|
|
||||||||
Regular Article |
a Dipartimento di Strutture, Funzioni, Patologie e Biotecnologie, Università di Teramo, Italy
b Department of Gene Expression & Development, The Roslin Institute, United Kingdom
c Research Institute of Animal Science, Praha, Czech Republic
d Istituto Zootecnico Caseario, 07040 Olmedo, Italy
e Institute of Molecular Genetics, CNRS, UMR-5535, Montpellier, France
f Laboratory of Protein Function, The Babraham Institute, Babraham, Cambridge, United Kingdom
g Department of Animal Reproduction, University of Agriculture, Krakòw, Poland
| ABSTRACT |
|---|
|
|
|---|
developmental biology, granulosa cells
| INTRODUCTION |
|---|
|
|
|---|
Nuclear reprogramming denotes the structural and functional modification associated with the reacquisition of pluripotency, or totipotency, of embryonic and somatic nuclei transplanted into enucleated oocytes [13]. At present, two main approaches have been advocated to induce reprogramming of somatic cell nuclei. The first focuses on exposing the transplanted nucleus for prolonged periods to the oocyte cytoplasmic environment during M phase when levels of maturation-promoting factor (MPF) are high [2, 3, 14, 15]. The second approach is to induce nuclear quiescence in cultured cells before their use as nuclear donors [2]. Both approaches aim to modify the structure and function of donor chromatin to render it more accessible to reprogramming factors in the oocyte cytoplasm. For example, the displacement of transcription factors is thought to be a common feature resulting both from the entry into G0 after serum deprivation [16] and from chromosome condensation induced by MPF [17]. However, despite the use of a highly defined, synchronous population of G0 nuclear donors [2] or the uniform, prolonged exposure of nuclei to MPF in nonactivated cytoplasts [3], complete reprogramming is achieved only occasionally, as indicated by the low number of reconstructed embryos that develop into live young [2, 3]. Moreover, work carried out in cattle [18] and, recently, in mice [19] provides evidence that actively proliferating fetal or embryonic stem cells that are in S or G2/M phase are equally permissive for cloning. Hence, factors unrelated to the cell cycle apparently restrict the development of embryos reconstructed using somatic cells, and an incomplete reprogramming of the somatic nucleus by the oocyte's cytoplasm remains the most suitable explanation for such developmental failures.
Reprogramming the parental genomes through the germ line is the logical result of a complex evolutionary process designed to induce a functional complementarity between maternal and paternal alleles. Current somatic cell cloning procedures induce a dramatic shortcut in which events that normally take place separately in germ line are grossly recapitulated in a common cytoplasmic environment, probably during the first 24 h of embryonic development. Therefore, it is unsurprising that the oocyte's reprogramming machinery, normally sufficient to realize the gamete's intrinsic totipotency, is often inefficient in reprogramming a somatic cell nucleus. In theory, an improvement in reprogramming efficiency might be achieved by altering the somatic cell chromatin to the level of organization normally present in gametes. Obviously, such a radical approach would involve destruction of the normal differentiated cellular organization.
In the present study, we tested whether somatic cell nuclei subjected to denaturing conditions are able to direct the development of enucleated sheep oocytes. Initially, we analyzed the effects of physical methods of denaturation on the accessibility of bulk chromatin structure to exogenous nucleases. In later developmental studies, we injected both chemically and physically denatured nuclei into enucleated oocytes and analyzed the subsequent capacity of such reconstructed embryos to form blastocysts, fetuses, and viable young.
| MATERIALS AND METHODS |
|---|
|
|
|---|
All animal experiments were performed in accordance with DPR 27/01/1992 (Animal Protection Regulation in Italy) in conformity with European Community regulation 86/609. Adult ewes were synchronized with s.c. progesterone implants (Crestar; Intervet, Boxmeer, Holland), and follicular growth was primed with 4.8 mg of ovine FSH (Ovagen, ICP, Auckland, New Zealand) given in 6 doses every 12 h. For oocyte recovery, donor ewes were anesthetized with acetopromazine maleate (0.05 mg/kg body wt) and pentothal sodium (10 mg/kg) before surgery. Sheep granulosa cells from in vitro-matured cumulus-oocyte complexes (COCs) were used throughout the experiments as nuclear donors because 1) they are easily harvested; 2) 98% of them are at the G0/G1 stage of the cell cycle (bromodeoxyuridine [BrdU] labeling in the present study and [3]) so they are functionally compatible with a metaphase II cytoplast [20]; and 3) they have been shown to be able to direct development to term after nuclear transfer [3]. Immature oocytes were aspirated from ovaries collected at the local abattoir or from hormonally primed, individual ewes.
Oocytes surrounded by at least 3 layers of granulosa cells were selected for in vitro maturation. Maturation medium (TCM-199; Sigma, St. Louis, MO) was enriched with 10% fetal bovine serum (Gibco Life Technology, Milan, Italy), 5 µg/ml of FSH (Ovagen), 5 µg/ml of LH (Sigma), 1 µg/ml of estradiol (Sigma), 0.3 mM sodium pyruvate, and 100 µM cysteamine. In vitro maturation was carried out in a humidified atmosphere of 5% CO2 at 39°C for 24 h. In two replicates, BrdU (100 µM; Sigma) was added to the medium at the end of the maturation process, and COCs were fixed and processed for indirect immunofluorescence to determine S-phase cells present at the time of nuclear transfer. Granulosa cells were washed in Hepes-synthetic oviductal fluid (H-SOF) and stored at room temperature until use for nuclear transfer. Granulosa cells were directly injected into individual enucleated oocytes [21], either fresh or after being fixed in 99.9% methanol for 30 min or heated in low-osmolarity (200 mOsm/kg) H-SOF in a water bath at 55°C or 75°C for 15 min. Injected oocytes were immediately activated in H-SOF plus 5 µM ionomycin and subsequently cultured in SOF with 10 µg/ml of cycloheximide and 7.5 µg/ml of cytochalasin B for 5 h. The inhibitor was then removed by washing, and cultures were continued in the conditions described below.
In each trial, an aliquot of injected oocytes were fixed and stained after 1, 3, 6, and 9 h for the assessment of nuclear dynamics following activation. As a control of cytoplast competence, in each trial a subsample of oocytes were fertilized in vitro using fresh semen obtained from a Sarda breed ram. The in vitro fertilization (IVF) medium was bicarbonate-buffered SOF enriched with 20% (v/v) heat-inactivated estrous sheep serum, 2.9 mM Ca2+ lactate, and 16 µM isoproterenol. The IVF was carried out at 39°C in a humidified atmosphere of 5% CO2 for 20 h. Successfully injected oocytes and control IVF embryos were transferred to 20-µl culture drops consisting of SOF supplemented with 2% (v/v) BME-essential amino acids (Sigma), 1% (v/v) minimum essential medium (MEM)-nonessential amino acids, 1 mM glutamine, and 8 mg/ml of BSA-fatty acid free. Cultures were carried out in a humidified atmosphere of 5% CO2, 7% O2, and 88% N2 at 39°C. At Days 3 and 5 of culture, 5% charcoal-stripped fetal bovine serum was added to the medium [22]. Cultures were maintained for 8 days, at which point embryos that had developed to the blastocyst stage were transferred to synchronized recipients. Those embryos that started to cleave but did not reach the blastocyst stage were fixed (acetic acid:ethanol, 3:1 [w/w]) and stained with aceto-lacmoid (1%) for the evaluation of cell number.
Nuclease Sensitivity Assay
To assess the effects of heat treatment on the global chromatin structure, we analyzed nuclease (DNase I [Boehringer-Mannheim, Roche Diagnostics S.p.A., Monza, Italy] and Micrococcal Nuclease [MNase; Amersham Pharmacia Biotech, Milan, Italy]) sensitivity of sheep granulosa cell nuclei (passage 5, three replicates); such analyses were also extended to sheep fetal fibroblasts (passage 4, three replicates) to confirm that the digestion patter observed was not exclusive of the type of cells used for cloning. Briefly, immediately after incubation in low-osmolarity SOF medium at 37°C or 55°C, cells were pelleted and lysed on ice for 2 min in a 0.3 M sucrose buffer containing 0.2% Nonidet P-40 (Sigma). Aliquots were then layered onto a 1.2 M sucrose buffer and centrifuged at 4°C. Nuclear pellets were resuspended in DNase I or MNase buffer at a concentration of 0.51 x 107 nuclei/ml. DNase I assays were performed in 0.3 M sucrose in 60 mM KCl, 15 mM NaCl, 5 mM MgCl2, 0.1 mM EGTA, 15 mM Tris-HCl (pH 7.5), and 0.5 mM dithiothreitol by adding different quantities of DNase I (grade 1) to 300-µl aliquots and incubating for 10 min at 25°C. The MNase digestions were performed in 300-µl aliquots at 100 U/ml of MNase for increasing periods of time at 37°C in 15 mM Tris-HCL (pH 7.5), 15 mM NaCl, 60 mM KCl, 0.15 mM B-mercapto-ethanol, 0.15 mM spermine, 0.15 mM spermidine, 0.34 M sucrose, 10 mM NaHSO3, and 1 mM CaCl2. Nuclease reactions were stopped by addition of an equal volume of 20 mM EDTA (pH 8.0) and 1.0% SDS. Proteinase K was added (to 200 µg/ml), and samples were incubated overnight at 37°C. Genomic DNA was extracted with phenol-chloroform, ethanol precipitated, and dissolved in TE buffer (10 mM Tris-HCL [pH 7.5] and 1 mM EDTA). The DNA samples were electrophoresed through 1.0% agarose gels in 1x TBE buffer and alkaline-blotted onto Hybond-N+ membrane (Amersham) and ultraviolet cross-linked (Stratalinker; Stratagene, Firenze, Italy). Hybridization with [32P]deoxycytidine triphosphate-labeled genomic sheep DNA and subsequent washing of membranes were performed as described elsewhere [23].
Genotyping of Microsatellite Markers
Total genomic DNA from the clones and the genetic mothers was prepared following the standard procedures from blood or from fragments of tongue in the case of dead animals. To confirm the clonal status of the newborns, individual identification and parentage diagnosis were performed with multiplex polymerase chain reaction using nine microsatellite markers (CP49, FCB11, AE129, FCB304, INRA063, MAF214, PZ963, CSRD247, and HSC). Based on the degree of polymorphism of these microsatellites in the breed used (Sarda), the probability that another animal randomly taken from the same population would have the same genotype was determined to be smaller that 2 x 10-11.
Genotypes were determined by polyacrylamide gel electrophoresis using an automated DNA sequencer (Perkin-Elmer, Bucks, U.K.) and were analyzed by Genscan and Genotyper software (Genscan Detection System, Woburn, MA).
Statistics
The experimental observations, expressed as percentage of embryonic development to the blastocyst stage, were processed by a one-way analysis of variance using the statistical package SPSS (SPSS, Inc., Chicago, IL) following the model
![]() |
i = effect caused by the treatment (i = 1 for control group, 2 for 50°C, 3 for 75°C, and 4 for fertilized), and
ij = casual effect of the error (0,
). | RESULTS |
|---|
|
|
|---|
As has been previously reported [3], only a small proportion of granulosa cells (1.5%) incorporated BrdU (Fig. 1, arrowheads) during the pulse period, indicating that almost all nuclei injected into the enucleated oocytes were in G0/G1 phase of the cell cycle.
|
DNase I and MNase Sensitivity Assays
Both the chemical and physical denaturation procedures rendered granulosa cells nonviable, as indicated by trypan blue staining and by the absence of subsequent growth in vitro. Cells heated to 75°C displayed the most extreme evidence of both nuclear and cytoplasmic disruption.
Because of technical difficulties in purifying intact nuclei from methanol and 75°C-treated cells, chromatin studies could not be performed on these experimental groups. As a general marker of chromatin structure, we determined the sensitivity of chromosomes in purified nuclei to the endonuclease DNase I, which, at the concentration used in the present study, normally digests different DNA sequences at comparable efficiencies [23]. Nuclei from granulosa cells and fibroblasts that had been incubated in SOF medium at either 37°C or 55°C were incubated at increasing concentrations of DNase I [23] in five separate replicates. Importantly, in the absence of DNase I (Fig. 2A, lane 1), the chromosomal content appeared fully intact in the 55°C-treated cells, with no sign of DNA degradation. At increasing concentrations of DNase I, both the 37°C- and 55°C-treated cells displayed a gradual increase in the degree of digestion by DNase I. However, chromatin of 55°C-treated cells was more sensitive than that of 37°C-treated cells. Hence, at the same concentrations of the enzyme, an approximately 2-fold higher degree of digestion was achieved in the 55°C-treated cells as compared to the 37°C-treated cells (Fig. 2A, lanes 57).
|
We next incubated nuclei from the 2 types of cells with MNase, an endonuclease that digests most readily the linker DNA between nucleosomes and, therefore, allows one to analyze the overall nucleosomal organization of chromatin; this assay was repeated twice. Both the control and the 55°C-treated cells displayed a canonical MNase digestion pattern, with bands corresponding to mononucleosomes, dinucleosomes, trinucleosomes, and so on. This established that the 55°C treatment had not altered the basic nucleosomal organization of chromatin. However, in both theDNase I and MNase digestions, we observed a higher degree of digestion in the 55°C-treated cells as compared to the control cells. This might indicate that heat treatment destabilizes high-order structures of chromatin, thereby facilitating access to the 30-nm chromatin fiber and, hence, digestion by exogenously added nucleases.
Nuclear Morphology in Fixed, Reconstructed Embryos
In the majority of reconstructed embryos (control, 55°C, 75°C, and methanol-fixed cells) that we fixed 1 h after injection, the chromosomes condensed within 1 h of injection (81 of 120 [68%]) (Fig. 3A), whereas unmodified nuclei were detected in the others (39 of 120 [33%]). Pronuclear organization was complete by 6 h after activation (Fig. 3B) in the following experimental groups (control, 38 of 45 [84%]; 55°C, 74 of 83, [90%]; 75°C, 45 of 51 [88%]). Remarkably, irregular pronuclei associated with condensed chromatin were observed in a higher proportion of oocytes injected with control nuclei as compared to those injected with heat-treated cells (untreated, 39%; 55°C, 16%; 75°C, 17%; P < 0.05). Embryos that were reconstructed using methanol-denatured nuclei also formed pronuclear-like structures in 65% of successfully injected oocytes (52 of 80). Moreover, 43% of these reconstructed embryos initiated cleavage (68 of 160), and 18% (12 of 68) progressed to the 12-cell stage on subsequent in vitro culture. However, nuclei were distinguished by large variations in size and were abnormally distributed within the embryo (Fig. 4, compare A and B), possibly suggesting that the centriole had lost the capacity to induce accurate sister-chromatid segregation. To restore centriolar function, fixed cells were coinjected with a sperm tail, but this resulted in poor survival of the oocytes. Therefore, chemical denaturation was discontinued, and all data presented in the remainder of this paper relate to the effects of physical denaturation of granulosa cells before nuclear transfer.
|
|
Developmental Capability of Cloned Embryos In Vitro and In Vivo
To test the developmental potential of nonviable cells, untreated and heat-denatured granulosa cells were injected into previously enucleated oocytes [21]. All the denaturation treatments resulted in a loss of cellular plasticity and necessitated the use of a large-bore pipette for injection (diameter of 15 µ as compared to 5 µ for fresh cells). Consequently, whereas the majority of oocytes injected with fresh granulosa cells survived intact, a significant proportion of oocytes injected with denatured cells underwent lysis. Only successfully injected cytoplasts were selected for further culture in SOF medium [22].
A similar proportion of reconstructed embryos initiated cleavage in all 3 experimental groups (fresh cells, 60%; both groups of heat-treated cells, 70%); however, the proportion of cloned embryos that reached the blastocyst stage tended to be higher in those reconstructed with heat-treated cells (Table 1). Control fertilized embryos consistently displayed a good viability, with 40% reaching the blastocyst stage after 7 days of culture (Table 1). Ninety blastocysts derived from all groups of cloned embryos and 52 IVF control embryos were transferred into synchronized recipient ewes, and pregnancies were monitored by high-resolution ultrasound scanning at 40, 60, and 80 days of gestation. Twenty recipient ewes that received fertilized embryos were pregnant at Day 40, and 18 of these ewes delivered 23 lambs (Table 2). Reconstructed embryos from all groups established pregnancy rates comparable to those with IVF embryos; however, fetal losses occurred in a high proportion of recipients after Day 40 and throughout the gestation. Although the small numbers of animals used preclude any statistical analysis, the fetal losses were comparatively less frequent in recipients carrying cloned embryos derived from heat-treated cells (Table 2). Two female lambs were obtained from untreated granulosa cells; these lambs were delivered by cesarean section after 148 days of gestation. One was dead at birth, whereas the other died within 24 h due to respiratory problems. Additionally, one was normal, whereas the other was affected by hydronephrosis. So far, we have produced 4 female lambs from heat-treated cells (Table 2). The first, named "Gavina" (Fig. 5A), weighed 4.8 kg, was delivered by cesarean section at Day 150 of pregnancy, and died unexpectedly at 15 days of age. A postmortem examination did not reveal any abnormalities. The second lamb of the 55°C-treatment group weighed 5.3 kg and was also delivered by cesarean section after 148 days of pregnancy; this lamb died after 24 h due to respiratory failure. A set of twins, weighing 4 and 3.6 kg and also derived from the 55°C-treated cells, was spontaneously delivered after 146 days of pregnancy. One died after 5 h due to pulmonary edema, whereas postnatal growth and development of the second twin has been normal as of this writing (Fig. 5B). No live offspring have yet been obtained from the 75°C-treated cells.
|
|
|
Apart from Gavina, which was produced after nuclear transfer of a granulosa cell taken from a pool of COCs collected at the abattoir, all other cloned lambs obtained so far have been derived from 2 granulosa cell donors: 3 lambs from ewe 5802, and 2 lambs from ewe 7207 (Fig. 6).
|
To ensure that development could not have resulted from a failure of enucleation, successful enucleation of each oocyte was confirmed by direct visualization of chromosomes (Hoechst 33342 staining) following aspiration of the metaphase plate. These procedures alone would preclude the possibility of parthenogenetic development, but we have previously demonstrated that parthenogenetic development of ovine embryos invariably arrests before Day 25 of gestation due to the absence of paternally expressed imprinted genes [24]. Moreover, microsatellite analysis confirmed unequivocally that the cloned lambs were isogenic with the granulosa cell donors (Fig. 6).
| DISCUSSION |
|---|
|
|
|---|
Unfortunately, despite the initial optimism, progress in improving somatic cell cloning has stalled, and the strategies proposed to optimize nuclear reprogramming, although scientifically sound [2, 3], do not significantly alter the outcome. An alternative approach, already applied to Xenopus, could be to reprogram quiescent somatic cells in vitro before nuclear transfer [29]. That paper reports the reactivation of Xenopus erythrocytes, permeabilized and pretreated with a series of enzymes, by incubating them in cytoplasmic extracts from meiotically arrested, unactivated eggs.
In the present study, we show that denatured somatic cells taken from adult sheep donors can be reactivated after nuclear transfer and develop to blastocysts in vitro and to viable offspring in vivo. The cloning pioneer, Marie DiBerardino [30], suggested in a recent review that for somatic cell cloning, ideally, only nuclear DNA, stripped of chromosomal proteins, and a functional centriole should be introduced into the egg cytoplasm. Our results represent the first evidence, to our knowledge, in support of this hypothesis in a mammalian model. Repressive nucleoprotein complexes could be destabilized by thermal or more efficient stimuli and chromatin remodeling factors applied in vitro before nuclear transfer. The feasibility of such an approach is supported by recent thermodynamic studies of protein-DNA interactions. For instance, the SOX family of high motility group (HMG) proteins, which are key regulators of embryonic development and are also involved in the regulation of cellular differentiation, start to unfold at greater than 46°C [31]; therefore, we can assume that even a mild heat treatment could displace it from the chromatin. Furthermore, dramatic conformational changes of high-order DNA-protein complexes were detected in interphase and metaphase chromosomes between 55°C and 75°C [32, 33]. Finally, the proteins of the nuclear matrix, which organize the DNA in inactive or active operational domains, are the most thermally labile proteins of the cells, undergoing denaturation at 4345°C [34]. Therefore, all these families of proteins are likely to have undergone denaturation under the thermal stress used in the present study.
Unfolded and denatured proteins are more readily degraded by proteolytic enzymes [35, 36]. Therefore, the oocyte proteolytic [37] and reprogramming machinery might more readily process nucleoprotein complexes, rendering the chromosomes more accessible to remodeling engines.
These experiments demonstrated that thermal treatment destabilizes high-order structures of chromatin without compromising nuclear reprogramming, as witnessed by the successful embryo development in vitro and by the relatively high postimplantation development of embryos reconstructed with heat-treated cells. Our results are in contrast with those of an analogous report in the mouse, in which no beneficial effect on development was observed in oocytes reconstructed with granulosa cells thermally stressed before nuclear transfer [38]. However, the method of nuclear transfer adopted in that study (electrofusion) prevented an evaluation of the effects exerted by the treatment, because any of the granulosa cells heated to relevant temperatures (
60°C) fused with the oocytes.
Although a slight improvement over current procedures, heat treatment did not invoke full and complete reprogramming, as evidenced by the fetal and postnatal losses observed in the present study. The failure rate reported here is similar to that described by others following the transfer of cloned embryos [2, 3, 3941], and it further confirms that the majority of fetal losses and postnatal mortality observed following transfer of cloned embryos is due to the incomplete reprogramming of transplanted nuclei.
It is essential to improve the success rate of generating nuclear transfer embryos from adult cells; such advances will ultimately increase the chances of isolating pluripotent stem cell lines for tissue engineering and transplantation medicine [42, 43]. The production of live mammalian offspring cloned from a nonviable cell, as reported in the present study, represents an essential step forward in the development of alternative strategies for nuclear reprogramming.
The growing body of knowledge regarding the molecular mechanism underlying genome remodeling [28, 44, 45], coupled with our demonstration that cell viability is no longer an absolute requisite for cloning, opens the possibility for use of unparalleled and more radical reprogramming strategies. For instance, thermally controlled conformational changes of high-order DNA-protein complexes associated with the serial addition of energy-driven, specific remodeling engines [45] might induce a complete reprogramming of permeabilized somatic cells/isolated nuclei in vitro, before being transplanted into enucleated eggs.
Freeze-dried spermatozoa retain their genetic integrity and are able to develop into healthy mice after ICSI (intracytoplasmatic sperm injection) [46]. We have shown, to our knowledge for the first time, that somatic cells rendered nonviable by heat are developmentally competent on nuclear transfer. These results clearly indicate that a contribution of structurally intact chromatin (and, of course, a functional centriole) is all that is required from a nuclear transfer donor cell. The next step should be to evaluate whether somatic cells stored in a freeze-dried state maintain their developmental competence. The possibility of storing freeze-dried cells in normal refrigerators (or, even better, at ambient temperature) would be of immense value for the establishment of genetic banks from endangered animal species, as has been recently proposed [47].
Although far from optimal, as indicated by the high rates of early and late fetal mortality, our results boost reprogramming approaches and offer alternative options to those aiming to use nuclear transfer to regenerate extinct species [48, 49].
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 Part of this work was supported by the Royal Society (to R.F., G.P., and P.L.). J.F. Jr. acknowledges the support from GACR 524/96/K162 and Mze NAZV EP 0960006200. Part of this work was covered by Es. Fin. 2001 (MIUR). ![]()
2 Correspondence: Pasqualino Loi, Dipartimento di Strutture, Funzioni, Patologie e Biotecnologie, Facoltà di Medicina Veterinaria, Piazza Aldo Moro 45, 64100 Teramo, Italy. FAX: 0861 558819; loi{at}ifv.vet.unite.it ![]()
Accepted: January 30, 2002.
Received: October 4, 2001.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Li and P. Mombaerts Nuclear Transfer-Mediated Rescue of the Nuclear Genome of Nonviable Mouse Cells Frozen Without Cryoprotectant Biol Reprod, October 1, 2008; 79(4): 588 - 593. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Shi, A. Hoeflich, H. Flaswinkel, M. Stojkovic, E. Wolf, and V. Zakhartchenko Induction of a Senescent-Like Phenotype Does Not Confer the Ability of Bovine Immortal Cells to Support the Development of Nuclear Transfer Embryos Biol Reprod, July 1, 2003; 69(1): 301 - 309. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |