Russell Laboratory Research Page

This is an introductory webpage describing our approaches toward understanding sexual plant reproduction. We are interested in the control of sexual reproduction in angiosperms. In particular, we have been interested in the transcriptome of male and female gametes of rice and Plumbago, which express typical and preferential fertilization, respectively. Control of expression through epigenetics is a focus as this influences the initial chromatin state of the zygote and young embryo.


Maternal to Zygotic Transition. Our current work on rice focuses on the transcriptome of the gametes and zygote at specific points in the timetable around fertilization using RNA-Seq to provide whole transcriptomic data. This is part of a collaboration with the laboratory of Professor V. Sundaresan of University of California-Davis. To date, we have worked out all of the needed technologies to conduct controlled pollinations of rice plants and to be able to collect fertilized zygotes at specific landmark times during the post-pollination development of rice. All isolation are conducted using living ovaries and rapid dissection of component cells, which are then frozen in liquid nitrogen and retain at -80°C until processing.

Male and Female Gametes Microarray results. Prior to the RNA-Seq approach, we used the 57K microarray of Affymetrix to characterize male and female genes expressed. The data is archive on GEO at http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE17002. Preferential Fertilization. In Plumbago zeylanica the two sperm cells that participate in double fertilization are different and have different predetermined fates. One important structural difference between these sperm cells is the one of them is physically associated with the nucleus of the pollen grain, known as the vegetative nucleus (or VN). The sperm cell associated with the VN is known as the Svn. The other sperm cell, in turn, is connected with the prior Svn, but is unassociated with the VN, so it has been termed the Sua. The former fertilizes the central cell and its sperm nucleus, together with the female polar nuclei, establishes the primary endosperm nucleus and the nutritive endosperm. The Sua fuses with the egg cell and forms the zygote, which divides to form the embryo and ultimately the new seedling. The control of this phenomenon is our principal interest.

We have used a number of methods to characterize this. First, we collected ESTs (expressed sequence tags) from the two different populations of sperm cells found in the pollen of Plumbago zeylanica. Additional structural distinctions between the cells are numerous, but one of the most significant is that the Svn contains many mitochondria (ave 252) but few plastids (usually none, but up to two have been observed). In contrast, the Sua contains numerous plastids (ave. 26) but only 20% of the number of mitochondria (ave. 52). The sperm cells preferentially fuse in the embryo sac and follow the pattern above in over 95% of observations. The technology of sperm cell collection is shown in Figure 1. In wind pollinated plants that produce over several grams of pollen per plant (e.g. maize), it is possible to collect cells en masse using FACS (fluorescence activated cell sorting). Plumbago is a limited model for mass collection, however, because it is insect pollinated. Each flower contains five anthers that under ideal conditions produce 256 pollen grains. This limits collection amount, but our collection method, because it is based on manual selection is more accurate at discriminating cell types. We select sperm cells from associated cell pairs. Cells that cannot be discriminated, if any, are omitted. Cell collections are frozen every 10 to 15 minutes and quickly frozen in liquid nitrogen. Thus, each collection is to a specific cell type. Collections are subsequently pooled as needed.

Figure 1. This illustrates the collection technology for collecting each population of sperm cells. Upper left: Differential interference contrast microscopy of sperm cells immediately after isolation from a pollen grain of Plumbago zeylanica. Sperm cell (Svn) associated with the vegetative nucleus (VN) is larger and contains numerous mitochondria and rare plastids. The other sperm cell (Sua) is unassociated, contains numerous plastids and few small mitochondria (Russell, 1984). Bar=10 m. Lower left: Sperm cells soon become spherical but can still be discriminated. The granular texture of the Sua reflects the presence of numerous plastids. Bar=10 m. Center: Population of Sua sperm cells collected using micropipette. Bar=30 m. Left: Sperm cells remain viable and intact, as indicated by accumulation of fluorescein diacetate in epifluorescence microscopy. Bar=30 m. (From Zhang et al., 1998).

Technology of EST Production. Once we obtained sufficient sperm cells of each sperm cell type, the next phase of our work was obtaining a high-quality library of ESTs. We constructed two libraries: one for each of the sperm cell morphs. We now have sequences from Svn and Sua cDNAs from both ends of the directionally cloned inserts. These are now present in GenBank and are also available on the web site of the OU Core DNA Facility, known as the Advanced Center for Genome Technology (ACGT). Sperm cells of flowering plants have less mRNA than many other types of animal cells. For example, in maize, 1.8 g of total RNA could be obtained from 1 million FACS-purified sperm (Engel et al. 2003). We used about 11000 sperm cells to construct a size-selected (>400bp) cDNA library for each cell type by using the SMART cDNA library construction kit from Clontech. In order to minimize the possibility of non-representative amplification, PCR cycle number was kept to a minimum (Chenchik et al. 1998). In our experience, 24 cycles of PCR are sufficient to generate enough cDNA for library construction. A portion of the cDNA was ligated into λTriplEx2 arms and packaged: the titer of the unamplified libraries was 2.1 x 107 and 3.2 x 107 pfu/ml, for Sua and Svn, respectively. The amplified libraries had a titer of 1.5 x 1010 and 1.6 x 1010 pfu/ml, for Sua and Svn, respectively. The libraries are of high quality: 90% of the clones had inserts and over 95% of the inserts were at least 500 bp. This phage-based system facilitates initial plating, color screening for successful insertion, and mass-excision into plasmid for sequencing. Randomly selected clones were collected from plates, grown and sequenced at the University of Oklahoma Advanced Center for Genome Technology (ACGT) Lab.

All clones were sequenced from both the 5 and 3 ends. After sequencing, they were assembled to contigs. To date, we have submitted 893 and 629 sequences of Sua and Svn sperm cell cDNA libraries, respectively, to GenBank. Among them, 716 and 471 sequences, respectively, are singletons (including sequences from the same clone that could not be assembled). After screening out ribosomal RNA, mitochondrial RNA, vector and small (less than 100 bp) inserts, we have 893 high quality sequences from the Sua representing 324,774 total bases that are archived on GenBank. After screening out ribosomal RNA, mitochondrial RNA, vector and small (less than 100bp) inserts, we have 629 high quality sequences from the Svn representing 241,007 total bases that are archived on GenBank. Assigned GenBank numbers are CB816827 to CB817719 for the Sua and CB817720 to CB818348 for the Svn. All of our data is available from our ftp site. Blast searches on this data can be made for the Sua and Svn datasets. A keyword search of a blastx search of GenBank for the Sua and Svn data is also available.

Svn search here:
Sua search here:

(These data are not yet linked to a unigene database as the number of EST's sequenced doesn't warrant this yet.)

These results indicate that sperm cells of flowering plants are intact cells with diverse messenger RNA populations. A portion of them could be successfully BLASTed to obtain putative protein sequences. These potential homologs can be categorized into different functional classes. For the Sua, 147 sequences were identified that are involved in information replication, storage, cellular process and metabolism pathways, but most sequences of the Sua could not be classified, including 229 hypothetical and unknown proteins. The other 340 sequences had no database match. This indicates the limitation of current databases, as there are only four flowering plants for which sperm cell EST sequences are currently available. That so too few plants have been sampled for this tissue to be meaningfully compared for the diversity of genes that may be involved in angiosperm gamete biology and double fertilization. Currently a few common homologs are found among the available sequences. Perhaps this indicates that current coverage is remarkably poor, but an alternative explanation is that sperm cells may be highly variable between maize, rice, tocacco and Plumbago. This also applies when the sample is expanded to include generative cells of lily, which currently represents the best described generative cell library. As stated above, we intend to obtain at least 10,000 high-quality sequences from each sperm cell cDNA library. This will help to establish a representative profile of relative gene expression between the two sperm cells. Enhanced gene sampling using other flowering plant sperm cells will also improve coverage.

Some recent publications (with PDF links):


Here are some further resources:

This and other pages on this site are under construction. Please forgive incomplete coverage of our current work as we do not always update it as often as we might.


This file was last modified on Wednesday, 12-Jun-2013 12:15:00 CDT 185278