|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Fels Institute for Cancer Research and Molecular Biology4
Department of Biochemistry,5 Temple University School of Medicine, Philadelphia, Pennsylvania 19140
Wisconsin National Primate Research Center,6 University of Wisconsin, Madison, Wisconsin 53715
Division of Reproductive Sciences,7 Huntsman Cancer Institute, University of Utah Health Sciences Center, Salt Lake City, Utah 84108
| ABSTRACT |
|---|
|
|
|---|
embryo, gametogenesis, gene regulation, ovum
| INTRODUCTION |
|---|
|
|
|---|
The acquisition of oocyte developmental competence, commonly referred to as cytoplasmic maturation, requires an ongoing and intimate relationship between the oocyte and follicular cells. Bidirectional signals between the oocyte and follicular cells likely occur throughout oogenesis to ensure that the oocyte and follicle cells differentiate in a coordinated manner [46]. Intimate associations between oocytes and granulosa cells are established through gap junctions as well as paracrine interactions [6, 7]. Signals between the oocyte and granulosa cells include signals that emanate from the oocyte, such as growth and differentiation factor 9 (GDF9), bone morphogenetic proteins BMP15 and BMP6, and oocyte-secreted protein, OOSP (reviewed in [6]). Oocytes appear to be responsible for directing the nearest follicular cells to differentiate into cumulus cells by supporting the expression of hyaluronidase and suppressing the expression of LH receptor [810]. Additionally, granulosa cells express Kit ligand, which can interact with its receptor on oocytes (reviewed in [4]). These interactions between the oocyte and the follicle cells are further influenced by hormonal signals [11, 12]. Available data from in vitro oocyte culture systems indicate that disruptions in these complex intercellular signaling processes can lead to the development of oocytes that are capable of many of the events associated with fertilization and oocyte activation, but nevertheless display substantial reductions in developmental competence [1214].
The degree to which correct oocyte-follicle cell interactions occur during in vitro maturation, when the oocyte-granulosa cell complex is exposed to an altered extracellular milieu, has not been well studied, particularly in nonhuman primates and humans. There is considerable interest in understanding the molecular factors that determine long-term oocyte developmental competence because such knowledge will be of value to basic, applied, and clinical areas of reproductive biology. The investigation of these factors is particularly relevant to understanding female infertility in humans, but progress in this area has been inhibited by ethical constraints and experimental limitations on the use of human oocytes and embryos. Studies in a suitable nonhuman primate model, however, provide one means of addressing such questions.
Comparative studies of the developmental competence of in vivo-matured oocytes (high developmental competence [1519]) and in vitro-matured oocytes from both FSH-primed monkeys (moderate developmental competence [15, 17, 20]) and nonstimulated monkeys (low developmental competence [15, 2022]) have demonstrated that cytoplasmic maturation is acquired progressively during the course of oocyte development and maturation. Developmental failure of oocytes derived from large follicles of FSH-primed monkeys occurs predominantly during the embryonically driven period of development [15, 17, 20] and may be caused in part by impairments in activation of the embryonic genome [17]. In contrast, developmental failure of in vitro-matured oocytes derived from small antral follicles of nonstimulated monkeys occurs to a large extent during the maternally driven period of embryogenesis [15, 2022]. These results are consistent with possible impairments in cytoplasmic maturation, which may have diverse effects on embryogenesis, depending on when they were incurred during the course of oocyte development and maturation. To evaluate the difference in ooplasmic properties between oocytes of differing developmental potentials, we compared the regulation and expression of maternal mRNAs in oocytes of the above three types and in the embryos derived from them. Our observations indicate that the normal changes in transcriptional activity and maternal mRNA stabilization during oocyte development fail to occur in the oocytes obtained from small antral follicles of nonstimulated females. Additionally, culture during in vitro maturation may impair the expression of mRNAs that support embryonic genome activation, leading to developmental failure during the embryonically driven period of development.
| MATERIALS AND METHODS |
|---|
|
|
|---|
The isolation and culture of rhesus oocytes and embryos was performed as described ([23] and references therein; see also http://www.preger.org). The PREGER sample set encompasses more than 160 samples. From among this collection, samples were employed representing germinal vesicle (GV)-stage oocytes, in vivo-matured metaphase II (MII)-stage oocytes, in vitro-matured MII-stage oocytes, and embryos derived from them, that were then cultured in vitro in HECM9 medium [24], as described [17]. Three types of oocytes were employed in these studies. In vivo-matured oocytes were obtained from females stimulated with both FSH and hCG (large follicles, 37 mm, high developmental competence). In vitro-matured oocytes were obtained from females stimulated only with FSH (large follicles, moderate developmental competence) and from nonstimulated females at random stages of the menstrual cycle (small antral follicles, 450 µm to 2 mm, low developmental competence). These oocytes will be referred to as hCG, FSH, and nonstimulated (NS) oocytes, respectively. GV-stage oocytes from hCG-stimulated females were those that failed to mature in vivo. All somatic cells and the zona pellucida were removed from oocytes and embryos before processing for reverse transcription-polymerase chain reaction (RT-PCR). For most stages/conditions, three or more samples of 14 oocytes/embryos were obtained. Because the entire mRNA population is amplified during the PCR procedure, the amount of input mRNA (i.e., the range of 14 rhesus oocytes/embryos) does not affect the quantitative representation of sequences within the population of amplified cDNAs. Only a single sample was obtained for the 8-cell stage of embryos from FSH-stimulated females and for early blastocyst-stage embryos obtained by in vitro maturation of oocytes taken from nonstimulated females. No conclusions were based on comparisons employing these single samples.
The general care and housing of rhesus monkeys (Macaca mulatta) at the Wisconsin National Primate Research Center have been described previously [25, 26]. The Wisconsin National Primate Research Center is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care, and animal protocols and experiments were approved by the Graduate School Animal Care and Use Committee and the University of Wisconsin-Madison. The animals were maintained according to recommendations of the Guide for the Care and Use of Laboratory Animals and the Animal Welfare Act with its subsequent amendments.
Messenger RNA Expression Analysis
The cDNA probes employed in these studies were described previously [23, 27]. Complementary DNA probes were obtained by RT-PCR or from other sources as indicated. The identities of amplified cDNAs were confirmed either by using diagnostic restriction digests or DNA sequencing. Blot preparation, probe preparation, hybridization, and quantitative analyses were performed as described [23, 28, 29]. Data were expressed as the mean (± SEM) cpm bound value for each stage/condition of oocytes and embryos included in the analysis. For some analyses, the ratios of expression were calculated among the three classes of oocytes or embryos derived from them, and the mean ratio (± SEM) was calculated for each stage. Significance of differences was evaluated using the t-test.
| RESULTS |
|---|
|
|
|---|
Comparisons among the three classes of oocytes (hCG, FSH, and NS) should provide the opportunity to evaluate the possible molecular basis for differences in gene expression, and specifically the effects of follicle size and oocyte maturation conditions on oocyte and embryo gene expression. Comparisons between FSH and hCG samples provides a chance to evaluate effects of in vitro versus in vivo maturation. The comparatively poor ability of NS oocytes to undergo meiotic maturation and to support preimplantation development [15, 20] indicates that gene expression is likely to be more severely disrupted. To explore these possibilities, we wished to compare first the expression of maternal mRNAs between these three classes of oocytes and embryos derived from them. From among the mRNAs that had been analyzed thus far on the PREGER sample set [23, 27], the expression of 23 mRNAs that exhibited significant expression (
100 cpm bound) in oocytes or early cleavage-stage embryos (Fig. 1; includes data reproduced from [23, 27]) was examined (Table 1). These include 4 housekeeping mRNAs, 6 transcription factor mRNAs, the oocyte-specific histone H1 linker H1FOO mRNA, 2 mRNAs encoding members of the chromatin accessibility complex that participates in DNA replication and gene transcription, and 10 mRNAs encoding chromatin modifying factors that regulate gene transcription.
|
|
Plotting the data for the hCG, FSH, and NS series of samples from the GV stage through the 8-cell stage revealed that for most of these mRNAs, the expression in FSH and hCG oocytes and embryos was comparatively similar at most stages, whereas the expression in NS oocytes and embryos was dramatically different for most stages and mRNAs (Fig. 2). For some mRNAs the expression in NS MII-stage oocytes (ACTB, HPRT1, TBP, HDAC2, RYBP, YY1, SMARCA5, SMARCC1, and SMARCE1), or NS pronucleate (PN)-stage embryos (ACTB, HPRT1, PDHA1, HSPA8, HAT1, HDAC2, RYBP, YY1,CHRAC1, POLE3, and SMARCA5) was significantly elevated relative to the hCG class. Significant differences likewise existed for 6 of these genes between NS and FSH MII-stage oocytes (annotated as a and c in Fig. 2). Overall, FSH and hCG MII-stage oocytes and PN-stage embryos were more similar to each other than to NS oocytes or embryos. For some mRNAs (ACTB, HPRT1, TBP, HDAC2, RYBP, YY1, and SMARCC1), expression in both FSH and NS MII-stage oocytes was significantly different from that of hCG oocytes (annotated as a and b in Fig. 2). With development to the 2-cell stage, these mRNAs displayed similar levels of expression between FSH and hCG embryos, but 2 of these mRNAs (ACTB and TBP) were repressed significantly in NS embryos relative to hCG embryos (annotated as "a" at the 2-cell stage). For some mRNAs (HPRT1, H1FOO, TBP, SMARCC1, and SMARCE1), expression in FSH oocytes, NS oocytes, or both differed from hCG oocytes, but during cleavage these differences diminished. The RYBP and YY1 mRNAs, which encode functionally related transcription molecules, were notable in that both mRNAs displayed a greater divergence in expression (NS:hCG > FSH:hCG) as developmental competence decreased (hCG > FSH > NS).
|
The numbers of mRNAs differing between hCG, FSH, and NS oocytes and embryos revealed that NS oocytes and embryos differed considerably from FSH and hCG oocytes and embryos (Table 2). A total of 25 significant differences were scored between NS and hCG oocytes and embryos, and 19 were scored between NS and FSH oocytes and embryos. By contrast, only 9 differences were scored overall between FSH and hCG oocytes and embryos. A majority of differences (20/25, 15/19, and 9/9, respectively) were displayed at the MII and PN stages.
|
Five of the 23 maternal mRNAs displayed significantly reduced expression in NS 2-cell stage embryos relative to hCG embryos, and an additional 10 displayed a trend toward reduced expression in NS 2-cell stage embryos that did not reach statistical significance. To more fully judge the overall effects of hormonal stimulation on maternal mRNA expression, the ratios of expression in NS:hCG, FSH:hCG, and NS:FSH were calculated for each mRNA, and then a mean ratio was calculated for each stage and plotted (Fig. 3, labeled combined). The average ratio of NS: hCG expression significantly increased (P < 0.01) during maturation from the GV stage to the MII stage, and then decreased significantly with development to the 2-cell stage (P < 1 x 105). To examine this pattern more completely, we divided the 23 mRNAs into 2 groups; group A (n = 19) having an NS:hCG ratio <1.0 at the 2-cell stage, and group B (n = 4) mRNAs having an NS:hCG ratio of 1.0 or greater at the 2-cell stage. Plotting the average ratios for these 2 groups separately revealed that the average NS:hCG ratio for group A mRNAs was 3.26 at the MII oocyte stage and 0.27 at the 2-cell stage, indicating an overabundance (P < 0.001) of these mRNAs initially in the NS oocytes, followed by a substantial reduction in the content of these mRNAs by the 2-cell stage (P < 1 x 107). For the group B mRNAs, the NS:hCG expression ratio was near parity in oocytes, and then increased at the PN stage (P < 0.05). The group B mRNAs (HAT1, CHRAC1, POLE3, and PDHA1) were among those that increase in expression during later preimplantation stages, (Fig. 1), indicating that these genes may be precociously up-regulated in the NS embryos. The average NS:FSH ratios varied in a similar fashion for the 2 groups of mRNAs.
|
For group A mRNAs, there was an approximately 10-fold (P < 1 x 106) increase in the average ratio of NS: hCG expression between the 2- and 8-cell stages (Figs. 2 and 3). The average maternal mRNA expression in 8-cell-stage samples thus rebounded in NS embryos from the low value at the 2-cell stage to a value that was approximately 2-fold greater than in hCG embryos at the 8-cell stage (Fig. 3).
Data for all 23 mRNAs combined revealed a significant increase in the FSH:hCG expression ratio in MII-stage oocytes (P < 0.001). Maternal mRNAs were less severely altered in FSH oocytes and embryos compared with NS embryos. The average FSH:hCG ratio was approximately 1.9, compared with 2.9 for the NS:hCG ratio in MII-stage oocytes. For group A mRNAs, the average ratio of FSH: HCG expression was elevated (2-fold; P < 0.002) in MII-stage oocytes, but decreased by the PN stage (P < 1 x 103) and was close to parity for fertilized embryos (average ratios of 0.84 and 1.19 for PN and 2-cell embryos, respectively), indicating an overall high degree of similarity in maternal mRNA expression between FSH and hCG embryos. The group B mRNAs were expressed nearly equally between FSH and hCG embryos from the GV oocyte through 2-cell stages (median 1.09). Thus, expression of these maternal mRNAs did not differ as dramatically between FSH and hCG embryos as between NS and hCG embryos, or between NS and FSH embryos, indicating that the most severe alteration occurs in NS embryos.
The expression of these mRNAs was quite similar between NS and FSH oocytes (ratios of 1.35 and 1.65, for GV and MII oocytes, respectively). Group A mRNAs displayed a significant increase in the NS:FSH ratio between the GV oocyte stage and the MII oocyte stage (P < 0.001), followed by a dramatic decrease at the 2-cell stage. The significant change in values for group A mRNAs, which comprise the majority of mRNAs analyzed, further indicates that the NS embryos displayed a more severely altered pattern of gene expression than the FSH embryos, and the dramatic decrease at the 2-cell stage resembles that observed for the NS:hCG ratio. The NS:FSH ratio for group B mRNAs did not vary significantly, and displayed a median value of 1.71 for the 4 stages.
Effects on Transcription Factor mRNA Expression
The data presented in Figure 2 reveal significant reductions in the expression of some maternal mRNAs encoding transcription factors (TFs) in 2-cell-stage NS embryos. Additionally, transcription factor mRNAs displayed many significant differences between NS and hCG embryos and between NS and FSH embryos at the PN and 2-cell stages (Table 2). Previous studies in mice revealed that some maternal mRNAs encoding TFs are recruited just before the major genome activation event, and that maternal TF mRNA recruitment may contribute to genome activation [30]. Studies in mice also revealed up-regulation in the expression of TF mRNAs after genome activation [30]. These earlier observations in mice, combined with the decreased expression of maternal TF mRNAs observed in Figure 2, raise the possibility that additional disruptions in the expression of TF mRNAs could contribute to the severe restriction in developmental potential for NS embryos. To explore this possibility, we compared between the 3 classes of embryos the expression of TF mRNAs at later stages (Fig. 4).
|
There were clear disruptions in the expression of TF mRNAs in embryos from both NS and FSH oocytes. Expression of some mRNAs at some stages was quite variable, leading to increased mean expression values in NS embryos. This was predominantly apparent at the morula stage, when 3 of 17 TF mRNAs (YY1, HAT1, and SMARCA5; annotated as a in Fig. 4) were elevated in NS embryos relative to hCG embryos. There were no statistically significant differences between FSH and hCG embryos at the morula stage. For 2 TF mRNAs (TCERG1 and SMARCC1), the combined elevation in NS embryos and repression in FSH embryos produced a significant difference between these 2 classes (annotated as c). Two other chromatin modifying mRNAs (HDAC2 and SMARCA4) were variably increased in NS morulae, but the difference did not reach statistical significance. Only 2 mRNAs (YY1 and SMARCE1) displayed a statistically significant difference between NS and hCG expanded blastocysts, and these 2 classes of embryos often displayed mRNA expression profiles that converged at the expanded blastocyst stage.
| DISCUSSION |
|---|
|
|
|---|
One of the most significant changes observed here was the altered expression of maternal mRNAs in NS oocytes and embryos. Metaphase II-stage oocytes and pronucleate-stage embryos from NS females displayed an overall increase in the apparent expression of many of the maternal mRNAs assayed. This may reflect a failure of these oocytes to undergo the normal pattern of transcriptional silencing during in vitro maturation. Earlier studies revealed that perinucleolar chromatin condensation, believed to be a possible indicator of processes leading to transcriptional silencing, does not occur as readily in the NS oocytes from follicles of <1 mm diameter [17]. Differences in maternal mRNA polyadenylation, stabilization, and translation-coupled degradation during maturation could also contribute to this effect by causing precocious polyadenylation, thereby affecting the efficiency of reverse transcription for some mRNAs within the population.
After the PN stage, NS embryos displayed a dramatic reduction in expression and in the overall abundance of many of the maternal mRNAs examined at the 2-cell stage. Some mRNAs were more severely affected than others, so it does not appear that there is simply a systematic, uniform depletion of all mRNAs. This decrease may reflect a deficiency in mRNA stabilization, possibly related to overexpression of certain mRNAs exceeding the capacity of the oocyte's mRNA stabilization mechanism, or precocious degradation following precocious polyadenylation.
Perturbations in transcription and maternal mRNA stabilization/regulation likely operate in concert to produce oocytes that are unable to provide the normal pattern of gene regulation, epigenetic modification, metabolism, and homeostasis, thereby compromising early development. The rebound in expression of these mRNAs in NS embryos observed after the 2-cell stage therefore most likely reflects the end result of preferential survival of a fraction of NS embryos that either experienced less severe perturbation, possibly as the result of being more developmentally advanced at the time of their isolation, or that by chance possessed a greater capacity for maternal mRNA stabilization. These observations are thus consistent with the hypothesis that the ooplasmic composition of NS oocytes is inadequate at the level of maternal mRNA content and, in addition, indicate other likely impairments related to maternal mRNA stability. Cytoplasmic changes acquired during oocyte development and maturation produce an oocyte that will support the normal pattern of gene expression, both before and after fertilization. These interactions appear to be disrupted with in vitro maturation, with the degree of this disruption depending on the developmental stage, follicle size, and prior hormonal exposure at the time of isolation.
Maternal mRNAs were less severely altered in FSH oocytes and embryos compared with NS embryos, but many mRNAs nevertheless displayed significant differences in FSH versus hCG MII-stage oocytes. These observations indicate that, overall, in vitro-matured oocytes from FSH-primed monkeys resemble in vivo-matured oocytes to some degree, but that in vitro maturation alters the abundance of certain mRNAs at the MII stage. This could reflect an effect of culture combined with disruption of the normal cellular associations within the follicle.
A second major difference among the three classes of oocytes and embryos was the tendency for disruptions in the regulation and expression of TF mRNAs. Some of these mRNAs were reduced in expression at the 2-cell or 8-cell stages in NS embryos. Because the initial wave of transcription occurs at the 2-cell stage and the major genome activation event, including the onset of nucleolar transcription occurs at the 8-cell stage in rhesus embryos [19], this reduction could interfere with timely and complete embryonic genome activation if an early increase in the expression of these TFs helps to promote broader embryonic genome activation. These results are consistent with the observed impairments in the onset of nucleolar transcription and high rate of developmental failure beyond the 8-cell stage in embryos derived from in vitro-matured rhesus monkey oocytes [17].
Some TF mRNAs were altered in expression at later stages in either NS or FSH embryos, indicating a likely failure to execute the normal developmental program even in the subset of embryos that attain those stages. This could contribute additional reductions in long-term development of embryos from in vitro-matured oocytes. Of interest, several other TF mRNAs were aberrantly expressed at later stages of development, or, in some cases, displayed aberrant fluctuations in expression. Some of these mRNAs, as well as the multifunctional CHRAC1 and POLE3 mRNAs, were underexpressed in FSH blastocysts. These observations indicate that even at these later stages, embryonic gene expression is likely to be abnormal in embryos derived from in vitro-matured oocytes.
Our results indicate that disruptions in the process of cytoplasmic maturation during oocyte development or maturation can have differential effects on expression of maternal mRNAs and expression of embryonically encoded genes, and that these are associated with developmental failure. This has important implications for assisted reproductive technologies in humans, with respect to methods employed for in vitro oocyte maturation. This is particularly important for women with polycystic ovarian syndrome (PCOS), a leading cause of infertility. Ovarian stimulation of women with PCOS often leads to hyperstimulation syndrome, and oocytes are typically impaired in their ability to lead to successful pregnancies. Suitable methods for in vitro maturation would be advantageous. Little progress has been made to enhance the developmental capacity of in vitro-matured human oocytes. It is clear that requirements for the acquisition of developmental competence by primate (human and nonhuman) oocytes differ substantially from those of rodent or bovine oocytes [31, 32]. Thus, progress in this field will benefit tremendously from a detailed understanding of the molecular processes involved in cytoplasmic maturation of primate oocytes, and how they are affected by extrinsic factors such as culture and endocrine abnormalities.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: Keith E. Latham, 3307 North Broad Street, Philadelphia, PA 19140. FAX: 215 707 1454; klatham{at}temple.edu ![]()
3 Current address: Department of Reproductive Medicine, Chiba University School of Medicine, 1-8-1 Inohana, Chuo-ku, Chiba 260-8677, Japan ![]()
Received: 2 September 2004.
First decision: 11 October 2004.
Accepted: 16 November 2004.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. K. Nyholt de Prada, Y. S. Lee, K. E. Latham, C. L. Chaffin, and C. A. VandeVoort Role for cumulus cell-produced EGF-like ligands during primate oocyte maturation in vitro Am J Physiol Endocrinol Metab, May 1, 2009; 296(5): E1049 - E1058. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. Wu Maternal depletion of NLRP5 blocks early embryogenesis in rhesus macaque monkeys (Macaca mulatta) Hum. Reprod., February 1, 2009; 24(2): 415 - 424. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. S. Lee, K. E. Latham, and C. A. VandeVoort Effects of in vitro maturation on gene expression in rhesus monkey oocytes Physiol Genomics, October 8, 2008; 35(2): 145 - 158. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. M. Jones, D. S. Cram, B. Song, G. Kokkali, K. Pantos, and A. O. Trounson Novel strategy with potential to identify developmentally competent IVF blastocysts Hum. Reprod., August 1, 2008; 23(8): 1748 - 1759. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. M. Jones, D. S. Cram, B. Song, M. C. Magli, L. Gianaroli, O. Lacham-Kaplan, J. K. Findlay, G. Jenkin, and A. O. Trounson Gene expression profiling of human oocytes following in vivo or in vitro maturation Hum. Reprod., May 1, 2008; 23(5): 1138 - 1144. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. R Mtango and K. E Latham Differential Expression of Cell Cycle Genes in Rhesus Monkey Oocytes and Embryos of Different Developmental Potentials Biol Reprod, February 1, 2008; 78(2): 254 - 266. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. R. Mtango and K. E. Latham Ubiquitin proteasome pathway gene expression varies in rhesus monkey oocytes and embryos of different developmental potential Physiol Genomics, September 11, 2007; 31(1): 1 - 14. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Zheng, R. Vassena, and K. E. Latham Effects of in vitro oocyte maturation and embryo culture on the expression of glucose transporters, glucose metabolism and insulin signaling genes in rhesus monkey oocytes and preimplantation embryos Mol. Hum. Reprod., June 1, 2007; 13(6): 361 - 371. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. A Maloney and W. D Rees Gene-nutrient interactions during fetal development Reproduction, October 1, 2005; 130(4): 401 - 410. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Zheng, R. D. Schramm, and K. E. Latham Developmental Regulation and In Vitro Culture Effects on Expression of DNA Repair and Cell Cycle Checkpoint Control Genes in Rhesus Monkey Oocytes and Embryos Biol Reprod, June 1, 2005; 72(6): 1359 - 1369. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |