Effect of a glucosamine derivative on impact-induced chondrocyte apoptosis in vitro. A preliminary report.

Here, we report that a lipophilic derivative of glucosamine, Glu5, is able to prevent impact-induced chondrocyte death by the putative mechanism of reducing mitochondrial depolarisation following a single impact load in vitro.

Calcium signaling leads to mitochondrial depolarization in impact-induced chondrocyte death in equine articular cartilage explants.

OBJECTIVE: Chondrocyte apoptosis is an important factor in the progression of osteoarthritis. This study aimed to elucidate the mechanisms involved upstream of caspase 9 activation and, in particular, calcium signaling and mitochondrial depolarization. METHODS: Articular cartilage explants obtained from healthy horses were subjected to a single impact load (500-gm weight dropped from a height of 50 mm) and cultured in vitro for up to 48 hours. Chondrocyte death was quantified by the TUNEL method. Release of proteoglycans was determined by the dimethylmethylene blue assay. Weight change was measured, and mitochondrial depolarization was determined using JC-1 staining. To assess the role of calcium signaling in impact-induced chondrocyte death, explants were preincubated in culture medium containing various concentrations of calcium. Inhibitors were used to assess the role of individual signaling components in impact-induced chondrocyte death. RESULTS: Calcium quenching, inhibitors of calpains, calcium/calmodulin-regulated kinase II (CaMKII), and mitochondrial depolarization reduced impact-induced chondrocyte death after 48 hours in culture. Transient mitochondrial depolarization was observed 3-6 hours following a single impact load. Mitochondrial depolarization was prevented by calcium quenching, inhibitors of calpain, CaMKII, permeability transition pore formation, ryanodine receptor, and the mitochondrial uniport transporter. Cathepsin B did not appear to be involved in impact-induced chondrocyte death. The calpain inhibitor prevented proteoglycan loss, but the percentage weight gain and proteoglycan loss were unaffected by all treatments used. CONCLUSION: Following a single impact load, calcium is released from the endoplasmic reticulum via the ryanodine receptor and is taken up by the mitochondria via the uniport transporter, causing mitochondrial depolarization and caspase 9 activation. In addition, calpains and CaMKII play important roles in causing mitochondrial depolarization.

Do the matrix degrading enzymes cathepsins B and D increase following a high intensity exercise regime?

OBJECTIVE: It has been shown by others that levels of matrix degrading enzymes are increased in osteoarthritis (OA) and so are proposed to be involved in the aetiopathogenesis of the disease, including exercise-associated OA. Therefore we hypothesised that cathepsin B and cathepsin D were increased in cartilage samples previously shown to have early stage OA from 2-year-old Thoroughbred horses, euthanased for reasons other than this study, that had a history of 19-week high intensity exercise (n=6) compared to age and sex-matched horses with a history of low intensity exercise (n=6). METHODS: Cartilage samples were used from four specific sites within the carpal joints. Standard immunolocalisation protocols and blind counting of positive and negative cells within the articular surface, mid-zone and deep zone (DZ) were used to test our hypothesis. RESULTS: A high intensity exercise regime did not significantly alter the number of chondrocytes positive for cathepsin B, whereas a significant decrease was found for cathepsin D in the DZ, indicating that these enzymes are regulated differently by mechanical loading. Furthermore, cathepsin D varied according to the topographical location within the joint, reflecting biomechanical differences experienced during a high compared to a low intensity exercise regime. CONCLUSION: This study disproves our hypothesis that cathepsins B and D are increased following a high intensity exercise regime unlike that reported for other matrix enzymes.

Inhibition of caspase-9 reduces chondrocyte apoptosis and proteoglycan loss following mechanical trauma.

OBJECTIVE: Chondrocyte death, a notable feature of osteoarthritis, may play a role in the initiation of cartilage degeneration. The present study was aimed at uncovering the nature and involvement of cell death in the initiation of cartilage degeneration induced by mechanical trauma. METHODS: Articular cartilage discs obtained from healthy skeletally mature horses were subjected to a single-impact load (500 g from 50 mm) using a simple drop-tower device and cultured in vitro for 48 h. Chondrocyte death was examined using two independent methods: transmission electron microscopy and caspase-3 activity assay. To elucidate the signalling mechanisms involved in impact-induced cell death measured by terminal deoxynucleotidyl transferase-deoxyuridine triphosphate (dUTP) nick-end labelling (TUNEL), cartilage discs were incubated with specific caspase-3, -8 and -9 inhibitors prior to impact. Additionally, weight gain and glycosaminoglycan (GAG) release, markers of cartilage degeneration were monitored. RESULTS: After 48 h, ultrastructural evidence of apoptosis was observed. Caspase-3 was activated after 12h of culture post-impact. When quantified by TUNEL, impact trauma induced death in 52.6% of superficial chondrocytes after 48 h in culture, compared to 4.2% in unimpacted controls. Specific caspases-3 and -9 inhibitors significantly reduced impact-induced apoptosis to 24.3% and 14.7%, respectively. Caspase-8 inhibition had no effect on chondrocyte death (60.3%). Impact-induced GAG release into the medium was significantly reduced by inhibition of cell death, but weight gain remained unaffected by caspase inhibition. CONCLUSION: These results suggest that impact trauma-induced chondrocyte death is predominantly due to caspase-9-dependent apoptosis and is linked to cartilage degeneration.

Promotion of the intrinsic damage-repair response in articular cartilage by fibroblastic growth factor-2.

OBJECTIVE: To identify the effect of fibroblastic growth factor-2 (FGF-2) on the intrinsic damage-repair response in articular cartilage in vitro. METHODS: Articular equine cartilage explants, without subchondral bone, had a single impact load of 500 g applied from a height of 2.5 cm. Explants were then cultured in 0, 12, 25, 50 or 100 ng/ml FGF-2 for up to 28 days. Unimpacted discs served as controls for each time-point. Histological and immunohistochemical techniques were used to quantify and characterise the response of putative chondrocyte progenitor cells (CPC) to damage and FGF-2 treatment. RESULTS: FGF-2 significantly accelerated the appearance and increased the numbers of de novo repair cells identified histologically at the cartilage surface. The response was affected by the dose of FGF-2. The repair cells were shown to be chondrocytes by their expression of collagen types II, IX/XI, but not of type I collagen. In addition, these cells, and those underlying the articular surface, were shown to be immunopositive for Notch-1 and PCNA, markers for proliferating cartilage progenitor cells. CONCLUSIONS: The results of this study indicate that, following single impact load, CPC can be stimulated in mature articular cartilage in vitro. These CPC and the cells arising from them appear to represent the cartilage's response to damage. The timing of the appearance of CPC and their overall numbers can be significantly increased by FGF-2, providing further evidence for an important role for FGF-2 in modulating cartilage repair. These results indicate that further study into the mechanisms of repair in mature cartilage using this in vitro model are vital in understanding the repair capacity of mature cartilage.

Homeostasis of intracellular Ca2+ in equine chondrocytes: response to hypotonic shock.

REASONS FOR PERFORMING STUDY: Ca2+ homeostasis in articular chondrocytes affects synthesis and degradation of the cartilage matrix, as well as other cellular functions, thereby contributing to joint integrity. Although it will be affected by mechanical loading, the sensitivity of intracellular Ca2+ concentration ([Ca2+]i) in equine articular chondrocytes to many stimuli remains unknown. HYPOTHESIS: An improved understanding of Ca2+ homeostasis in equine articular chondrocytes, and how it is altered during joint loading and pathology, will be important in understanding how joints respond to mechanical loads. METHODS: [Ca2+]i was determined using the fluorophore fura-2. We examined the effects of hypotonic shock, a perturbation experienced in vivo during mechanical loading cycles. We used inhibitors of Ca2+ transporters to ascertain the important factors in Ca2+ homeostasis. RESULTS: Under isotonic conditions, [Ca2+]i was 148 +/- 23 nmol/l, increasing by 216 +/- 66 nmol/l in response to reduction in extracellular osmolality of 50%. Resting [Ca2+]i, and the increase following hypotonic shock, were decreased by Ca2+ removal; they were both elevated when extracellular [Ca2+] ([Ca2+]o) was raised or following Na+ removal. The hypotonicity-induced rise in [Ca2+]i was inhibited by exposure of cells to gadolinium (Gd3+; 10 micromol/l), an inhibitor of mechanosensitive channels. [Ca2+]i was also elevated following treatment of cells with thapsigargin (10 micromol/l), an inhibitor of the Ca2+ pump of intracellular stores. CONCLUSIONS: A model is presented which interprets these findings in relation to Ca2+ homeostasis in equine articular chondrocytes, including the presence of mechanosensitive channels allowing Ca2+ entry, a Na+/Ca2+ exchanger for removal of intracellular Ca2+ and intracellular stores sensitive to thapsigargin. POTENTIAL RELEVANCE: A more complete understanding of Ca2+ homeostasis in equine chondrocytes may allow development of future therapeutic regimes to ameliorate joint disease.

Differential distribution of cathepsins B and L in articular cartilage during skeletal development in the horse.

REASONS FOR PERFORMING STUDY: This study was designed to examine a new role for cysteine proteinases in the process of endochondral ossification. OBJECTIVES: The aim of the present study was to investigate the presence and distribution of cathepsin B and cathepsin L in equine articular cartilage during development. METHODS: Full-depth cartilage samples from a total of 40 horses (age range: 4 month fetuses to 2 years) were examined and enzymes detected by immunocytochemical localisation. RESULTS: Observations on the presence of cathepsins B and L revealed significant age-related differences, resulting in clear division of the animals into 2 age groups: i) fetuses and neonates; ii) young growing horses (age 4 weeks to 2 years). Cathepsin B was not detected in cartilage from the majority of fetuses and neonates but was located characteristically in chondrocytes at the articular surface and hypertrophic zone in all growing horses. In contrast, cathepsin L was predominantly present in fetal and neonatal cartilage, located primarily in proliferating chondrocytes. CONCLUSIONS: This study is the first to demonstrate differential and site-specific roles for cathepsin B and cathepsin L in skeletal development in the horse. POTENTIAL RELEVANCE: The demonstrated involvement of cathepsins B and L in endochondral ossification is of relevance to developmental orthopaedic diseases such as osteochondrosis in which there is a focal failure of bone formation.

Localisation and activity of cathepsins K and B in equine osteoclasts.

Cathepsin K and cathepsin B were immunolocalised in equine osteoclasts (OC s) present in ex vivo cartilage/subchondral bone samples. Samples were obtained post mortem from the lateral trochlear ridge (LTR) of six horses and ponies aged between 303 days gestation to 8 months. Strong expression of cathepsin K was detected in OC s, particularly those located at the osteochondral junction, apparently involved in the resorption of calcified cartilage. Cathepsin K expression was also detected in hypertrophic chondrocytes and in the endothelial cells of some blood vessels penetrating the hypertrophic zone of cartilage. By contrast, cathepsin B was either absent or present at very low levels in OC s.Osteoclast-like cells (OCL s) were generated in vitro from bone marrow (BM), obtained from the femurs of one horse and two ponies. High levels of cathepsin K activity but only very low levels of cathepsin B activity were demonstrated in OCL s using fluorogenic substrates for these enzymes. The cathepsin K activity could be blocked by the general cysteine proteinase inhibitor, E-64, but not by the cathepsin B inhibitor, CA-074Me. The cathepsin B activity was completely blocked by both CA-074Me and E-64. Taken together, these results suggest that cathepsin K is more important than cathepsin B in the osteoclastic resorption of bone and calcified cartilage of developing equine long bones. Given the apparent importance of cathepsin K in equine endochondral ossification further investigation into the possibility that abnormal expression of this enzyme is involved in the pathogenesis of equine developmental orthopaedic disease is warranted.

Generation and activity of equine osteoclasts in vitro: effects of the bisphosphonate pamidronate (APD).

Equine osteoclast-like cells (OCLs) were generated from the bone marrow (BM) of two ponies and one horse in the presence of RANKL, the receptor activator of NF kappa B ligand and macrophage colony-stimulating factor (M-CSF). The phenotype of these cells was confirmed by demonstration of characteristics typical of osteoclasts (OCs) including: the expression of tartrate-resistant acid phosphatase (TRAP), the vitronectin receptor (VNR) and the calcitonin receptor (CTR), the demonstration of responsiveness to calcitonin (CT) and the ability to form resorption lacunae on ivory slices and calcium phosphate films. The bisphosphonate pamidronate (APD) dose-dependently inhibited resorption of calcium phosphate films by equine OCLs with an IC(50) of 5.8 x 10(-7) M in one horse. APD also dose-dependently inhibited the number of OCLs present in BM cultures after 7 days. However, this effect is most likely attributable to increased OCL death rather than decreased OCL formation. Paradoxically, ADP appeared to cause an early, transient, increase in OCL formation in BM cultures, however, this effect was reversed after 7 days. These preliminary in vitro data support the potential use of APD in clinical conditions characterised by increased bone turnover such as osteomyelitis, osteitis, septic osteoarthritis, navicular disease, cystic bone lesions and immobilisation-induced osteoporosis and provide useful information for future pharmacokinetic studies and clinical trials in vivo.

Control of viremia and prevention of clinical AIDS in rhesus monkeys by cytokine-augmented DNA vaccination.

With accumulating evidence indicating the importance of cytotoxic T lymphocytes (CTLs) in containing human immunodeficiency virus-1 (HIV-1) replication in infected individuals, strategies are being pursued to elicit virus-specific CTLs with prototype HIV-1 vaccines. Here, we report the protective efficacy of vaccine-elicited immune responses against a pathogenic SHIV-89.6P challenge in rhesus monkeys. Immune responses were elicited by DNA vaccines expressing SIVmac239 Gag and HIV-1 89.6P Env, augmented by the administration of the purified fusion protein IL-2/Ig, consisting of interleukin-2 (IL-2) and the Fc portion of immunoglobulin G (IgG), or a plasmid encoding IL-2/Ig. After SHIV-89.6P infection, sham-vaccinated monkeys developed weak CTL responses, rapid loss of CD4+ T cells, no virus-specific CD4+ T cell responses, high setpoint viral loads, significant clinical disease progression, and death in half of the animals by day 140 after challenge. In contrast, all monkeys that received the DNA vaccines augmented with IL-2/Ig were infected, but demonstrated potent secondary CTL responses, stable CD4+ T cell counts, preserved virus-specific CD4+ T cell responses, low to undetectable setpoint viral loads, and no evidence of clinical disease or mortality by day 140 after challenge.

Equine osteoclast-like cells generated in vitro demonstrate similar characteristics to directly isolated mature osteoclasts.

We report on novel methods to isolate osteoclasts (OC s) and generate osteoclast-like cells (OCL s) from the bone and bone marrow of the equine femur. OC s were successfully isolated from bone scrapings taken from the endosteal surface of the femurs of three horses. OCL s were generated from bone marrow cells taken from the same animals. The validity of using the formation of OCL s as a method for studying OC differentiation and activity was confirmed by the similar characteristics of these two cells. In particular, they both were multinuclear, expressed the enzyme tartrate resistant acid phosphatase and the vitronectin receptor. Most importantly, both were able to resorb bone as demonstrated by the formation of extensive resorption pits when cultured on dentine slices. The generation of OCL s from bone marrow obtained from the equine femur can therefore be used to study equine OC differentiation and for studies requiring the generation of large numbers of these cells. OC s isolated directly from the same bones may be used to examine the effect of a variety of factors on bone resorption in vitro and to continually reaffirm the validity of using OCL s for large-scale studies on OC biology. Such research is essential for improved understanding of bone turnover and endochondral ossification in the horse.

Augmentation of immune responses to HIV-1 and simian immunodeficiency virus DNA vaccines by IL-2/Ig plasmid administration in rhesus monkeys.

The potential utility of plasmid DNA as an HIV-1 vaccination modality currently is an area of active investigation. However, recent studies have raised doubts as to whether plasmid DNA alone will elicit immune responses of sufficient magnitude to protect against pathogenic AIDS virus challenges. We therefore investigated whether DNA vaccine-elicited immune responses in rhesus monkeys could be augmented by using either an IL-2/Ig fusion protein or a plasmid expressing IL-2/Ig. Sixteen monkeys, divided into four experimental groups, were immunized with (i) sham plasmid, (ii) HIV-1 Env 89.6P and simian immunodeficiency virus mac239 Gag DNA vaccines alone, (iii) these DNA vaccines and IL-2/Ig protein, or (iv) these DNA vaccines and IL-2/Ig plasmid. The administration of both IL-2/Ig protein and IL-2/Ig plasmid induced a significant and sustained in vivo activation of peripheral T cells in the vaccinated monkeys. The monkeys that received IL-2/Ig plasmid generated 30-fold higher Env-specific antibody titers and 5-fold higher Gag-specific, tetramer-positive CD8+ T cell levels than the monkeys receiving the DNA vaccines alone. IL-2/Ig protein also augmented the vaccine-elicited immune responses, but less effectively than IL-2/Ig plasmid. Augmentation of the immune responses by IL-2/Ig was evident after the primary immunization and increased with subsequent boost immunizations. These results demonstrate that the administration of IL-2/Ig plasmid can substantially augment vaccine-elicited humoral and cellular immune responses in higher primates.

Epitope insertion into variable loops of HIV-1 gp120 as a potential means to improve immunogenicity of viral envelope protein.

We report on the properties of a set of HIV-1 IIIB Env mutants carrying a linear gp41 epitope insertion (LLELDKWASL) in the V1, V2, V3 or V4 variable loop. Insertion of the epitope, which is defined by the HIV-1 neutralizing MAb 2F5, was well tolerated in the V1, V2 and V4 loops, as these mutants were properly expressed, retained reactivity to conformation-dependent monoclonal antibodies and exhibited patterns similar to the parental Env molecule. However, insertion of this epitope in the V3 loop was associated with drastically reduced protein expression. Relative to parental Env molecule, the V1, V2 and V4 insertion mutants demonstrated significantly increased binding to mAb 2F5 in vitro. To evaluate immunogenicity, mice and guinea pigs were immunized with plasmid expression vectors for the mutant proteins. For both mice and guinea pigs, all four mutants elicited anti-gp120 antibody responses. In mice the V1 and V3 insertion mutants, but neither the V2 or V4 insertion mutant nor the parental env, elicited significant titers against the epitope peptide, whereas in guinea pigs, V2 insertion mutant was most effective in eliciting anti-2F5 peptide antibody responses. While original V2 2F5 insertion mutant failed to elicit anti-2F5 peptide responses in mice, studies with 14 additional V2 insertion mutants revealed several insertion sites at which the epitope was able to induce epitope-specific antibody responses. This indicates that the precise position at which the epitope insertion takes place dictates the ability of the mutant to induce the epitope-specific antibody responses. When tested for virus neutralization activity, the guinea pig sera that contain high titers of anti-2F5 peptide antibody failed to enhance the virus neutralizing activity, suggesting that the configuration of 2F5 epitope plays a critical role in inducing neutralizing antibody responses. The results from this study may have potential implications with respect to modification of the HIV-1 Env molecule for the purpose of improving HIV-1 Env immunogenicity.

In vitro generation of equine osteoclasts from bone marrow cells using a novel culture system.

We report on preliminary results of a novel in vitro culture system designed to generate equine osteoclasts in large numbers. Osteoclast generation, as determined by the expression of tartrate resistant acid phosphatase (TRAP) and ability to resorb bone, was enhanced in equine bone marrow cultures supplemented with fibroblastic cell (L929) conditioned medium (L929-CM). Bone marrow was collected from a total of 12 horses and ponies and TRAP-positive cells with bone resorbing ability were generated in significant numbers in the last seven. TRAP-positive mononuclear cells appeared after three days in culture and steadily increased in number and size up to seven days when multinuclear TRAP-positive cells started to appear. L929-CM caused a significant increase in the number of TRAP-positive mononuclear cells generated after four and seven days, after which time the number of TRAP-positive cells decreased. The development of this in vitro system provides a reproducible means of generating large numbers of equine osteoclasts necessary for the study of bone remodelling and endochondral ossification in the horse.

Augmentation and suppression of immune responses to an HIV-1 DNA vaccine by plasmid cytokine/Ig administration.

The use of cytokines has shown promise as an approach for amplifying vaccine-elicited immune responses, but the application of these immunomodulatory molecules in this setting has not been systematically explored. In this report we investigate the use of protein- and plasmid-based cytokines to augment immune responses elicited by an HIV-1 gp120 plasmid DNA vaccine (pV1J-gp120) in mice. We demonstrate that immune responses elicited by pV1J-gp120 can be either augmented or suppressed by administration of plasmid cytokines. A dicistronic plasmid expressing both gp120 and IL-2 induced a surprisingly weaker gp120-specific immune response than did the monocistronic pV1J-gp120 plasmid. In contrast, systemic delivery of soluble IL-2/Ig fusion protein following pV1J-gp120 vaccination significantly amplified the gp120-specific immune response as measured by Ab, proliferative, and CTL levels. Administration of plasmid IL-2/Ig had different effects on the DNA vaccine-elicited immune response that depended on the temporal relationship between Ag and cytokine delivery. Injection of plasmid IL-2/Ig either before or coincident with pV1J-gp120 suppressed the gp120-specific immune response, whereas injection of plasmid IL-2/Ig after pV1J-gp120 amplified this immune response. To maximize immune responses elicited by a DNA vaccine, therefore, it appears that the immune system should first be primed with a specific Ag and then amplified with cytokines. The data also show that IL-2/Ig is more effective than native IL-2 as a DNA vaccine adjuvant.

Early views of the martian surface from the Mars Orbiter Camera of Mars Global Surveyor.

High-resolution images of the martian surface at scales of a few meters show ubiquitous erosional and depositional eolian landforms. Dunes, sandsheets, and drifts are prevalent and exhibit a range of morphology, composition (inferred from albedo), and age (as seen in occurrences of different dune orientations at the same location). Steep walls of topographic depressions such as canyons, valleys, and impact craters show the martian crust to be stratified at scales of a few tens of meters. The south polar layered terrain and superposed permanent ice cap display diverse surface textures that may reflect the complex interplay of volatile and non-volatile components. Low resolution regional views of the planet provide synoptic observations of polar cap retreat, condensate clouds, and the lifecycle of local and regional dust storms.

Evidence for a subsurface ocean on Europa.

Ground-based spectroscopy of Jupiter's moon Europa, combined with gravity data, suggests that the satellite has an icy crust roughly 150 km thick and a rocky interior. In addition, images obtained by the Voyager spacecraft revealed that Europa's surface is crossed by numerous intersecting ridges and dark bands (called lineae) and is sparsely cratered, indicating that the terrain is probably significantly younger than that of Ganymede and Callisto. It has been suggested that Europa's thin outer ice shell might be separated from the moon's silicate interior by a liquid water layer, delayed or prevented from freezing by tidal heating; in this model, the lineae could be explained by repetitive tidal deformation of the outer ice shell. However, observational confirmation of a subsurface ocean was largely frustrated by the low resolution (>2 km per pixel) of the Voyager images. Here we present high-resolution (54 m per pixel) Galileo spacecraft images of Europa, in which we find evidence for mobile 'icebergs'. The detailed morphology of the terrain strongly supports the presence of liquid water at shallow depths below the surface, either today or at some time in the past. Moreover, lower-resolution observations of much larger regions suggest that the phenomena reported here are widespread.

Immunolocalization of cathepsin B in equine dyschondroplastic articular cartilage.

A polyclonal antiserum raised in sheep against human cathepsin B was tested for specificity and cross-reactivity with the horse homologue by SDS-PAGE and Western blotting, prior to being used for immunolocalization of the enzyme in equine articular cartilage. In Western blots, the antiserum recognized the 30 kDa single chain and 25 kDa heavy chain of the mature enzyme in purified bovine cathepsin B, and corresponding bands at 32 and 27 kDa in equine chondrocyte and fibroblast lysates. This antiserum was then used to compare the expression and distribution of cathepsin B in normal and dyschondroplastic cartilage of young horses. In normal articular cartilage (n = 6 animals), significant amounts of enzyme were detected only in hypertrophic chondrocytes in the deep zone. The enzyme was intracellular, located in the lysosomal granules. No extracellular matrix staining was observed. Levels of cathepsin B were increased slightly above normal in the deep zone in age-matched dyschondroplastic cartilage (n = 5 animals). The most striking finding, however, was the abundance of the enzyme in chondrocyte clonal clusters associated with the lesions. Cathepsin B levels were low in chondrocytes isolated from normal cartilage (n = 6), but increased progressively during serial subculture, reaching a maximum at passage 5-6. In contrast, primary cultures of dyschondroplastic chondrocytes (n = 3) expressed abundant cathepsin B.

Nonlinear noise reduction through Monte Carlo sampling.

We consider the problem of nonlinear noise reduction within the framework of Bayesian Theory. This enables us to place appropriate weights on the measurement and dynamic errors and thereby avoid over cleaning the data. Using a Metropolis-Hastings sampler, we are able to achieve robust noise reduction without the introduction of ad hoc parameters but at the expense of higher computational complexity. Such an algorithm should also allow us to explore the potential and limitations of other noise reduction methods. (c) 1998 American Institute of Physics.