. [PubMed]
  • Steer, M.W. (1977). Differentiation of the tapetum in Avena. I. The cell surface. J. Cell Sci. 25, 125–138. [PubMed]
  • Stevens, V.A., and Murray, B.G. (1981). Studies on heteromorphic self-incompatibility systems: The cytochemistry and ultrastructure of the tapetum of Primula obconica. J. Cell Sci. 50, 419–431. [PubMed]
  • Stieglitz, H., and Stern, H. (1973). Regulation of β-1,3-glucanase activity in developing anthers of Lilium. Dev. Biol. 34, 169–173. [PubMed]
  • Takatsuji, H. (1996). A single amino acid determines the specificity for the target sequence of two zinc-finger proteins in plants. Biochem. Biophys. Res. Commun. 224, 219–223. [PubMed]
  • Takatsuji, H. (1999). Zinc-finger proteins: The classical zinc finger emerges in contemporary plant science. Plant Mol. Biol. 39, 1073–1078. [PubMed]
  • Takatsuji, H., and Matsumoto, T. (1996). Target-sequence recognition by separate-type Cys2/His2 zinc finger proteins in plants. J. Biol. Chem. 271, 23368–23373. [PubMed]
  • Theiben, G. (2001). Development of floral organ identity genes: Stories from the MADS house. Curr. Opin. Plant Biol. 4, 75–85. [PubMed]
  • van der Meer, I.M., Stam, M.E., van Tunen, A.J., Mol, J.N.M., and Stuitje, A.R. (1992). Antisense inhibition of flavonoid biosynthesis in petunia anthers results in male sterility. Plant Cell 4, 253–262. [PMC free article] [PubMed]
  • van Engelen, F.A., Molthoff, J.W., Conner, A.J., Nap, J.P., Pereira, A., and Stiekema, W.J. (1995). pBINPLUS: An improved plant transformation vector based on pBIN19. Transgenic Res. 4, 288–290. [PubMed]
  • van Tunen, A.J., Mur, L.A., Brouns, G.S., Rienstra, J.D., Koes, R.E., and Mol, J.N. (1990). Pollen- and anther-specific chi promoters from petunia: Tandem promoter regulation of the chiA gene. Plant Cell 2, 393–401. [PMC free article] [PubMed]
  • Vaucheret, H., Béclin, C., Elmayan, T., Feuerbach, F., Godon, C., Morel, J.-B., Mourrain, P., Palauqui, J.-C., and Vernhettes, S. (1998). Transgene-induced gene silencing in plants. Plant J. 16, 651–659. [PubMed]
  • Wilson, Z.A., Morroll, S.M., Dawson, J., Swarup, R., and Tighe, P.J. (2001). The Arabidopsis MALESTERILITY1 (MS1) gene is a transcriptional regulator of male gametogenesis, with homology to the PHD-finger family of transcription factors. Plant J. 28, 27–39. [PubMed]
  • Ylstra, B., Busscher, J., Franken, J., Hollman, P.C.H., Mol, J.N.M., and van Tunen, A.J. (1994). Flavonols and fertilization in Petunia hybrida: Localization and mode of action during pollen tube growth. Plant J. 6, 201–212.

Articles from The Plant Cell are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...