Comparative antiviral efficacies of cidofovir, trifluridine, and acyclovir in the HSV-1 rabbit keratitis model.

PURPOSE: To determine the relative antiviral inhibitory activity of topical 1% and 0.5% cidofovir, topical trifluridine (Viroptic; Burroughs-Wellcome, Research Triangle Park, NC), and topical acyclovir (Zovirax; The Wellcome Foundation, London, UK) during a 7-day period for the treatment of herpes simplex virus type 1 (HSV-1) keratitis and HSV-1 replication in the New Zealand rabbit ocular model. METHODS: In a series of four experiments using a two-eye design, a total of 80 New Zealand rabbits were inoculated in both eyes with HSV-1 McKrae after epithelial scarification. Forty-eight hours after inoculation, the rabbits were randomly assigned to a treatment group. Five treatment groups (16 rabbits/group) were evaluated: I, 1% cidofovir, twice daily for 7 days; II, 0.5% cidofovir, twice daily for 7 days; III, 3% acyclovir ointment, five times daily for 7 days; IV, 1% trifluridine, nine times daily for 3 days, then 4 times daily for 4 days; and V, control vehicle twice daily for 7 days. HSV-1 dendritic keratitis was graded in a masked fashion by slit-lamp examination on days 2, 3, 5, 7, 9, 11, and 14. Ocular viral cultures were obtained after slit-lamp examination on days 1, 3, 5, 7, 9, 11, and 14. RESULTS: Compared with the control group, all four treatment groups demonstrated significantly lower viral titers, fewer HSV-1-positive eyes/total during the treatment period, lower keratitis scores, fewer eyes with keratitis/total, and a shorter time to resolution of keratitis. Within the treatment groups, the 1% and 0.5% cidofovir treatments were significantly more effective than acyclovir and trifluridine as measured by the previous viral and keratitis parameters. CONCLUSIONS: Topical 1% and 0.5% cidofovir both appeared to be significantly more efficacious than topical trifluridine and acyclovir, during a 7-day course, in the treatment of experimental HSV-1 ocular disease in the New Zealand rabbit keratitis model.

Contrast-enhanced magnetic resonance imaging confirmation of an anterior protein pathway in normal rabbit eyes.

PURPOSE: Contrast-enhanced proton magnetic resonance imaging (1H MRI) has been used as a quantitative, noninvasive method to corroborate a pathway for the diffusion of plasma-derived protein into the aqueous humor in the normal rabbit eye. METHODS: T1-weighted magnetic resonance images were produced over 1- to 3-hour periods after the intravenous injection of gadolinium diethylenetriamine-pentaacetic acid. RESULTS: Analysis of the images yielded the time dependence of signal enhancements within the areas of interest. The ciliary body showed an immediate sharp increase, followed by a gradual decrease in signal enhancement with time. Although a gradual increase in signal enhancement was found in the anterior chamber, no significant change occurred in the posterior chamber. A similar MRI experiment with an owl monkey produced parallel, though smaller, signal enhancements in the ciliary body and anterior chamber. Again, however, no significant change was found in the posterior chamber. CONCLUSIONS: These results support and extend those of recent fluorophotometric, tracer-localization, and modeling studies demonstrating that in the normal rabbit and monkey eye, plasma-derived proteins bypass the posterior chamber, entering the anterior chamber directly via the iris root.

The relationship between aqueous humor flow and anterior chamber protein concentration in rabbits.

PURPOSE: Protein concentration in the anterior chamber of noninflamed eyes is determined by three factors: the rate of protein entry into the aqueous humor (AH), the removal rate by bulk flow of AH, and the anterior chamber volume. On the basis of observations by previous investigators and the authors' computational modeling, it was hypothesized that a direct reciprocal relationship exists between aqueous flare and AH flow. This relationship was studied in pigmented rabbits, under several conditions, to determine the validity of the hypothesis that changes in aqueous flare intensity reflect variations in AH flow rate. METHODS: Aqueous flare and AH flow were measured in rabbits entrained to a 12 hr light: 12 hr dark cycle, starting at 6 AM, which subsequently was phase-shifted 6 hr earlier to allow measurements over the light-to-dark transition period. AH flow rates were determined fluorophotometrically using the clearance method after corneal deposition of fluorescein. A Kowa FC1000 cell/flare meter was used to measure aqueous flare. Predictions of anterior chamber protein concentration were made using a computer model of plasma-derived protein diffusion into AH. The response to changes primarily of AH flow were evaluated by concurrently determining flare and flow before and after administration of intravenous acetazolamide. RESULTS: Aqueous flare decreased 40% during constant light and 47% over the light-dark transition in a parallel, monophasic manner. AH flow did not change significantly under either condition. In contrast, acetazolamide-induced changes in AH flow resulted in reciprocal changes in flare that could be simulated with the computer model through flow parameter changes alone. CONCLUSIONS: Anterior chamber protein concentration in rabbits appears to be modulated both by factors affecting AH flow and protein entry into the aqueous. Thus, in rabbits, changes in aqueous flare do not necessarily reflect AH flow changes.

Morphologic correlations with fluorophotometric data from monkey eyes with anterior uveitis.

Acute anterior uveitis was induced in monkeys by unilateral intravitreal injection of 1.0 ng of Escherichia coli endotoxin. Twenty-four hours later, each animal received an intravenous injection of 250 mg/kg body weight of fluoresceinated horseradish peroxidase (F-HRP), and fluorophotometric measurements were taken for 90 min. The animals were killed, and both eyes were processed for HRP demonstration. In the anterior chamber aqueous humor of normal control eyes, F-HRP concentrations were less than 0.002 mg/ml at 90 min. The F-HRP concentration was elevated consistently in the endotoxin-injected eyes; however, the magnitude of the effect varied. By fluorophotometry, inflamed eyes fell into two distinct groups. At 90 min, most had an anterior chamber F-HRP concentration of 0.014-0.06 mg/ml, although others had 0.39 mg/ml. In the latter group, an appreciably shorter latency was observed between the time of tracer injection and its detection in the anterior chamber. Aqueous humor protein concentrations, although highest in the most F-HRP-permeable eyes, followed more of a continuum in their distribution and identified less clearly the subpopulations seen by fluorophotometry. Normal eyes had no tracer leakage across either the ciliary epithelial or iris vascular endothelial barriers. All inflamed eyes had HRP leakage across the ciliary epithelium, but the subpopulation of eyes with shorter latencies and higher F-HRP concentrations by fluorophotometry also had iris vascular leakage.

The source of protein in the aqueous humor of the normal monkey eye.

In vivo aqueous fluorophotometry, morphology, and computational modeling were combined to examine the source of protein and the pathway by which protein enters the aqueous humor of monkeys. A computational model was developed to determine the likelihood of a diffusional route for delivering plasma proteins from ciliary body capillaries via the iris to anterior chamber aqueous humor, bypassing the posterior chamber. Model predictions were compared to aqueous fluorophotometric data obtained from monkeys following a single intravenous injection of fluoresceinated horseradish peroxidase (F-HRP, 250 mg/kg body mass). Model predictions of the magnitude and time course of anterior chamber F-HRP concentration agree with the fluorophotometric measurements. For example, of anterior chamber F-HRP concentration as a percentage of initial plasma F-HRP concentration at 90 min and 180 min post-injection was predicted to be 0.02% and 0.05%, respectively, and was measured to be 0.01-0.03% and 0.03-0.06%, respectively. In addition, model predictions in the case of a constant plasma protein level also are consistent with experimental data. The steady-state anterior chamber total protein concentration as a percentage of plasma protein concentration was predicted to be 0.2% and was assayed to be 0.05-0.2%. As in our previous study of the normal rabbit eye, morphologic and tracer localization evidence combined with the good agreement between model predictions and experimental data lead to the conclusion that a significant amount of the plasma protein normally present in monkey aqueous humor originates in ciliary body capillaries and diffuses anteriorly through the iris and into the anterior chamber.

Adrenergic mechanism in circadian elevation of intraocular pressure in rabbits.

Stimulation of the ocular sympathetic nerves is essential for the circadian elevation of intraocular pressure (IOP) in rabbits. Adrenergic mechanisms that participate in this elevation of IOP around the onset of darkness were investigated using selective adrenergic agents. Unilateral topical administration of 0.0001-0.1% prazosin, an alpha-1-adrenergic antagonist, at the onset of darkness caused a dose-dependent reduction of IOP elevation. After treatment with 0.1% prazosin, the concentration of norepinephrine (NE) in aqueous humor was not changed, but the aqueous humor protein concentration was reduced. The increase of aqueous flow, determined by fluorophotometry, after the onset of darkness was not affected by 0.1% prazosin treatment. Treatment of rauwolscine, an alpha-2-adrenergic antagonist, or timolol, a beta-adrenergic antagonist, was ineffective in reducing the circadian elevation of IOP. Apraclonidine, an alpha-2-adrenergic agonist, with concentrations of 0.0001-1% caused a dose-dependent reduction of IOP elevation. Treatment with 1% apraclonidine caused a 70% reduction of aqueous humor NE, a significant attenuation of the increase of aqueous flow, and no change of aqueous humor protein concentration. These results suggest that both an increase of aqueous outflow resistance (by alpha-1-adrenergic receptors) and an increase of aqueous flow (not exclusively by beta-adrenergic receptors) contribute to the circadian elevation of IOP. Prejunctional alpha-2-adrenergic receptors may serve as an autoregulating mechanism to limit the excess release of NE and hypertensive spike in IOP.

Chronic electrical stimulation of sympathetic nerves: effects on blood-aqueous barrier.

The effect of electrical stimulation of the sympathetic nerves on the blood-aqueous barrier was investigated in rabbits. The permeability of the barrier was assessed during either acute, chronic or following chronic nerve stimulation. During acute and chronic stimulation of the sympathetic nerves, fluorescein entered the anterior chamber at a rate significantly slower than in control eyes. After chronic stimulation, both the rate of entry of fluorescein and the aqueous humor protein concentration were much greater than in control eyes indicating breakdown of the blood-aqueous barrier. Treatment with the non-steroidal anti-inflammatory drugs, indomethacin and suprofen, completely blocked the breakdown of the blood-aqueous barrier. These results indicate that sympathetic nerve stimulation can cause the local synthesis of prostaglandins and that these can affect the blood-aqueous barrier.

Background-protein effects on fluorophotometric data.

Fluorescent tracers are commonly used in fluorophotometric studies of ocular fluids and tissues that contain background protein. Background-protein concentrations were found to decrease or increase significantly the measure of fluorescence emitted from solutions containing sodium fluorescein, fluorescein-labeled dextran, or fluorescein-labeled horseradish peroxidase. The effect of background protein on fluorescence was expressed as a function of the specific fluorescent tracer, tracer concentration, and background-protein concentration; it can be corrected in the analysis of fluorophotometric data. Fluorophotometric studies--particularly those in which the background-protein level is expected to be abnormally high, such as postoperative and pathologic studies--may need to include either a data correction based on measured effects of background protein on tracer fluorescence or, in the case of clinical investigations, recognize at least the potential for a range of possible interpretations.

The source of proteins in the aqueous humor of the normal rabbit.

Aqueous fluorophotometric, tracer localization and modeling methods were combined to document the existence of a pathway in the normal rabbit for the diffusion of proteins from the ciliary and iridial process stromas through the iris stroma into the aqueous humor of the anterior chamber. A new custom-conjugated tracer, fluoresceinated horseradish peroxidase (F-HRP), was used. Anesthetized rabbits were injected intravenously with F-HRP (250 mg/kg). In some animals, aqueous fluorophotometric and tracer localization studies were performed on the same eyes. Anterior chamber fluorescence was detected 2-10 min post-injection and rose to concentrations of 0.01-0.05 mg/ml 60 min post-injection. Subsequent tracer localization studies of these eyes revealed that the morphologic components of the blood-aqueous barrier were intact, that is, no leakage of F-HRP from the iris vasculature or across the ciliary epithelium was observed. Separate tracer localization studies were performed to examine the time course of the route(s) by which tracer entered the anterior chamber. These studies revealed a "wave" of tracer that migrated from the ciliary and iridial process stromas, through the iris, and arrived at the anterior iris surface approximately 8 min post-injection. A pharmacokinetic model based on the diffusional pathway was developed to describe the time course of the concentration of plasma macromolecules in the ciliary body, iris and anterior chamber. Model predictions were consistent with aqueous fluorophotometric and tracer localization results. The diffusion model can account for a major fraction of protein entering the aqueous humor of normal rabbit eyes.

Thyrotropin releasing hormone increases intraocular pressure. Mechanism of action.

Intravenous injections of 1-100 micrograms thyrotropin releasing hormone (TRH) in rabbits elevated intraocular pressure (IOP). The 2-5 mm Hg increase of IOP lasted for less than 2 hr. No change of pupil size was observed. This IOP elevation was not due to a direct effect of TRH on ocular tissues since intravitreal injections of 0.1 and 1 micrograms TRH did not change IOP. Concentrations of thyroid stimulating hormone (TSH), triiodothyronine (T-3), epinephrine (Epi) and norepinephrine (NE) in the plasma were elevated at 30 min after an i.v. injection of 10 micrograms TRH. Plasma levels of prolactin and thyroxine were not changed. Bolus i.v. injections of 0.1-1 micrograms TSH and 0.1-1 micrograms T-3, which would produce an equivalent increase of relevant hormones in the circulation, did not increase IOP. However, similar i.v. injections of 10-100 ng Epi and 100 ng NE caused a 1.5-3 mm Hg IOP elevation for 15-30 min. Thus, the IOP elevation following TRH administration probably is caused by the increase of circulating endogenous catecholamines and not by the stimulation of the TSH-thyroid hormone axis. Heart rate, but not blood pressure, was increased with 10 micrograms TRH. After unilateral transection of the cervical sympathetic trunk, the IOP elevation in the decentralized eye was larger than that in the intact eye. Topical treatment of 0.1% or 1% timolol in the decentralized eye inhibited the IOP elevations in both eyes, but 0.1% prazosin was not effective. Topical 1% atropine and atropine given subcutaneously at 0.6 mg/kg decreased the bilateral IOP elevations. These observations indicate that beta-adrenergic and muscarinic mechanisms, not an alpha-1-adrenergic mechanism, are involved.

Aqueous humor flow measured with fluorophotometry in timolol-treated primates.

Anterior chamber aqueous humor flow rate was measured in unanesthetized owl monkeys using fluorophotometry and anterior chamber photogrammetry. The mean anterior chamber turnover constant (ko) was 0.01, the mean aqueous humor flow rate was 2.75 microliters min-1, and the mean anterior chamber volume was 317 microliters in 16 eyes of eight monkeys. A significantly lower (11%, P less than 0.002) flow rate was measured in the afternoon compared to flow rates measured in the morning. This diurnal cycle is analogous to the fluctuations in flow in humans and supports the hypothesis that aqueous humor formation in primates is regulated by endogenous mechanisms. To investigate the role of adrenergic mechanisms in regulating flow, the dose-response effect of topical timolol (0.5-100 micrograms) was measured. Five microliters drops were used to minimize systemic absorption. Lower concentrations (0.01% and 0.05%) caused sporadic changes in flow. Topical 0.1% timolol (5 micrograms) significantly decreased aqueous humor flow in the treated eye compared to baseline flow while flow in the contralateral eye was not significantly different from baseline flow. Higher concentrations of timolol (0.5% and 1%) caused a dose-related bilateral decrease in flow. These results indicate that even with small topical volumes systemic absorption of the higher concentrations of timolol occurs. Significant local inhibition of flow occurred following 5 micrograms of topical timolol whereas an equal bilateral decrease in flow occurred following a 100 microgram dose. These results suggest that the standard clinical dose (approximately equal to 100 micrograms) is supramaximal for decreasing aqueous flow.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of cyclocryotherapy on aqueous humor dynamics in cats.

The effects of graded cyclocryotherapy were studied using pneumatonometry, fluorophotometry, and light and electron microscopy. Cats were treated with either 90 degrees, 180 degrees, or 270 degrees of cyclocryotherapy and were followed up for six weeks. The response of the eyes to treatment was evaluated in terms of intraocular pressure (IOP) response, aqueous humor flow rate, and permeability of the blood-aqueous barrier compared with the untreated fellow eye. The average IOP response was determined to be -20%, -44%, and -47% for the 90 degrees, 180 degrees, and 270 degrees treated eyes, respectively. The aqueous humor flow rate was estimated to be -14%, -44%, and -52% for each of the groups, respectively. Finally, the permeability of the blood-aqueous barrier, measured 60 minutes after injection of fluorescein, was estimated to be a percent increase of 22%, 332%, and 285% for each of the groups, respectively. Histologically, the maximally treated eyes (270 degrees) demonstrated greater disruption of the ciliary body compared with the control eyes. Thus, graded cyclocryotherapy causes graded destruction of the ciliary epithelium and proportionally related changes in IOP and aqueous humor dynamics.

Effects of stimulation of the ocular sympathetic nerves on IOP and aqueous humor flow.

Ocular sympathetic nerves were stimulated chronically in awake rabbits using electrodes unilaterally implanted on the cervical sympathetic trunk. IOP was measured by pneumatonometry and aqueous inflow was measured by fluorophotometry. In each animal, continuous trains of 1 msec pulses were delivered by means of a portable electrical stimulator. Experiments were spaced by 1 week recovery periods. Stimulation was varied over a range of amplitudes (5-15 V) and frequencies (3-12 Hz). Continuous sympathetic stimulation produced an immediate sharp decrease in IOP followed by a gradual rise to pre-stimulation values which were attained 60-90 min after onset. A rebound increase in IOP occurred when stimulation was terminated. The magnitude of the initial IOP drop, the delay in the return to pre-stimulation IOP, and the rebound rise in IOP subsequent to termination of electrical stimulation were proportional to the stimulation frequency. Maximal effects were observed at 12 Hz, and stimulation with 8-10 Hz for 180 min caused a sustained reduction in anterior chamber aqueous humor flow. Topical 2% phentolamine 1 hr before stimulation markedly reduced IOP and abolished the acute IOP changes observed in untreated stimulated animals. Topical 1% timolol did not affect either the initial IOP drop or the rebound; however, the IOP recovered during stimulation to values greater than pre-stimulation IOP. We conclude that in rabbits the beta-adrenergic effect of prolonged sympathetic nerve stimulation is to decrease aqueous flow. Chronic electrical stimulation in awake animals provides an experimental model for studying the role of the ocular sympathetic nerves.

Adrenergic stimulation of ciliary process epithelium causes surface membrane internalization.

An ultrastructural change induced in the nonpigmented epithelium (NPE) of the ciliary processes by adrenergic stimulation in the albino rabbit was studied. Thirty min after topical treatment with 2% isoproterenol, an extensive intracellular membranous network, previously reported to be smooth endoplasmic reticulum, was revealed by electron microscopy. It was postulated that this network originated from the plasma membrane. Using cationized ferritin (CF) as an ultrastructural tracer, freshly isolated anterior segments were incubated in buffer containing 10(-5) M isoproterenol and 0.2% CF. As early as 10 min, and for at least 30 min, the isoproterenol-treated NPE cells contained a membranous network that was morphologically similar to that which occurs in vivo. CF particles were present within the network, indicating that the membranous network had originated at the cell surface. This labeling was prevented by pretreatment with the beta-adrenergic antagonist timolol maleate. In both treated and control ciliary processes, CF was present in the ciliary canals between the NPE and the underlying pigmented epithelium after 10 min incubation. This suggests that the NPE is able to transport CF from its basilar to apical surface. These experiments imply that the NPE is able to internalize rapidly large amounts of plasma membrane in response to adrenergic stimulation. This response may be part of the mechanism of adrenergic receptor desensitization, alteration of aqueous humor production, or another adrenergic response.

The effects of forskolin on cyclic AMP, intraocular pressure and aqueous humor formation in rabbits.

Forskolin was used to study cyclic AMP-mediated regulation of aqueous humor dynamics in rabbits. Crystalline forskolin was solubilized in oil and its pharmacological effects were studied both in vitro and following topical ocular administration. In vitro, using cultured corneal epithelial cells, forskolin rapidly stimulated cyclic AMP production and in vivo increased cyclic AMP concentration in the aqueous humor 10-fold following topical administration. The effect of topical forskolin on intraocular pressure and aqueous humor formation was determined in vivo using pneumatonometry and fluorophotometry, respectively. Forskolin caused a prolonged reduction of intraocular pressure and decreased aqueous humor formation. The ability of forskolin to potentiate the ocular hypotensive effect of epinephrine was investigated. Forskolin in combination with epinephrine caused a decrease in intraocular pressure of longer duration than either 0.1% epinephrine or 1% forskolin administered separately. Forskolin caused a small but significant increase in the permeability of the blood-aqueous barrier at the time of maximal intraocular pressure reduction. This effect on the blood-aqueous barrier may explain the inhibitory effect of forskolin on aqueous humor formation.

Rabbit corneal endothelial cells in vitro: effects of EGF.

Rabbit corneal endothelial cells were grown in tissue culture. Epidermal growth factor (EGF) increased the mitotic rate during the growth phase by 70% over control without affecting the plating efficiency. Within 48 hr of exposure to EGF, the endothelial cells became spindle-shaped. This morphological change was quantitated by morphometry; cells treated with EGF had a major axis 1.5 X larger than that of non-EGF treated cells. The spindle-shaped morphological change did not occur in response to other growth factors, was not related to cell density, and was reversible within 24 hr after removal of EGF from the media or subculture in the absence of EGF. The addition of 5-fluorouracil blocked cell division but did not affect the EGF-induced morphological change. The appearance of the endothelial cells following EGF stimulation is similar to migrating cells closing a wound in vivo.

Aqueous to cornea fluorescein distribution ratio in normal and swollen cornea.

Intravitreal sodium fluorescein was used to simulate equilibrium fluorescein kinetics, thereby allowing simple measurement of the aqueous to cornea fluorescein distribution ratio. Two groups of rabbit corneas were studied: normal corneas and corneas wounded by freezing. The aqueous to cornea fluorescein distribution ratio was approximately 0.4, was not significantly different in groups of normal or wounded eyes and little variability was noted. In addition, a comparison of in vivo and in vitro measurements of corneal fluorescein concentration in wounded eyes suggests that in vivo protein-bound fluorescein in the cornea fluoresces less efficiently than free fluorescein.

Cellular migration and morphology in corneal endothelial wound repair.

After a mechanical denudation of rabbit corneal endothelial cells, the healing process was followed with wide-field specular microscopy. Individual cell migration and morphologic changes were analyzed by computer-assisted morphometry. The cells surrounding the wound migrated to cover the defect without producing intercellular gaps. The greatest cellular migration and morphologic alterations occurred close to the wound edge. As the cells migrated toward the wound, they elongated and increased their surface area in the direction of the migration. As the healing proceeded, the cells lost their original hexagonal pattern, which returned after coverage was complete. The wound was covered completely by large, irregularly shaped cells showing mitotic figures between 24 and 48 hr. During this period, cellular migration decreased and normal cellular morphology began to recover. When mitosis decreased, the normal cellular pattern rearranged towards a more hexagonal shape. During the healing process, the degree and direction of cellular migration varied from cell to cell. Additionally, changes in cell-to-cell contact (positional changes of neighboring cells) occurred in one-third of migrating cells. Such cellular migration can account for monolayered cells sliding without producing gaps between individual cells.

Corneal endothelial function and structure following cryo-injury in the rabbit.

Wide-field specular microscopy, fluorophotometry, pachymetry, and scanning electron microscopy are used to characterize a reproducible, in vivo model of corneal endothelial injury and recovery in the rabbit. Following an 8-mm central cryo-injury, the cornea remains thickened for as long as 3 weeks. Mean endothelial permeability to fluorescein is above normal for 10 days following injury, but by 14 days postinjury the endothelial permeability to fluorescein is not statistically significantly different from preinjury control values, thus indicating that endothelial permeability probably returns to normal by approximately 2 weeks postinjury. Cell morphology, as determined by scanning electron microscopy, is also essentially normal by 2 weeks postinjury. Endothelial permeability appears to recover before stromal thickness normalizes, suggesting a lag in recovery of endothelial pump function.

Effects of timolol on intraocular pressure following ocular adrenergic denervation.

The effects of timolol on the elevation of intraocular pressure induced by orogastric water-loading were studied in conscious pigmented rabbits which had undergone unilateral, superior cervical ganglionectomy. Each rabbit was studied without timolol treatment and with unilateral 2% timolol treatment, either to the innervated eye or to the denervated eye, 90 min before water-loading. Timolol, applied to the innervated eye, significantly reduced the elevation of intraocular pressure in that eye, but not in the fellow eye. Timolol, applied to the denervated eye, did not affect the elevated intraocular pressure in either the denervated or the fellow eye. These results demonstrate that ocular adrenergic innervation participates in the mechanism of ocular hypotensive action of timolol in water-loaded pigmented rabbits.