If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This work was supported in part by National Institutes of Health Grants HD 12304 and HD 29968, Center Grants HD 33994 from Reproductive Biology and HD 02528 from Mental Retardation provided access to various core facilities.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The infertility phenotype of cyclooxygenase-2 (Cox-2)-deficient female mice establishes the important role of Cox-2 in pregnancy. Cox-2 deficiency results in defective ovulation, fertilization, implantation, and decidualization; the latter of which can be restored in part by the prostacyclin analog carbaprostacyclin. Uterine Cox-2 expression during early pregnancy shows distinct localization and kinetics in the uterine luminal epithelium and underlying stromal cells, suggesting that expression is tightly regulated. Several intracellular signaling cascades including ERK, p38, and JNK are implicated in vitro as critical components of regulated Cox-2 expression in response to mitogens, growth factors, and cytokines. We investigated the involvement of these signaling pathways during Cox-2 induction in vivo by monitoring uterine kinase activity after intraluminal application of a deciduogenic stimulus. Our results show that the ERK and p38 pathways are activated in uterine preparations as early as 5-min post-stimulation. ERK activation was sustained for several hours with a return to baseline levels by 4 h. p38 activation was rapid with a peak at 5-min post-stimulation and returned to near baseline levels after 45 min. Systemic administration of a MEK inhibitor completely inhibited ERK activation, but did not affect early (2 h) luminal epithelial or late (24 h) stromal Cox-2 expression and only modestly affected decidualization. In contrast, administration of a p38 inhibitor modestly inhibited early Cox-2 expression in the luminal epithelium, while dramatically diminishing late stromal expression. In parallel, induced stromal peroxisomal proliferator activated receptor-δ (PPARδ) expression is blunted by p38 inhibition. p38 inhibition also significantly inhibited decidualization. These results suggest that p38, but not ERK, activation is required for induced Cox-2 and PPARδ expression during decidualization. In addition, inhibition of p38 led to decreased decidualization suggesting that an intracrine prostanoid pathway consisting of Cox-2, prostacyclin, and PPARδ is required for maintenance of early pregnancy.
Prostaglandins (PGs)1are arachidonic acid-derived lipids that mediate pain, fever, and other symptoms associated with inflammation. PGs also contribute to cellular functions such as proliferation and differentiation (
). Two distinct Cox isozymes have been identified. Cox-1 is constitutively expressed and is reported to be critical for cytoprotective and homeostatic functions, whereas Cox-2 is inducible and is the major isozyme found in inflammatory and proliferating cell types (
) purified a novel form of cyclooxygenase from ovarian follicular granulosa cells that displayed properties distinct from the well studied seminal vesicle Cox enzyme. This novel Cox isozyme was hormonally inducible and required for ovulation. Its identity with Cox-2 has now been established (
). The infertility phenotype observed in Cox-2-deficient female mice further substantiates the important role of Cox-2 in early pregnancy. Indeed, Cox-2 deficiency affects all stages of early pregnancy leading to failure in ovulation, fertilization, implantation, and decidualization (
). Increasing levels of P4 from newly formed corpora lutea from day 3 of pregnancy direct stromal cell proliferation, which is further potentiated by ovarian estrogen secretion in the morning of day 4. In contrast, the luminal epithelium becomes differentiated on day 4 for interactions with the blastocyst during the attachment reaction, which occurs at 2200–2300 h on this day. The first conspicuous sign of the implantation process is an increased endometrial vascular permeability at the sites of blastocyst apposition (
). This coincides with the initial attachment reaction between the blastocyst trophectoderm and uterine luminal epithelium, resulting in the adherence and penetration by trophoblast cells through the underlying basement membrane. The attachment reaction is rapidly followed by proliferation and differentiation of stromal cells into decidual cells (decidualization).
The decidual reaction can also be elicited experimentally by the application of a mechanical trauma to the uterus or an intraluminal infusion of a small amount of oil in pseudopregnant or steroid hormonally prepared mice. This experimentally induced decidualization (deciduoma) has widely been used to study the events that occur during early pregnancy (
). On the morning of day 4 of pregnancy, uterine Cox-2 expression is undetectable in the uterine epithelium of pregnant or pseudopregnant mice. However, Cox-2 is expressed in the luminal epithelium and underlying stroma at the site of blastocyst apposition during the attachment reaction on day 4 night (
). In contrast, initiation of the uterine decidual reaction experimentally by intraluminal infusion of oil (a deciduogenic stimulus) at 1000 h on day 4 of pseudopregnancy results in specific accumulation of Cox-2 mRNA in the luminal epithelium of the oil-infused horn within 2 h followed by a marked decline to undetectable levels by 1800 h. Surprisingly, Cox-2 expression reappears at 2400 h post-stimulation, but its localization is in the uterine stroma. This is similar to the localization and kinetic profile observed for Cox-2 in the uterus during normal embryo implantation (
). Furthermore, the stromal Cox-2 expression observed during implantation and decidualization coincides in time and place with that of the nuclear hormone receptor, peroxisome proliferator-activated receptor-δ (PPARδ) (
). The restoration of dedicualization in Cox-2null mice by a prostacyclin analog cPGI (carbaprostacyclin) that binds PPARδ suggests that an intracrine Cox-2-prostacyclin-PPARδ signaling network exists in the uterus to support early pregnancy (
The unique kinetic and expression profiles observed for Cox-2 and PPARδ during early pregnancy suggests that expression of these two genes is tightly regulated. Multiple signaling cascades are thought to be involved in the regulated expression of Cox-2. The MAP kinase signaling cascades of ERK, JNK, and p38 as well as the NF-κB pathway have been implicated as critical components contributing to regulated Cox-2 expression in response to mitogens, growth factors, and cytokines (
). Similarly, the p38 MAP kinase cascade is also implicated in the induced expression of PPARγ during adipogenesis in vitro. The presence of either a dominant-negative mutant of p38 or a p38 inhibitor prevented adipocyte differentiation and impaired PPARγ expression (
). Thus, we studied the involvement of various signaling pathways leading to Cox-2 and PPARδ expression in the mouse uterus in vivo after intraluminal application of a deciduogenic stimulus. Our results show that p38, but not ERK, JNK, or NF-κB, activation is required for induced Cox-2 and PPARδ expression during the decidualization reaction in the uterus. In addition, inhibition of p38 leads to a decreased decidualization reaction, suggesting that p38 signaling is critical for establishing a receptive uterus during early pregnancy perhaps via regulation of uterine Cox-2 and PPARδ induction.
MATERIALS AND METHODS
Animals and Treatment Schedules
All of the mice (129/B6) were housed in the animal care facility at the University of Kansas Medical Center in accordance with National Institutes of Health standards for the care and use of experimental animals. Pseudopregnancy was produced by mating females with vasectomized males of the same strain (Day 1=vaginal plug). To induce decidual cell reaction (Fig.1), sesame oil (25 μl) was infused in one uterine horn on day 4 of pseudopregnancy; the non-infused contralateral horn served as a control. ERK inhibitor (SL327 or U0126) or p38 inhibitor (SB203580) was dissolved first in absolute ethanol followed by dilution in 40% polyethylene glycol. Mice were injected with specific inhibitors intraperitoneally 30 min before and/or various times after intraluminal infusion of oil. The control animals received only the vehicle. Mice were killed at different times to process infused and non-infused uterine horns for various assays. Uterine weights of the infused and non-infused horns were recorded on day 8 and the fold increases in uterine weights were used as an index of decidualization (
Extracts were prepared from uterine tissues at various times after oil infusion. Tissue was collected and homogenized in lysis buffer (1% Triton X-100, 1% deoxycholate, 10 mm Tris, pH 7.2, 150 mm NaCl, 1 mm sodium vanadate, 50 mm NaF, 1 mm phenylmethylsulfonyl fluoride, 50 μg/ml aprotinin, 50 μg/ml leupeptin). The samples were transferred to Eppendorf tubes and centrifuged at 14,000 × g for 15 min in the cold. The supernatants were transferred to fresh tubes, and protein concentrations were determined. These extracts were used for kinase assays and Western blot analysis.
Western Blot Analysis
Lysates (50 μg/sample) were analyzed by SDS-polyacrylamide gel electrophoresis on 12% Tris-glycine gels (Novex, San Diego, CA). Protein was electrotransferred to polyvinylidene difluoride membranes and blocked with a solution of phosphate-buffered saline containing 5% milk and 0.1% Tween 20. Blots were probed with a polyclonal phosphospecific antibody against ERK, p38, or IκB (New England BioLabs, Beverly, MA) or JNK (Promega, Madison, WI) followed by a peroxidase-conjugated goat anti-rabbit IgG polyclonal antibody (New England BioLabs). Washes following incubation with the antibodies were done in phosphate-buffered saline, 0.1% Tween 20. Bands were detected using chemiluminescent detection reagents (Pierce, Rockford, IL) and quantitated using a Molecular Dynamics personal densitometer.
MAPKAP Kinase-2 Assay
Lysates (100 μg/sample) were incubated with 10 μg of an anti-MAPKAP kinase-2 antibody (Upstate Biotechnology, Lake Placid, NY) on a rocking platform at 4 °C. After 3 h, 50 μl of a protein A/G-agarose slurry was added (Santa Cruz Biotechnology, Santa Cruz, CA), and tubes were rocked for another 1 h at 4 °C. Agarose beads were pelleted by centrifugation at 1500 × g for 5 min., washed three times with lysis buffer and washed once with 20 mm Hepes, pH 7.0 buffer. Immunoprecipitations were resuspended in 75 μl of kinase assay buffer (20 mm Hepes, pH 7.0, 5 mm 2-mercaptoethanol, 10 mm MgCl2, 0.1 mg/ml bovine serum albumin, containing 2 μg of hsp27 (Stressgen Biotechnology, Victoria, BC Canada). Kinase reactions were initiated by the addition of 10 μm ATP plus 10 μCi of [γ-33P]ATP (PerkinElmer Life Sciences) and incubated at 25 °C for 30 min. Reactions were terminated by the addition of Laemmli SDS sample buffer, boiled for 5 min, electrophoresed on a 12% Tris-glycine gel, dried, and quantitated using a Molecular Dynamics phosphorimager.
Polyclonal phosphospecific antibodies against ERK or p38 were used for immunohistochemistry. Immunolocalization was performed on paraformaldehyde-fixed frozen sections using a Zymed Laboratories Inc.-Histostain-SP kit (Zymed Laboratories Inc., San Francisco, CA) as described previously (
). Frozen sections (10 μm) were mounted onto poly-l-lysine-coated slides and fixed in 4% paraformaldehyde in phosphate-buffered saline for 10 min at 4 °C. After prehybridization, sections were hybridized at 45 °C for 4 h in 50% formamide buffer containing 35S-labeled sense or antisense cRNA probes for Cox-2 or PPARδ (
). After hybridization and washings, sections were incubated with RNase A (20 μg/ml) at 37 °C for 20 min, and RNase A-resistant hybrids were detected by autoradiography using Kodak NTB-2 liquid emulsion. Sections were post-stained with hematoxylin and eosin. Sections hybridized with sense probes served as negative controls.
All reagents not specifically described were of the highest quality commercially available. U0126, SL327 and SB203580 were provided by the DuPont Pharmaceuticals Chemical and Physical Sciences Department.
Oil-induced Decidualization in Mice Activates ERK, p38, and JNK, but Not NF-κB
Our first experiments were to determine which signaling pathways were activated in the uterus during decidualization. We monitored the activation status of intracellular signaling cascades in uterine extracts using phosphospecific antibodies to several MAP kinases along with IκB. As seen in Fig.2, ERK (panel A), p38 (panel B), and JNK (panel C) showed an increased phosphorylation state indicative of activation with the onset of an oil-induced decidual cell reaction. The ERK and p38 pathways were activated in uterine preparations as early as 5-min post-stimulation. ERK activation was sustained for several hours with a return to baseline levels by 4 h (data not shown).
The activation of p38 was rapid with a peak at 15-min post-stimulation and returning to near baseline levels after 45 min. JNK phosphorylation proceeded on a more relaxed time frame with elevated levels not observed until 45-min post-treatment. In contrast to the MAP kinase cascades, the level of uterine phosphorylated IκB (Fig. 2,panel D) remained unaffected by the application of the deciduogenic stimulus, suggesting that the NF-κB pathway is not activated under this condition.
ERK and p38 activation have been implicated in the induction of Cox-2 in inflammatory cell types. In fact, pharmacological inhibition of p38 by SB203580 or inhibition of ERK activation by the MEK inhibitor, U0126, blocks lipopolysaccharide-included Cox-2 expression in monocytes (
). Given the utility of these agents to specifically block individual kinase cascades, we employed these agents to investigate the role of p38 and ERK activation in the regulation of Cox-2 expression in the uterus.
Uterine Luminal Epithelial ERK Activation Is Blocked by MEK Inhibition
To test whether the MAP kinase inhibitors could influence the regulated expression of Cox-2 in vivo, we administered the selective kinase inhibitors systemically and monitored kinase activity in extracts of uteri isolated 15-min post-stimulation by intraluminal oil infusion. The results of these studies are shown in Fig. 3A. Administration of the MEK inhibitor, U0126, blocked ERK activation detected in the uterine extracts. Similarly, another MEK inhibitor, SL327, was just as or even more effective than U0126 in preventing uterine ERK activation by this treatment. The specificity of SL327 toward various kinase enzymes is shown in Table I along with data for U0126 for comparison. It can be seen that SL327 is a highly selective MEK1/2 inhibitor. The pharmacological properties of SL327 favor its use in systemic dosing paradigms,
Immunohistochemical localization of activated phospho-ERK in the uterus following oil stimulation is shown in Fig. 3B (panels a and b). Clearly, phospho-ERK was detected primarily in the uterine glandular and luminal epithelia 15 min after oil treatment. Critical for our further studies was the demonstration that this specific staining was markedly abrogated by SL327 (Fig.3B, panels c and d). Collectively, these results indicate that SL327 effectively blocks ERK activation in the uterus at the onset of the decidual reaction.
Uterine Stromal p38 Is Inhibited by SB203580
We next turned our attention to the p38 MAP kinase pathway. p38 is known to activate a number of substrates through phosphorylation. MAPKAP kinase is a p38 substrate that undergoes phosphorylation by p38 subsequent to its activation (
). Thus, the phosphorylation status and kinase activity of MAPKAP kinase have been used to monitor the activity of p38 under various physiological states. We monitored MAPKAP kinase activity and p38 phosphorylation status as a measure of p38 activity in the uterus following intraluminal oil infusion. The results of these studies are shown in Fig. 4. MAPKAP kinase activity measured in the extracts of uteri isolated 15-min post-stimulation with oil was elevated over control (non-infused uterine lysates) levels. Systemic administration of SB203580 blocked the activation of MAPKAP kinase activity indicating that p38 was inhibited (Fig. 4A). Immunohistochemical localization of activated phospho-p38 in the uterus is shown in Fig. 4B. Immunostaining was primarily localized in the glandular epithelium in the non-infused uterine sections. However, with the initiation of decidual cell reaction by intraluminal oil infusion, distinct accumulation of activated p38 was noted in the uterine stroma in addition to the glandular epithelium by 15 min after oil infusion. The specific staining in the stroma was completely abrogated by SB203580. This is consistent with previous findings showing that SB203580 can inhibit stimulus-induced phosphorylation of p38 in addition to blocking its kinase activity (
). Collectively, these results indicate that SB203580 effectively inhibits p38 in the uterus during the initiation of decidualization.
Uterine Stromal Cox-2 and PPARδ Expression Are Blocked by p38, but Not MEK Inhibition
Our next series of experiments were designed to test which of the kinase inhibitors effectively blocks the expression of Cox-2 and PPARδ in the uterus during decidualization. Previously, we have shown that Cox-2 and PPARδ expression display a unique temporal and spatial profile in the uterus after administration of a deciduogenic stimulus. Consistent with these reports (
), the present data show that Cox-2 expression is restricted to the uterine luminal epithelium at early (2 h) time points following oil infusion and to the stroma at later (24 h) time points (Fig.5). Administration of the MEK inhibitor, SL327, did not affect the expression pattern of Cox-2 in the uterus. In contrast, administration of the p38 inhibitor, SB203580, dramatically abrogated the late stromal Cox-2 expression, whereas this inhibitor did not affect the early Cox-2 expression in the luminal epithelium (Fig.5). PPARδ expression, which is found exclusively in the uterine stroma 24-h post-oil infusion (Fig. 6), was also completely prevented by the p38 inhibitor, SB203580. SL327 did not affect PPARδ expression (data not shown).
p38 Inhibition with SB203580 Blocks Uterine Decidualization
Our final series of studies investigated the effects of MEK or p38 inhibition on decidualization in the mouse. The details of this protocol are outlined in Fig. 1. Animals were treated with either SL327 or SB203580 once or twice at various doses on day 4 of pseudopregnancy prior to and/or after application of the decidualizing stimulus (oil infusion). Mice were scored for decidual response (uterine weights) on day 8. The results are shown in Fig.7. SL327 did not affect decidualization regardless of the dosing paradigm. On the other hand, SB203580 showed modest effects when administered 30 min prior to oil infusion used to initiate the decidual cell reaction. However, administration of this inhibitor 30 min prior to intraluminal oil infusion followed by a second administration of the drug at 6-h post-oil infusion produced a significant diminution in decidual response when scored 4 days later (Fig. 7). These results indicate that p38 inhibition at relatively early time points during the initiation of the decidualization reaction can dramatically influence the later course of this response.
Although induction of Cox-2 is critical to establishing and maintaining early pregnancy, the underlying biochemical events leading to expression of this gene product during pregnancy have not been well characterized. This is in contrast to the study of Cox-2 up-regulation during inflammation where a network of signaling molecules have been defined as critical for the regulated expression of Cox-2 in response to inflammatory stimuli (
). Specifically, involvement of the MAP kinase pathways have been demonstrated in Cox-2 activation by a number of stimuli including cytokines, growth factors, and mitogens. These pathways transduce extracellular signals to the nucleus where cis-acting regulatory factors control Cox-2 gene transcription. Analysis of the Cox-2 gene have revealed that specific response elements for NF-κB, NF-IL6 (C/EBP), ATF/CRE, and E-box binding proteins are present in the Cox-2 promoter (
). Thus, activation of a number of signaling pathways and transcription factors contribute to the regulation of Cox-2gene transcription after exposure to inflammatory stimuli. We sought to determine whether these same factors regulating induced Cox-2 expression observed during inflammation were applicable to the induction of Cox-2 in the mouse uterus during the early phase of embryo implantation, which is considered a “proinflammatory” response.
We investigated the response of the MAP kinase pathways in the mouse uterus in response to a deciduogenic stimulus and their relationship to Cox-2 expression. A robust in vivo inductive response of the MAP kinase pathways was found in the uterus following exposure to a deciduogenic stimulus. The cascades affected included ERK, p38, and JNK, whereas the NF-κB pathway was not significantly induced during decidualization. The kinetics of MAP kinase pathway activation are consistent with the findings in other systems where a rapid increase in kinase activity is observed in response to an external stimulus (
). The rapid increase in MAP kinase pathway activation fits with a proposed involvement of these cascades as mediators of the signaling events that control the biochemical processes of early pregnancy. Indeed, further support for this hypothesis as it relates to Cox-2 induction and decidualization was obtained from the location and kinetics of activation of these pathways in specific regions of the uterus.
Our results demonstrate that the ERK pathway is specifically activated in the uterine glandular and luminal epithelia within minutes following oil stimulation. In contrast, the p38 pathway is rapidly induced in the uterine stroma and glandular epithelium following the same stimulus. These findings are remarkable in that the signaling pathways are differentially regulated within the uterus in juxtaposed cellular populations experiencing an identical input. Although unique cellular responses to extracellular signals are known to preferentially induce specific signaling cascades in isolated cell populations in vitro, our present observations represent a unique demonstration of such control in vivo. Indeed, the cell-specific pathway utilization perhaps specifies the unique spatial and temporal expression profile seen for uterine Cox-2 and PPARδ during early pregnancy. This notion is supported by results obtained with specific pharmacological inhibitors of the individual kinase pathways.
Exposure of the uterus to the MEK inhibitor, SL327, completely abrogated ERK activation in the luminal epithelium with the onset of the decidual reaction. Similarly exposure to the p38 inhibitor, SB203580, completely blocked the ability of p38 to phosphorylate its downstream substrate MAPKAP kinase 2 in the stroma. We had previously shown that Cox-2 displays a unique expression profile, which includes early luminal expression, followed later by its expression in the stroma (
). Based upon these findings, we anticipated differential kinase cascade involvement regulating Cox-2 expression in the uterus. To our surprise, this was not the case. The consequence of MEK inhibition upon expression of Cox-2 was insignificant when studied at either 2- or 24-h post-oil stimulation. In contrast, exposure to the p38 inhibitor resulted in selective blockade of Cox-2 expression. The luminal expression at 2 h was unaffected, which is consistent with the lack of activation of p38 in the uterine luminal epithelium. The late stromal Cox-2 expression, however, was completely abrogated by p38 inhibition, supporting the hypothesis that p38 is a critical signaling component that regulates Cox-2 expression in the uterus.
We have previously demonstrated that Cox-2 and PPARδ expression coincide in the uterine stroma during the initiation of implantation (2300 h) on day 4 of pregnancy (
). Coordinate expression of Cox-2 and PPARδ in stromal cells suggests a possible signaling cascade within uterine cells that regulates the implantation process during early pregnancy. Remarkably, uterine stromal PPARδ expression was also blocked by p38 inhibition. This finding suggests that the coordinate expression of Cox-2 and PPARδ is further connected at the level of intracellular signaling pathway involvement and potentially transcription factor utilization. The p38 MAP kinase cascade is implicated in the controlled expression of other PPARs. During adipocyte differentiation, SB203580 prevents 3T3-L1 fibroblasts from differentiating into adipocytes after treatment with an adipogenic stimulus (
). The expression of PPARγ, a gene that is critical for adipocyte differentiation, is prevented by p38 inhibition. The induction of PPARγ in this system is dependent upon a phosphorylated form of the transcription factor C/EBPβ. C/EBPβ has also been shown to be critical for Cox-2 induction in granulosa cells prior to ovulation (
). C/EBPβ is one of three C/EBP family members that are induced during adipogenesis. Sequence alignment of the C/EBP family members reveals that a p38 kinase recognition motif is conserved among all members. Although we did not examine the involvement of C/EBP in uterine induction of Cox-2 and PPARδ, we speculate that members of this transcription factor family may be involved in the coordinate expression of Cox-2 and PPARδ. Also, in a preliminary experiment, we have found that the protein synthesis inhibitor cycloheximide can prevent Cox-2 gene expression at 24 h following oil treatment when administered during the 12–18-h window following induction.
P. A. Scherle, S. K. Dey, and J. M. Trzaskos, unpublished observations.
This would suggest that synthesis of another protein is critical forCox-2 gene expression in the uterus after receiving a deciduogenic stimulus. This factor could be C/EBPβ or another C/EBP family member.
Finally, exposing mice to the p38 inhibitor, SB203580, leads to attenuated uterine decidualization, a process that takes 4 days to develop fully in the mouse. This retarded development of decidualization as recorded on day 8 was observed after only a limited treatment with the drug at the time of the initial application of the deciduogenic stimulus on day 4 of pseudopregnancy. These results imply that interruption of key signaling events that occur in response to initiating deciduogenic signals can dramatically impede the process of early pregnancy. We would propose that the effects seen with p38 inhibition are due in part to the inhibition of Cox-2 and PPARδ induction. A model for this process is presented in Fig.8. We propose that a deciduogenic stimulus initiates a signaling cascade in the uterus that includes p38 activation. Downstream factors including ATF, CREB, and C/EBP undergo activation that drives induction of Cox-2 and PPARδ in addition to prostacyclin synthase (
). These three gene products establish an intracrine signaling network in uterine stromal cells that involves the conversion of arachidonic acid to prostacyclin, an endogenous ligand for PPARδ. The ligand-bound nuclear receptor complex would then control expression of a number of gene products that characterize the deciduogenic phenotype. The induction of this network is critical for implantation and maintaining early pregnancy and is self-sustained until other factors lead to the controlled repression of the system. The validity of this model awaits further experimentation.
The authors thank Janice I. Hytak, A. Christine Tabaka, James S. Piecara, Michael X. Sa, Jia-Sheng Yan, and Christopher A. Teleha for synthesis of U0126, SL327, and SB203580 used in this studies.
Inflammation: Basic Principles and Clinical Correlates. 2nd Ed. Raven Press, Ltd.,
New York, NY1992: 1127-1137