Download Gene Expression Profiles of Cold

Plant Cell Physiol. 45(10): 1519–1528 (2004)
JSPP © 2004
Gene Expression Profiles of Cold-stored and Fresh Pollen to Investigate
Pollen Germination and Growth
Min-Long Wang 1, 2, 4, Chia-Mei Hsu 1, 4, Liang-Chi Chang 1, Co-Shine Wang 2, Ting-Ho Su 1, 2, Yih-Jong
John Huang 1, Liwen Jiang 3 and Guang-Yuh Jauh 1, 5
1
Institute of Botany, Academia Sinica, Nankang, Taipei 11529, Taiwan, R.O.C.
Graduate Institute of Biotechnology, National Chung Hsing University, Taichung 40227, Taiwan, R.O.C.
3
Department of Biology, Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China
2
;
In lily (Lilium longiflorum cv. Avita) pollen cold-stored
(–20°C) for 2 months, typical in vitro germination/growth
was delayed by about 1 h compared with fresh pollen. We
hypothesized that some proteins and mRNAs stored in
mature pollen were degraded during storage periods and
that re-synthesis of them was essential to resume normal
germination and growth. Cold-stored and fresh pollen
grains were used to investigate the regulatory mechanism of
pollen germination and tube growth in terms of both total
protein profile and gene expression. Total protein profiles of
cold-stored pollen differed qualitatively and quantitatively
from fresh pollen. Actinomycin D significantly inhibited
both germination and tube growth of cold-stored pollen and
later tube growth of fresh pollen but had no effect on fresh
pollen germination and early tube growth. Suppression
subtractive hybridization screening revealed 99 cDNAs
enriched in fresh mature pollen, and 22 were selected for
further characterization. Most of these 22 cDNAs gradually disappeared during cold storage, but full recovery was
achieved by incubating the cold-stored pollen in culture
medium for 2 h. Because of different sensitivities to cold
storage and actinomycin D, the transcripts were divided
into three groups according to their possible roles in pollen
germination and tube growth. Several cDNAs encoding
novel proteins showed pollen-specific expression patterns
and may participate in drought tolerance (an Na+/H+ antiporter), endomembrane trafficking (DnaJ), division of the
generative cell (Sgt1), pollen wall precursor uptake from
stylar exudate (an Na+/myoinositol symporter) and chemotropism of the pollen tube (peptide transporter) during pollination.
Keywords: Actinomycin D — Gene expression — Lilium
longiflorum — Low temperature — Pollen germination — Pollen tube growth.
Abbreviations: ActD, actinomycin D; 2D-PAGE, two-dimensional polyacrylamide gel electrophoresis; RT-PCR, reverse transcription-polymerase chain reaction.
4
5
Introduction
Pollen development takes place within the anther and
provides a good system for investigating the regulation of
several fundamental cellular events such as cell division, cell
fate determination, gene regulation, cell-cell communication
and cellular differentiation during pollination. Mature pollen
released from the anther is dehydrated and metabolically
quiescent; after reaching a receptive stigma or in vitro cultured medium, the pollen rehydrates and germinates quickly
(Mascarenhas 1993, McCormick 1993, Lord and Russell
2002). Pollen germination and tube growth are very rapid
events, and the rates of pollen tube growth are extremely high.
To support this unusually fast growth, large amounts of protein, rRNA, mRNA, bioactive small molecules and reserved
material are stored in mature pollen grains (Mascarenhas 1993,
Taylor and Hepler 1997, Twell 2002).
Analyses of the expression patterns of a number of pollenexpressed and/or pollen-specific mRNAs have shown that at
least two broad classes of genes, early and late, are involved
(Mascarenhas 1990). The early genes are first expressed in
microspores, but their expression decreases significantly before
pollen maturation, whereas the late genes are not expressed
until after pollen mitosis I and their transcripts accumulate until
pollen maturation. It is believed that the pre-synthesized
mRNAs and proteins are mainly products of the late genes of
the vegetative cell. The mRNAs in mature pollen grains of Tradescantia (Willing and Mascarenhas 1984) and maize (Willing
et al. 1988) are the products of about 20,000–24,000 different
genes, and 10% of these appear to be pollen-specific transcripts. Only a small fraction of these genes, approximately
150 genes from 28 species, have been well characterized individually, and most were obtained from the developing pollen
grain (Twell 2002). Guyon et al. (2000) showed the rapid and
synchronous in vitro germination response of conditional-malefertility pollen to exogenous kaempferol, and isolated several
novel germinating pollen cDNAs during the earliest moments
of pollen germination in petunia. With the use of novel experimental strategies and the increasing availability of new genetic
tools, global gene expression profiles have been explored in
Arabidopsis pollen (Becker et al. 2003, Honys and Twell 2003,
These authors contributed equally to this work.
Corresponding author: E-mail, [email protected]; Fax, +886-2-27827954.
1519
1520
Gene expression profile in germinating lily pollen
Lee and Lee 2003), maize sperm cells (Engel et al. 2003) and
tobacco sperm cells (Xu et al. 2002). These studies greatly
broaden our current knowledge of transcripts present in the
mature pollen grain. However, further gene expression and
functional analyses of individual transcripts (da Costa-Nunes
and Grossniklaus 2003), as well as the establishment and characterization of the gene expression profiles in germinating pollen and/or pollen tubes, can expand our comprehension of the
complex events during pollination (Mascarenhas 1993).
In this study, Lilium longiflorum cv. Avita pollen stored at
–20°C for 2 months showed retarded (by 1 h) germination and
growth rate compared with fresh pollen. Some pre-synthesized
proteins and mRNAs stored in mature pollen might have
degraded gradually during cold storage and had to be resynthesized to carry on pollination. To test this hypothesis,
cold-stored and fresh pollen grains were analyzed for total protein and gene expression profiles during pollen germination and
tube growth. The total protein profile of cold-stored pollen was
qualitatively and quantitatively decreased compared with that
of fresh pollen. After suppression subtractive hybridization
screening, thousands of clones were obtained, 99 cDNAs were
enriched in the fresh mature pollen, and expression in both
fresh and stored pollen grains and in pollen development was
further characterized in 22. The expression of most of the 22
gradually disappeared during cold storage but reappeared on
incubation of the stored pollen grains in germination medium
for 2 h. Actinomycin D (ActD) significantly inhibited both germination and tube growth of cold-stored pollen and later tube
growth of fresh pollen but had no effect on fresh pollen germination and early tube growth. Because of different sensitivities
to cold storage and ActD, these 22 transcripts were divided into
three groups according to their possible roles in pollen germination and tube growth. Several novel cDNAs (i.e. the first to
be identified in pollen) were specifically expressed in the
mature lily pollen grains and germinating pollen tubes. Their
possible functions during pollination are discussed.
Results
Cold storage significantly altered lily pollen germination,
growth and total protein composition.
Lily pollen freshly collected or stored at –20°C for 1 and 2
months was germinated in vitro to evaluate the effect of cold
storage on germination and growth rates (Fig. 1). Compared
with fresh pollen, cold-stored pollen was delayed in germination (Fig. 1A) and initially had a slower tube growth rate (Fig.
1B). These consequences were correlated with the length of
storage. However, stored pollen grains were still capable of
pollination, as shown by the in vivo pollen tubes isolated from
pollinated 24-h styles (Fig. 1C), and fertilization, as evidenced
by ovule development (data not shown). To examine the possible causes of growth retardation at the protein level, the total
protein profiles of fresh pollen and pollen cold-stored for 2
Fig. 1 Effects of cold storage on lily pollen germination and pollen
tube growth. Lily pollen grains freshly collected or stored at –20°C for
1 or 2 months were cultured in vitro. Pollen germination (A) and tube
growth (B) rates were measured. Compared with freshly collected pollen grains, cold-stored pollen showed greatly decreased germination
and tube growth rates, which increased with longer storage time. (C)
Pollen grains stored for 2 months at –20°C were checked for viability
by in vivo pollination. After 24 h, the pollen tubes were pulled out of
the style. The arrowhead and arrow indicate the stigma and aggregated
in vivo pollen tubes, respectively. All results from three independent
experiments were similar. Bar = 1 cm.
months were compared on 2D-PAGE separation (Fig. 2). The
profiles of fresh (Fig. 2A) and stored (Fig. 2B) pollen differed
quantitatively and qualitatively. Many proteins were no longer
detectable (solid arrows in Fig. 2A) or had greatly decreased
levels (open arrows in Fig. 2A) in pollen cold-stored for 2
months. In addition, several new proteins and/or the products of
the degraded proteins were present in the pollen (open arrowheads in Fig. 2B).
Gene expression profile in germinating lily pollen
Fig. 2 Protein degradation in cold-stored lily pollen grains. A total of
200 µg of total proteins extracted from freshly collected lily pollen
grains (A) and those stored at –20°C for 2 months (B) were separated
by IEF gel electrophoresis (pH 4.2–7.5) and 10% SDS-PAGE. Quantitative (open arrow) and qualitative (solid arrow) differences were
observed. Several new proteins (open arrowhead) were also found in
cold-stored pollen. The open boxes are the coordinates for comparison
of the corresponding protein spots on the gels. M is the middle range
protein molecular weight marker (kDa).
De novo synthesis of essential mRNAs/proteins allows resumption of normal germination and tube growth in cold-stored
pollen.
The finding that cold-stored pollen needs a longer incubation time (about 1 h) than does fresh pollen to resume normal
pollen germination and tube growth suggests that de novo synthesis of mRNA/protein, perhaps degraded during storage, may
occur in cold-stored pollen during incubation. To test this
hypothesis, fresh and cold-stored (2 months) pollen grains were
germinated in medium with or without ActD and with or without cycloheximide, which are effective transcriptional and
translational inhibitors, respectively. ActD treatment had no
effect on fresh pollen germination (Fig. 3A) and early tube
growth (the first 3 h in Fig. 3B), but later tube growth was
greatly inhibited (Fig. 3B). ActD had a slight effect on germination of cold-stored pollen (Fig. 3A), but pollen tube growth
was almost completely inhibited (Fig. 3B). Another efficient
1521
Fig. 3 Effects of ActD and cycloheximide on the restoration of germination competence of the cold-stored pollen grains. A total of
100 µg ml–1 of ActD or 100 µM of cycloheximide (CHX) was added
to germination medium to determine the effect on pollen germination
(A) and tube growth (B) rates of freshly collected and cold-stored (2
months) lily pollen grains. Cycloheximide completely inhibited both
pollen germination and growth. ActD had no effect on fresh pollen
germination (A) and early tube growth (B), but later tube growth was
greatly inhibited (B). ActD had a stronger effect on the germination of
cold-stored pollen (A), and pollen tube growth was almost completely
inhibited (B). Results from three independent experiments were similar.
transcriptional inhibitor, cordecypin (100 µg ml–1), gave a similar effect (data not shown). Cycloheximide efficiently inhibited
both germination and tube growth of both fresh and cold-stored
pollen (Fig. 3). The differences in germination rates of fresh
pollen shown in Fig. 1, 3 are a consequence of different samples of pollen, but results from three independent experiments
with either the same or different batches of collected pollen
grains were similar. All these results supported the hypothesis,
and suppression subtractive hybridization screening was conducted to clone the expressed and functional transcripts persisting in mature pollen grains.
1522
Table 1
Gene expression profile in germinating lily pollen
Identity of selected cDNAs obtained from suppression subtractive hybridization screening
Clone
Amino acid
identity (%)
Amino acid
Homolog in other organism
similarity (%) (with Best Blast P)
69 (118/169) a
66 (18/27)
70 (129/183)
56 (53/94)
48 (46/94)
67 (105/156)
82 (140/169)
77 (21/27)
82 (151/183)
71 (68/94)
62 (60/94)
74 (117/156)
Arabidopsis thaliana (AP000603)
Medicago truncatula (AJ249611)
A. thaliana (AP000735)
Solanum tuberosum (AJ133765)
A. thaliana (AC005825)
A. thaliana (AL162691)
27 (32/115)
64 (90/139)
46 (54/115)
74 (105/139)
Peptide transporter
Plasma membrane H+-ATPase
Monosaccharide transporter
Na+/H+ antiporter
54 (92/168)
77 (35/45)
75 (90/119)
39 (101/254)
68 (117/168)
88 (40/45)
86 (104/119)
54 (141/254)
A. thaliana (AL050352)
Mesembryanthemum crystallinum This study
(AF280432)
A. thaliana (AL161555)
This study
Kosteletzkya virginica (AF029258)
A. thaliana (AL161555)
b
A. thaliana (AC007063)
This study
RNA binding protein
Acid phosphatase
DnaJ
Sgt1
COP8
Carbonic anhydrase
NTP303 precursor
26S proteasome regulatory subunit
36 (22/60)
59 (65/109)
53 (68/126)
52 (62/119)
90 (147/162)
46 (36/77)
81 (63/77)
85 (91/106)
51 (31/60)
73 (81/109)
72 (93/126)
67 (81/119)
95 (156/162)
65 (51/77)
92 (72/77)
94 (101/106)
A. thaliana (AC011437)
A. thaliana (AJ243527)
A. thaliana (U95973)
Oryza sativa (AF192467)
A. thaliana (AF176089)
A. thaliana (U73462)
Nicotiana tabacum (P29162)
A. thaliana (AC073395)
Unknown protein
Hypothetical protein
65 (122/186)
54 (73/135)
79 (149/186)
65 (89/135)
A. thaliana (AC018907)
A. thaliana (AL035602)
Name
Cell wall/plasma membrane protein
LLPB7
Pectin esterase
LLPB9
Pectin methylesterase
LLPC8
Extensin
LLPD9
Invertase
LLPD16 β-Galactosidase
LLPE21 Cellulase
Special transporter
LLPC12 Ca2+-transporting ATPase
LLPD5
Na+/myo-inositol symporter
LLPD11
LLPD21
LLPD22
LLPE49
Others
LLPB3
LLPC1
LLPC3
LLPC4
LLPC14
LLPD2
LLPD13
LLPD27
Unknown
LLPB8
LLPD3
a
b
Pollen
specificity
b
This study
This study
b
This study
b
This study
Parentheses indicate the number of amino acids for each clone used for the comparison.
Pollen specificity was obtained from the Arabidopsis pollen transcriptome (Honys and Twell 2003).
Cloning and characterization of the transcripts degraded in
stored pollen by suppression subtractive hybridization screening.
The cDNAs from cold-stored (2 months) pollen (driver)
were used to subtract those from the fresh pollen (tester). Thousands of clones were obtained after nested PCR amplification,
and 99 cDNAs greatly enriched in the fresh pollen (supplementary Table), as confirmed by reverse Northern blotting (data not
shown), were sequenced and BLAST analyzed. The cDNAs
were divided into five categories according to the nature of
deduced amino acid sequences: 17% are involved in cell wall
metabolism, 12% are various transporters, 6% participate in
signaling pathways, 34% are components of other cellular
processes and 30% are hypothetical proteins or clones with no
significant similarity. Twenty-two cDNAs (Table 1) with higher
expression levels in fresh than in cold-stored pollen, as examined by use of reverse Northern blotting (data not shown), were
further investigated for expression patterns using Northern blotting (Fig. 4). For the cDNAs that encoded wall and/or plasma
membrane proteins, most of the transcripts, except for that for
LLPB7, were degraded after cold storage for 1 month (Fig. 4A,
lane 2) and 2 months (Fig. 4A, lane 3), as compared with fresh
pollen (Fig. 4A, lane 1). Those degraded mRNAs, however,
reappeared and accumulated to normal or even higher levels in
some cases after 2 h in vitro incubation (Fig. 4A, lane 4). These
results suggest that the cold-stored pollen still has the ability,
but needs a longer culture time, to re-synthesize the mRNAs
gradually degraded during cold storage. Similar results were
obtained for most of the cDNAs that encode several transporters (Fig. 4B), as well as novel ones (Fig. 4C), or proteins with
unknown function (Fig. 4D). Some transcripts, including pectin esterase (LLPB7), acid phosphatase (LLPC1) and carbonic
anhydrase (LLPD2), showed high stability even after 2 months’
storage. Also, ActD produced different responses during the 2h incubation (Fig. 4, lane 4 vs. lane 5): (i) a significant inhibitory effect on the message levels of LLPD21 (Fig. 4B), LLPB3,
LLPC14 (Fig. 4C) and LLPB8 (Fig. 4D); (ii) a moderate effect
on the levels of LLPB9, LLPC8, LLPE21 (Fig. 4A), LLPD5,
LLPC12 (Fig. 4B) and LLPC3 (Fig. 4C); (iii) no effect on the
Gene expression profile in germinating lily pollen
Fig. 4 Low-temperature storage results in significant decreases in the
expression of several cell wall/plasma membrane, transporter and
novel genes in lily pollen. Twenty micrograms of total RNA from fresh
pollen grains (lane 1) and pollen grains stored at –20°C for 1 month
(lane 2), 2 months (lane 3) or 2 months, then germinated in vitro without (lane 4) or with (lane 5) actinomycin D for 2 h were analyzed by
Northern blotting. (A) Wall/plasma membrane cDNAs used as probes:
LLPB7 (pectin esterase), LLPB9 (pectin methylesterase), LLPC8
(extensin-like protein), LLPD9 (invertase), LLPD16 (β-galactosidase)
and LLPE21 (cellulase). (B) Transporter cDNAs used as probes:
LLPC12 (Ca2+-transporting ATPase), LLPD5 (Na+/myo-inositol symporter), LLPD11 (peptide transporter), LLPD21 (plasma membrane
H+-ATPase), LLPD22 (monosaccharide transporter) and LLPE49
(Na+/H+ antiporter). (C) Novel cDNAs used as probes: LLPB3 (RNA
binding protein), LLPC1 (acid phosphatase type 5), LLPC3 (DnaJ),
LLPC4 (Sgt1), LLPC14 (COP8), LLPD2 (carbonic anhydrase)
LLPD13 (1-ascorbate oxidase) and LLPD27 (26S proteasome regulatory subunit S12). (D) Two cDNAs of unknown function were used as
probes. EtBr-stained rRNA was used as an internal standard to monitor equal loading of total RNA.
levels of LLPB7, LLPD16 (Fig. 4A), LLPD11 (Fig. 4B),
LLPC1, LLPC4, LLPD2, LLPD13, LLPD27 (Fig. 4C) and
LLPD3 (Fig. 4D); and (iv) a slightly enhanced effect on the
levels of LLPD9 (Fig. 4A), LLPD22, LLPE49 (Fig. 4B). The
variable expression patterns of the characterized transcripts
were probably caused by the different strengths of their promoters or mRNA stability and would be interesting for further
study.
Since pollen tube growth of cold-stored pollen was almost
completely inhibited by ActD (Fig. 3B), the cDNA expression
patterns sensitive to ActD as shown in Fig. 4 may play an
1523
Fig. 5 Expression of several genes in low-temperature-stored pollen
is steadily inhibited by transcription inhibitor during pollen germination. mRNAs were purified from low-temperature-stored pollens cultured in the media with or without (control) 100 µg ml–1 cordecypin
for 0, 3, 6 or 9 h and were used as templates for corresponding cDNA
synthesis. Primer pairs specific to each cDNA were used for RT-PCR
amplification to determine cDNA expression. Actin was the internal
control.
important role in pollen tube growth. To test this idea, 2-month
low-temperature-stored pollen was cultured in media with or
without another efficient transcription inhibitor, cordecypin, for
0, 3, 6 and 9 h and was used for total RNA extraction, mRNA
purification and corresponding cDNA synthesis. Primer pairs
designed specifically for each cDNA (listed in Table 2) were
used for semi-quantitative reverse transcriptase-polymerase
chain reaction (RT-PCR) amplification to determine their
expression profiles after different incubation intervals. LLPD21
cDNA was too short for proper RT-PCR amplification and was
not included in this experiment. For the pollen cultured in normal medium, all the cDNAs, including the ActD-insensitive
LLPD27 and LLPC1 cDNAs shown in Fig. 4, started to accumulate after 3 h of incubation and continued to 9 h incubation
(control in Fig. 5). However, this increased expression was significantly blocked by cordycepin, as evidenced by the steady
expression levels of these cDNAs during all incubation periods
(cordycepin in Fig. 5). In addition, the RT-PCR expression
patterns correlated well with the results of Northern blotting
(compare 0 and 3 h in Fig. 5 with lanes 1 and 2 in Fig. 4,
respectively). All these results suggest that these ActD- and
cordecypin-sensitive transcripts are essential for normal pollen
tube growth.
1524
Gene expression profile in germinating lily pollen
Table 2 Primers of chosen cDNAs and length of corresponding amplified products after RT-PCR analysis
Clone
Primer sequence
Product size (bp)
LLPB3 (RNA binding protein)
Sense: 5′-TAGCATGGCACATATGGACCACGG-3′
Antisense: 5′-CGGAAGTGCCTTGCTATCCATTTGAG-3′
LLPB8 (unknown protein)
Sense: 5′-CTGCATAAAGGTCGAAGATGTGGAGCC-3′
Antisense: 5′-CGAGGATAAGGTTGGATACGTC-3′
LLPB9 (pectin methylesterase)
Sense: 5′-GGGCAGGTACGGCAACCGTGGC-3′
Antisense: 5′-TTCATAGCCTCGGCCCTGCTG-3′
LLPC3 (DnaJ)
Sense: 5′-GATCACATTCGAAGGGATGGGG-3′
Antisense: 5′-GCTTAGCCTCTTAGGGAAGACAACAC-3′
LLPC4 (Sgt1)
Sense: 5′-GCGTCGCAGG ATTTTGAAAAGCTG-3′
Antisense: 5′-GAGAGATTGATGCCCACATAG-3′
LLPC1 (acid phosphatase)
Sense: 5′-GCGTCGCAGG ATTTTGAAAAGCTG-3′
Antisense: 5′-GAGAGATTGATGCCCACATAG-3′
LLPC8 (extensin)
Sense: 5′-CCGACGAAGCGGTTGTTGCTCA-3′
Antisense: 5′-TGCCCTATCCGATGCCGAAGTC-3′
LLPC12 (Ca2+-transporting ATPase)
Sense: 5′-GGATGGATTAGCTTCTCCCGAA-3′
Antisense: 5′-GGAAAGCCTTCGGAGAGTAGTTTG-3′
LLPC14 (COP8)
Sense: 5′-GCTCGTCTCTATCTTGAGGATG-3′
Antisense: 5′-GTAGCAAGAACCCGAGAGCGTTGTG-3′
LLPD5 (Na+/myo-inositol symporter)
Sense: 5′-ACGAATGGTCTGTTGATCACC-3′
Antisense: 5′-ACATGACTGTGTTGATACCAAC-3′
LLPD11 (peptide transporter)
Sense: 5′-GGCTATCAGCCATCAATTGC-3′
Antisense: 5′-CTCCATGGATTGGTTTTCTCTG-3′
LLPD27 (26S proteasome regulatory subunit S12) Sense: 5′-CTCCCGAAGTTTGGGACCAGTGCTAT-3′
Antisense: 5′-GCACCCTCTCGTTCTGCTTTCC-3′
LLPE49 (Na+/H+ antiporter)
Sense: 5′-GGGTGGGATGCCGCTCGG-3′
Antisense: 5′-CCTCGTCCAGTCGTGCTC-3′
actin
Sense: 5′-GGCY*GGHTTYGCNGGNGAYGATGC-3′
Antisense: 5′CCKGATRTCVACRTCACACTTCAT-3′
358
415
93
334
261
327
459
371
311
415
488
320
1,000
761
*IUB codes: H = A, C or T; K = G or T; N = any base; R = A or G; V = A, C or G; Y = A or G.
Spatial and temporal expression of several novel cDNAs in lily
pollen
Several cDNAs coded for novel proteins, including DnaJ
(LLPC3), Sgt1 (LLPC4), Na+/myo-inositol symporter (LLPD5),
peptide transporter (LLPD11), Na+/H+ antiporter (LLPE49) and
the 26S proteasome regulatory subunit S12 (LLPD27). These
cDNAs were chosen to explore expression profiles during pollen development. Their spatial and temporal expression patterns were examined by semi-quantitative RT-PCR of the
cDNAs obtained from total RNAs of lily roots (R), leaves (L),
pistils (Pi), pollen grains (Po and 0 h), in vivo- (24 h) and in
vitro- (12- and –24-h) grown pollen tubes with specific primer
sets for each cDNA (Table 2). As shown in Fig. 6A, except for
the cDNAs of actin (the internal total RNA-loading quantitative control) and LLPD27 (26S proteasome regulatory subunit
S12), all cDNAs were specifically expressed in the pollen
grain, and in the in vitro- and in vivo-grown pollen tube. The
detailed temporal and spatial expression patterns of LLDNAJ
and LLSGT1 transcripts were further investigated by Northern
blot analyses. Both transcripts were pollen- (Po) specific, and
no signal was detected in total RNA extracted from leaves (L),
roots (R), tepals (T), filaments (F) or pistils (Pi, Fig. 6B, upper
panel). The transcripts also belong to the late pollen-expressed
genes with specific expression of LLDNAJ in anthers of 130mm buds (Fig. 6B, lower panel, lane 6); and for both LLDNAJ
and LLSGT1 in mature anthers (Fig. 6B, lower panel, lane 7),
mature pollen (Fig. 6B, lower panel, lane 8), and 3-h in vitrogerminated pollen tubes (Fig. 6B, lower panel, lane 9).
Discussion
A distinctive cold-storage-induced, artificial ‘depleted pollen’ system was used to systematically characterize the general
profiles of the genes/proteins participating in lily pollen germination and tube growth. After lily pollen had been cold-stored
for 2 months, typical germination and growth was significantly
retarded and accompanied by quantitative and qualitative degradation of numerous pre-synthesized proteins and mRNAs
(Fig. 2, 4). Thousands of cDNAs were obtained by suppression
subtractive hybridization screening, and 99 cDNA clones
Gene expression profile in germinating lily pollen
enriched in fresh pollen (supplementary Table) were confirmed
by reverse Northern blotting. Twenty-two of these with higher
expression profiles in fresh pollen were chosen to examine
their expression profile in fresh and cold-stored pollen grains.
Northern blotting results revealed that most of these 19 cDNAs
gradually disappeared during –20°C storage and that full recovery of these degraded transcripts, as well as resumption of germination and growth competence of the growth-retarded pollen
grains, were achieved by immersion in culture medium for 2 h
(Fig. 1, 4). Several novel cDNA clones showed pollen-grain-
1525
and pollen-tube-specific expression patterns and were the products of late genes.
Thus far, most pollen-expressed genes have been obtained
from studies of developing microspore and/or mature pollen,
and only a few genes expressed in the germinating pollen
and/or pollen tube have been well characterized (reviewed in
Mascarenhas 1993, McCormick 1993, Twell 2002). The fact
that cold-stored pollen germinated and grew after immersion in
culture medium for an extended time (1 h) suggests that de
novo synthesis of degraded mRNAs and proteins is essential to
resume normal pollen germination and tube growth of coldstored pollen. This suggestion is supported by the significant
inhibitory effect of ActD on cold-stored pollen germination and
tube growth. For fresh pollen, ActD had no effect on germination and early tube growth, but it significantly inhibited
later tube growth, as has been reported for other species
(Mascarenhas 1975, Mascarenhas 1993). Cycloheximide prevented pollen germination and tube growth in both fresh and
cold-stored pollen (Hoekstra and Bruinsma 1979). All these
results imply that the persisting mRNAs in the mature pollen
grain are sufficient, but the translation of these mRNAs is
obligatory for pollen germination and early tube growth. The
transcripts can be divided into three groups according to degree
of response to ActD and cold storage, which may have a role in
different developmental phases, such as pollen germination/
early pollen tube growth, later tube growth or both events
(Fig. 3–5). The first group of transcripts, LLPD9, LLPD16,
LLPD11, LLPD22, LLPE49, LLPD13, LLPD27 and LLPD3, is
sensitive to cold storage but not ActD and involved in pollen
germination and early pollen tube growth. The second group,
LLPB9, LLPC8, LLPC12, LLPD5, LLPD21, LLPB3, LLPC3,
LLPC14 and LLPB8, is sensitive to both cold storage and ActD
(with varying degrees of sensitivity) and may participate in
later pollen tube growth. The relationship between the expression of these transcripts and pollen tube growth is supported by
Fig. 6 Pollen-specific expression patterns of several novel transcripts.
(A) mRNAs isolated from lily root (R), leaves (L), pistils (Pi), pollen grains (P), in vivo-grown (24 h) pollen tubes (Vo) and in vitrogerminated (0, 12 and 24 h) pollen tubes were used as templates for
corresponding cDNA synthesis. Actin was the internal control. Primer
pairs specific to LLPC3 (DnaJ), LLPC4 (Sgt1), LLPD5 (Na+/myoinositol symporter), LLPD11 (peptide transporter), LLPE49 (Na+/H+
antiporter) and LLPD27 (26S proteasome regulatory subunit S12)
cDNAs were used for RT-PCR amplification to determine the cDNA
expression. (B) Twenty micrograms of DNase I-treated total RNA
were isolated from lily leaves (L), roots (R), tepals (T), filaments (F),
pistils (P), pollen (Po) and anthers/stamens of various-sized floral
buds: 15 mm (1), 35 mm (2), 65 mm (3), 90 mm (4), 110 mm (5) and
130 mm (6); mature anther (7), mature pollen (8) and in vitrogerminated 3-h pollen tubes (9). Total RNA was denatured, separated
on formaldehyde-agarose gels, transferred onto a nylon membrane and
hybridized with DIG-labeled LLDNAJ or LLSGT1 cDNA inserts.
Almost equal amounts of total RNA were loaded in each lane, as evidenced by ethidium bromide (EtBr) staining of the gel.
1526
Gene expression profile in germinating lily pollen
the inhibition of cordycepin on both expression profiles and
continuous pollen tube growth. The last group, LLPB7,
LLPC1, LLPD2, is insensitive to both ActD and cold storage
and may have a role in all phases of pollen germination and
pollen tube growth. Nevertheless, we cannot rule out other
essential, as yet to be identified, transcripts involved in these
events. Another interesting phenomenon is the varying stability of the transcripts characterized here under cold stress. For
example, LLPE21 (cellulase), LLPD21 (plasma membrane H+ATPase), LLPE49 (Na+/H+ antiporter), LLPB 3 (RNA binding
protein), LLPC14 (COP8), LLPD27 (26S proteasome regulatory subunit S12) and LLPB8 (an unknown protein) were more
sensitive to cold storage than the others, but some transcripts
were extremely stable, with little or no degradation after long
periods of storage, including LLPB7 (pectin esterase), LLPC1
(acid phosphatase) and LLPD2 (carbonic anhydrase) (Fig. 4).
The results regarding these mRNAs, with quantitatively different degradation patterns, are consistent with the idea of differential RNA decay mechanisms in pollen as proposed by Ylstra
and McCormick (1999).
The precise functions of the novel cDNAs listed in Table 1
remain to be experimentally revealed, but their deduced protein sequences give us some clues about the fundamental nature
of pollination. For example, LLPE21 cDNA, encoding a cellulase, may be responsible for penetration of the pollen tube into
the stigmatic papillary cells during pollination. For the first
time, we report several transcripts coding for novel proteins
such as DnaJ, Sgt1, Na+/myo-inositol symporter, peptide transporter and the 26S proteasome regulatory subunit S12, in pollen, and their expression patterns are pollen specific, except for
the 26S proteasome regulatory subunit S12 (Fig. 6A). The 26S
proteasome is involved in many cellular protein degradation
processes, so it is not surprising to find its transcripts present in
all organs tested. The pollen-specific pattern of other transcripts suggests that their encoded proteins play important roles
during pollination. It is known that style-secreted exudates are
enriched with the pectin precursor myo-inositol (Rosen 1971),
which is likely taken up into in vivo pollen tubes by the plasma
membrane Na+/myo-inositol symporter for pollen wall synthesis during pollination. Lily stigma exudate contains γ-amino
butyric acid in a bound form of N-acylated or peptide-like
derivatives (Rosen 1971), which may be taken up into in vivo
pollen tubes as by the peptide transporter (LLPD11) to support
pollen tube growth as found in Arabidopsis (Palanivelu et al.
2003). The expression pattern of the lily Na+/H+ antiporter
(LLPE49) transcript may reflect its function under stress conditions to protect the dehydrated mature pollen grain (Chauhan et
al. 2000). The DnaJ protein, a member of the Hsp40 family,
functions as a co-factor of the heat-shock protein 70 (Hsp70)
chaperone machine in facilitation of protein folding (Kelley
1998, Lemmon 2001). SGT1 is a component of SCF ubiquitin
ligase and participates in regulation of the cell cycle in yeast
(Kitagawa et al. 1999) and meiosis in the Arabidopsis anther
(Yang et al. 1999). Lily pollen belongs to the two-celled pollen
type, and LLSGT1 may participate in the second mitosis of the
generative cell to produce two sperm cells after pollen tube
emergence.
By taking advantage of the availability of the complete
genomic sequence of Arabidopsis and contemporary genetic
tools, two different strategies have been conducted recently to
investigate the genome-wide gene expression of Arabidopsis
pollen: Affymetrix ATH8K Genechip (Becker et al. 2003,
Honys and Twell 2003) and serial analysis of gene expression
(SAGE; Lee and Lee 2003). In addition, sequence analysis of a
cDNA library constructed from maize sperm cells reveals
sperm cell-specific transcripts that may be involved in gamete
interaction during double fertilization (Engel et al. 2003).
These studies significantly increase our current knowledge of
the gene expression profile of a pollen grain; however, general
and systematic gene and/or protein expression patterns during
pollen germination and tube growth are lacking. In addition,
the fact that ATH8K GeneChip contains only about 30% of the
estimated genes in the Arabidopsis genome suggests that more
pollen-expressed and/or -specific genes remain to be identified
(da Costa-Nunes and Grossniklaus 2003). In this study, ActD
significantly inhibited continuous tube growth of fresh pollen,
which suggests that some genes essential for effective later tube
growth may not be expressed until the later stages of pollination. Pollination involves very complicated cell-cell interactions, and the characters of pollen tubes in vivo differ
significantly from those in vitro (Lord 2000). Pollen grown in
vitro does not completely mimic that grown in vivo, and even
with highly optimized germination medium, in vitro tubes
reach only 30–40% of in vivo lengths (Read et al. 1993). Some
transcripts might be up-regulated or induced in later phases of
pollination as the in vivo pollen tube travels in the style and
during double fertilization. Isolation and characterization of
these pollen tube-specific or pollination-induced transcripts
will allow us to explore the mechanisms of the complex cellcell interactions during pollination and double fertilization in
flowering plants.
Materials and Methods
Plant materials
Easter lily (Lilium longiflorum Thunb. cv. Avita) bulbs obtained
from a local farm (Foreport Enterprises Co., Ltd, Taipei, Taiwan) were
planted in a greenhouse under ambient conditions at Academia Sinica,
Taiwan. Mature pollen grains from anthers were collected 1 day after
anthesis and dried on a bench for 2 d. Dried pollen grains were used
directly (fresh pollen) or stored at –20°C for later use (cold-stored pollen). To measure germination and growth rates, pollen was placed in
germination medium (1.27 mM CaCl2, 0.162 mM H3BO3, 0.99 mM
KNO3, 290 mM sucrose, pH 5.2) with or without 100 µg ml–1 ActD,
cordecypin or 100 µM cycloheximide for various times. Tube length
and germination rate were measured under an Olympus BH2 microscope. For in vivo pollination studies, ‘Snow Queen’ styles were pollinated 2 d after anthesis with Avita pollen stored at –20°C for 2 months,
and in vivo pollen tubes were collected as described previously (Jauh
and Lord 1995, Jauh and Lord 1996).
Gene expression profile in germinating lily pollen
Pollen total protein extraction and 2D-PAGE
Total protein extraction, 2D-PAGE and Western blotting were
performed according to Wang et al. (1992). The phenol-extracted total
protein was precipitated, and the pellet was washed twice with cold
acetone, then vacuum-dried and dissolved in solubilization buffer
[9.5 M urea, 5 mM K2CO3, 2% (w/v) Triton X-100, 500 µg ml–1 Llysine, 0.5% DTT, 1.6% ampholines, pH 5–7, and 0.4% ampholines,
pH 3–10] for 2D-PAGE or in SDS sample buffer for SDS-PAGE and
Western blotting.
mRNA purification, suppression subtractive hybridization and Northern blotting
Total RNA was extracted by use of the Ultraspec™ RNA isolation system according to the manufacturer’s manual (Biotecx Laboratories, Houston, TX, U.S.A.). To avoid DNA contamination, total RNA
was treated with of the Message Clean™ kit (Gen Hunter, Nashville,
TN, U.S.A.), then dissolved in 20 µl DEPC-H2O and stored at –80°C
for further use. mRNAs from fresh and cold-stored pollen were purified from the respective total RNA (1 mg) by use of an mRNA purification kit (Amersham Biosciences, Buckinghamshire, U.K.), and used
directly for cDNA synthesis with use of the cDNA subtraction kit
(Clontech, Palo Alto, CA, U.S.A.). mRNAs from fresh and cold-stored
pollen grains were chosen as tester and driver material, respectively.
After double hybridization and nested PCR, the cDNAs were inserted
directly into the pGEM-T Easy vector (Promega, Madison, WI,
U.S.A.) for sequencing, reverse Northern and Northern blotting analyses. The sequence data were analyzed by use of the Vector NTI Suite
(V.7) (IndorMax, Bethesda, MD, U.S.A.) and BLAST program of
NCBI. For Northern blotting, 20 µg of DNase I-treated total RNA was
separated on a 1% agarose gel, transferred onto a nylon membrane and
cross-linked by UV light. Probe synthesis and hybridization were performed by use of a DIG Northern Starter kit according to the manufacturer’s manual (Roche, Mannheim, Germany). After being incubated
with anti-DIG antibody at 25°C for 1 h, membranes were washed with
buffer twice for 15 min. Membranes were incubated with detection
buffer containing CDP-star ready-to-use reagent and exposed to X-ray
film. The partial sequences of the transcripts characterized in this
study are available upon request.
Semi-quantitative RT-PCR analysis of the expression pattern of genes
in different organs
By using an Absolutely RNA® RT-PCR Miniprep Kit (Stratagene, La Jolla, CA, U.S.A.), total RNA was isolated from 0.1 g of lily
tissues: leaves, roots, tepals, filaments, pistils, pollen and anthers/
stamens of various-sized floral buds, including the mature anther and
in vitro- and in vivo-grown pollen tubes. First-strand cDNA was synthesized from 3 µg total RNA with oligo(dT) primer and random primers according to the manufacturer’s protocol (M-MLV Reverse
Transcriptase; Invitrogen Life Technologies, Carlsbad, CA, U.S.A.).
To study the expression pattern of genes in the pollen grain, in vitrogerminated (12 h) pollen tubes and different organs, RT-PCR analyses
were performed using the gene-specific primer sets as listed in Table 2.
The PCR mix consisted of 1.5 mM magnesium chloride, 0.2 mM
dNTPs, 0.5 µM each of sense and antisense primer, 2.5 U of Taq
polymerase (MDBio, Taipei, Taiwan) and 1× PCR buffer supplied with
the Taq polymerase. The reaction was conducted under the following
protocol: 94°C for 10 min; 25–30 cycles at 94°C for 1 min, annealing
at 55°C for 1 min, 72°C for 1 min; final elongation at 72°C for 10 min
with use of a Biometra® T3 Thermocycler (Whatman Biometra, Goettingen, Germany). The resulting PCR products were run on a 1% agarose gel containing 0.01% ethidium bromide. To monitor the efficacy of
cDNA synthesis by PCR amplification, the housekeeping gene actin
1527
was used as a positive control for PCR amplification, and a 761-bp
fragment was obtained by use of the primer sets listed in Table 2.
Supplementary material
Supplementary material mentioned in the article is available to
online subscribers at the journal website www.pcp.oupjournals.org.
Acknowledgments
We thank Drs. E.M. Lord and Yue-Ie Hsing for critically reading
and providing comments on the manuscript. This work was supported
by grants NSC 91-2317-B-001-014 and NSC 92-2311-B-001-064 from
National Science Council, Academia Sinica (Taiwan, ROC) and the Li
Foundation (U.S.A.) to G.Y.J.
References
Becker, J.D., Boavida, L.C., Carneiro, J., Haury, M. and Feijo, J.A. (2003) Transcriptional profiling of Arabidopsis tissues reveals the unique characteristics
of the pollen transcriptome. Plant Physiol. 133: 713–725.
Chauhan, S., Forsthoefel, N., Ran, Y., Quigley, F., Nelson, D.E. and Bohnert,
H.J. (2000) Na+/myo-inositol symporters and Na+/H+-antiport in Mesembryanthemum crystallinum. Plant J. 24: 511–522.
da Costa-Nunes, J.A. and Grossniklaus, U. (2003) Unveiling the gene-expression profile of pollen. Genome Biol. 5: 205.
Engel, M.L., Chaboud, A., Dumas, C. and McCormick, S. (2003) Sperm cells of
Zea mays have a complex complement of mRNAs. Plant J. 34: 697–707.
Guyon, V.N., Astwood, J.D., Garner, E.C., Dunker, A.K. and Taylor, L.P. (2000)
Isolation and characterization of cDNAs expressed in the early stages of flavonol-induced pollen germination in petunia. Plant Physiol. 123: 699–710.
Hoekstra, F.A. and Bruinsma, J. (1979) Protein synthesis of binucleate and trinucleate pollen and its relationship to tube emergence and growth. Planta
146: 559–566.
Honys, D. and Twell, D. (2003) Comparative analysis of the Arabidopsis pollen
transcriptome. Plant Physiol. 132: 640–652.
Jauh, G.Y. and Lord, E.M. (1995) Movement of the tube cell in lily style in the
absence of the pollen grain and the spent pollen tube. Sex. Plant Reprod. 8:
168–172.
Jauh, G.Y. and Lord, E.M. (1996) Localization of pectins and arabinogalactan
proteins in lily (Lilium longiflorum L.) pollen tube and style and their possible roles in pollination. Planta 199: 251–261.
Kelley, W.L. (1998) The J-domain family and the recruitment of chaperone
power. Trends Biochem. Sci. 23: 222–227.
Kitagawa, K., Skowyra, D., Elledge, S.J., Harper, J.W. and Hieter, P. (1999)
SGT1 encodes an essential component of the yeast kinetochore assembly
pathway and a novel subunit of the SCF ubiquitin ligase complex. Mol. Cell
4: 21–33.
Lee, J.Y. and Lee, D.H. (2003) Use of serial analysis of gene expression technology to reveal changes in gene expression in Arabidopsis pollen undergoing
cold stress. Plant Physiol. 132: 517–529.
Lemmon, S.K. (2001) Clathrin uncoating: Auxilin comes to life. Curr. Biol. 49–
52.
Lord, E. (2000) Adhesion and cell movement during pollination: cherchez la
femme. Trends Plant Sci. 5: 368–373.
Lord, E.M. and Russell, S.D. (2002) The mechanisms of pollination and fertilization in plants. Annu. Rev. Cell. Dev. Biol. 18: 81–105.
Mascarenhas, J.P. (1975) The biochemistry of angiosperm pollen development.
Bot. Rev. 41: 259–314.
Mascarenhas, J.P. (1990) Gene activity during pollen development. Annu. Rev.
Plant Physiol. Plant Mol. Biol. 41: 317–338.
Mascarenhas, J.P. (1993) Molecular mechanisms of pollen tube growth and differentiation. Plant Cell 5: 1303–1314.
McCormick, S. (1993) Male gametophyte development. Plant Cell 5: 1265–
1275.
Palanivelu, R., Brass, L., Edlund, A.F. and Preuss, D. (2003) Pollen tube growth
and guidance is regulated by POP2, an Arabidopsis gene that controls GABA
levels. Cell 114: 47–59.
1528
Gene expression profile in germinating lily pollen
Read, S.M., Clark, A.E. and Bacic, A. (1993) Stimulation of growth of cultured
Nicotiana tabacum W38 pollen tubes by poly(ethylene glycol) and Cu (II)
salts. Protoplasma 177: 1–14.
Rosen, W.G. (1971) Pistil-pollen intreactions in Lilium. In Pollen: Development
and Physiology. Edited by Harrison, J.H. pp. 239–254. Butterworth & Co
Ltd, New York.
Taylor, L.P. and Hepler, P.K. (1997) Pollen germination and tube growth. Annu.
Rev. Plant Physiol. Plant Mol. Biol. 48: 461–491.
Twell, D. (2002) The developmental biology of pollen. In Plant Reproduction.
Edited by O’Neill, S.D. and Roberts, J.A. pp. 86–153. Sheffield Academic
Press, Sheffield, U.K.
Wang, C.S., Walling, L.L., Eckard, K.J. and Lord, E.M. (1992) Immunological
characterization of a tapetal protein in developing anthers of Lilium longiflorum. Plant Physiol. 99: 822–829.
Willing, R.P. and Mascarenhas, J.P. (1984) Analysis of mRNA from pollen and
shoots of Tradescantia. Plant Physiol. 75: 865–868.
Willing, R.P., Bashe, D. and Mascarenhas, J.P. (1988) An analysis of the quantity and diversity of messenger RNAs from pollen and shoot. of Zea mays.
Theor. Appl. Genet. 75: 751–753.
Xu, H., Weterings, K., Vriezen, W., Feron, R., Xue, Y., Derksen, J. and Mariani,
C. (2002) Isolation and characterization of male-germ-cell transcripts in
Nicotiana tabacum. Sex. Plant Reprod. 14: 339–346.
Yang, M., Hu, Y., Lodhi, M., McCombie, W.R. and Ma, H. (1999) The Arabidopsis SKP1-LIKE1 gene is essential for male meiosis and may control
homologue separation. Proc. Natl Acad. Sci. USA 96: 11416–11421.
Ylstra, B. and McCormick, S. (1999) Analysis of mRNA stabilities during pollen development and in BY2 cells. Plant J. 20: 101–108.
(Received March 26, 2004; Accepted August 5, 2004)