| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Consequently, two major initiatives were recently announced, with the intention of increasing our knowledge of the molecular genetics of hookworms (The Wellcome Trust Beowulf Initiative) and to develop rationally designed vaccines for hookworm infection (The Hookworm Vaccine Initiative, Sabin Vaccine Institute, and George Washington University).
Furthering these initiatives will be partially dependent on the use of models of hookworm infection and disease. However, many of the models used historically to investigate the immunobiology of hookworm infection have been biologically deficient. For example murine models, in which few parasites reach the gut and thus cannot exhibit either patency or gut pathology, may have provided valuable information on the protective inflammatory responses to hookworm challenge but cannot support the full life cycle.5–9 Canine models infected with the dog hookworm Ancylostoma caninum could also be of value in proof of principle studies with trial vaccines but these studies use infection with an inappropriate species. Therefore, it would be scientifically valuable to establish a model of human hookworm infection that maintained the full life cycle and exhibited many of the pathologic and immunologic features of human hookworm disease. In a pilot study, patent and pathologic infections were established in marmosets using a laboratory maintained isolate (passaged at Nottingham since 1983) or a fresh field isolate (PNG) collected in October 2001 from Papua New Guinea (Haven, Madang Province). This was followed by a more extensive study of the PNG isolate in a larger group of animals.
From these studies, it would appear that patent infections can be established in adult marmosets by using laboratory and fresh field isolates. In addition, our study may have detected key physiologic and biochemical differences between the two parasite isolates, with the fresh field isolate inducing a profile of pathology more comparable to infection in humans, which suggested passage-induced attenuation of the laboratory strain.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To maintain the laboratory isolate, 2–4-day-old hamsters were infected percutaneously with 100 infective third-stage larvae and the infection was allowed to proceed until adult worms in the small intestine became fecund, approximately 42 days post-infection. Fecal material containing N. americanus eggs was cultured by a method described previously.14
Acquisition of a fresh N. americanus isolate from Papua New Guinea. A fresh field isolate of N. americanus was obtained from Papua New Guinea in October 2001. Fecal material obtained from a hookworm-infected person living in Haven village on the Bogia Coast Road of Madang Province was cultured as previously described.15 Freshly cultured larvae were transported back to the United Kingdom and used to infect neonate hamsters in Nottingham and three marmosets at the Defence Science and Technology Laboratory in Porton Down 15 days after arrival in the United Kingdom and 22 days after collection in Papua New Guinea.
Collection of N. americanus ES products. Excretory/secretory (ES) products from adult worms were collected as described previously.16 Necator americanus–infected hamsters were killed 35 days post-infection and the small intestines were removed, cut along their length, and placed in Petri dishes containing Hanks’ balanced salt solution. The Petri dishes were incubated at 37°C to enable adult worms to detach voluntarily from the intestines, thus minimizing the possibility of host tissue contaminating subsequent ES cultures. Once detached, adult worms were washed over a period of two hours in RPMI 1640 medium containing 100 IU/mL of penicillin and 100 µg/mL of streptomycin and cultured in RPMI 1640 medium for 24 hours. The ES products obtained after 24 hours were stored at –20°C until required.
Proteolytic activity and inhibition of factor Xa by ES products. The proteolytic activity of normal and heat-inactivated (100°C for 30 minutes) ES products was determined using fluorescein isothiocyanate–labeled casein (FITC-casein).17 The ES products (12 µg in 20 µL) were mixed with FITC-casein (10 µL of a 0.5 mg/mL stock) and 170 µL of 50 mM phosphate buffer, pH 6.5, containing 5 mM cysteine, and incubated at 37°C for two hours. To stop the reaction and precipitate undigested protein, trichloroacetic acid (120 µL of 5% [w/v]) was added and allowed to stand at room temperature for one hour. Precipitated protein was removed by centrifugation at 13,000 x g for 10 minutes. Aliquots of the supernatant (20 µL) were added to 80 µL of 0.5 M Tris, pH 8.5 and the fluorescence was measured (excitation at 490 nm and emission detection at 525 nm) by using an MFX microplate fluorimeter (Dynes Technologies, Chantilly, VA). Under these conditions, untreated ES products released 2,439 ± 66 fluorescence units from FITC-labeled casein over two hours. No activity was detected in heat-inactivated ES products.
Inhibition of human factor Xa activity was measured by monitoring the release of 7-amino-4-methylcoumarin (AMC) by Factor Xa (Calbiochem) from the fluorogenic substrate Boc-Ile-Glu-Gly-Arg-AMC.18 Ten micrograms of ES products were pre-incubated for 15 minutes with 0.03 units of factor Xa in 0.05 M Tris, 0.05 M NaCl, pH 7.4, prior to the addition of substrate to a final concentration of 5 µM. The release of AMC was measured over a 15-minute period (excitation at 365 nm and emission detection at 465 nm) using a microplate fluorimeter (Dynes TEechnologies). Unihibited factor Xa released 1,536 fluorescence units over 15 minutes. The ES products from the fresh field isolate reduced this to 446 units, a reduction of 71%. The ES products from the laboratory strain failed to inhibit human factor Xa.
Expression and purification of recombinant calreticulin. Recombinant calreticulin was expressed as described previously19 and purified using a combination of Bugbuster protein extraction reagent (Novagen, Madison, WI) and a His Bind Purification kit (Novagen). Harvested cells were resuspended in Bugbuster reagent (5 mL per g of cell pellet) containing 25 units benzonase/mL of Bugbuster reagent and incubated at room temperature for 20 minutes. Insoluble cell debris was removed by centrifugation at 16,000 x g for 20 minutes at 4°C, and the supernatant was loaded directly onto a 5-mL His bind resin column previously equilibrated with 5 column volumes of 50 mM NiSO4 and three column volumes of binding buffer (5 mM imidazole, 0.5 M NaCl, 20 mM Tris-Cl, pH 7.9). After application, the column was washed with 10 volumes of binding buffer and 6 volumes of wash buffer (60 mM imidazole, 0.5 M NaCl, 20 mM Tris-Cl, pH 7.9) prior to elution. Bound calreticulin was eluted with six column volumes of elution buffer (1 M imidazole, 0.5 M NaCl, 20 mM Tris-Cl, pH 7.9). Fractions containing calreticulin, as determined by protein estimation (Bio-Rad, Hercules, CA), were pooled, dialyzed against phosphate-buffered saline (PBS), and stored at –20°C until required. Purified recombinant protein was sequenced by matrix assisted laser desorption/ionization time-of-flight mass spectrometry to confirm its identity.
Animal husbandry. Common marmosets (three males and three females in the pilot study, and five males and five females in the follow-up study) bred at the Defence Science and Technology Laboratory in Porton Down and weighing 325–430 g were used. The animals were housed as mixed sex pairs, the males having previously been vasectomized. Each pair had access to four cage units (each with a height of 72 cm, a width of 47 cm, and a length of 60 cm) linked by one vertical and two rigid extensions.
Animals were fed daily in the afternoon and each received primate pellets (Special Dietary Services, Witham, Essex, United Kingdom) as well as fruit, eggs, and other supplements, including access to forage mixture, over the course of the week. Water was freely available at all times. Various items of cage furniture, including hanging wooden dowels, buckets, and other playthings, were also placed in the cages. Illumination was provided by sodium lighting at a level of 350–400 lux 1 meter from the ground by using a 12-hour light/dark cycle with dusk and dawn effects.
Infection of marmosets with N. americanus. Marmosets were anesthetized with ketamine (40 mg/kg) and an area of skin approximately 2 cm2 below the scapulae was close clipped. Infective larvae were placed on a gauze fixed to a self-adhesive bandage (International Market Supply, Congleton, Cheshire, United Kingdom), which was wrapped around the thorax and held in place with a Tubigrip jacketTM (Seton Products Ltd., Oldham, U.K.). The jacket and bandage were removed after 24 hours.
The pilot study used six marmosets that were divided into three treatment groups. Treatment group 1 (marmosets 1–3) received 300 laboratory strain infective larvae on two occasions. Treatment group 2 (marmoset 4) received 300 laboratory strain infective larvae followed by re-infection with 300 larvae from the PNG fresh isolate. Treatment group 3 (marmosets 5 and 6) were infected on one occasion with the PNG fresh isolate (300 and 600 larvae, respectively). The infection protocols are summarized in Figure 1
. In the second, more extensive study, 10 marmosets were infected with one dose of 300 larvae from the PNG isolate. This dose was chosen to represent the lower end of the infectivity scale shown to induce a degree of pathology in the model during the pilot study. To ensure that the PNG strain had no opportunity to adapt to rodents, for the pilot study, marmosets were infected with larvae cultured from a person in Papua New Guinea. These larvae were also used to infect a human volunteer in the United Kingdom previously shown to be negative for human immunodeficiency virus and hepatitis B and C viruses. The larvae used in the second study were freshly cultured from the volunteer in the United Kingdom.
|
Samples of feces were collected at regular intervals. Fecal matter was weighed, and eggs harvested by salt flotation and counted.21 Egg counts were expressed as eggs per gram of feces (epg). In the second study, all blood samples were taken at 14-day intervals and fecal samples were taken at 7-day intervals. The infection was terminated 70 days post-infection; this time point corresponded to the peak of anemia found in the pilot study.
Immunologic and pathologic analysis. Total IgE determination. Ninety-six well polystyrene plates were coated overnight at 4°C with mouse anti-human IgE (50 µL, 5 µg/mL diluted in 0.05 M carbonate/bicarbonate buffer, pH 9.6, clone no. G7-18; BD Pharmingen, San Diego, CA). The plates were washed with PBS/Tween 20 (0.05% [v/v]), pH 7.2 (PBS/Tween) and blocked with bovine serum albumin (BSA) (200 µL of 1% [w/v]) in PBS (BSA/PBS) for one hour at room temperature. After blocking, the plates were washed and 50 µof marmoset serum (diluted 1:5 in 1% BSA/PBS) added to each well and incubated overnight at 4°C. In addition 50 µL of human IgE standards (doubling dilutions from 100 IU/mL or 100 µg/mL) were included on each plate. All assays were carried out in duplicate. After overnight incubation, the plates were washed and 50 µL of biotinylated mouse anti-human IgE (2 µg/mL diluted in 1% BSA/PBS, clone no. G7-26, BD Pharmingen) added to each well and incubated at room temperature for two hours. After two hours, the plates were washed and 50 µL of streptavidin conjugated to horseradish peroxidase (diluted 1: 1,000 in 1% BSA/PBS) added to each well and incubated for one hour at room temperature. The plates were washed and developed with 100 µL of 0.1 mg/mL 3,3',5,5'-tetramethylbenzidine (TMB) containing 2 µL of hydrogen peroxide/10 mL of TMB. The reaction was stopped by adding 20 µL of 2.5 M sulfuric acid and the absorbance of each well was measured at 450 nm.
Specific IgG determination. Fifty microliters of N. americanus ES products (5 µg/mL in 0.05 M sodium carbonate/ bicarbonate buffer, pH 9.6) were used to coat the wells of a 96-well polystyrene plate overnight at 4°C. The plate was washed with PBS/Tween and the wells were blocked with 200 µL of 5% skimmed milk powder/PBS (blocking agent) for one hour at 37°C. The plate was washed as before and 50 µL of marmoset serum (diluted 1:100 in skimmed milk powder/ PBS) was added to individual wells and the plate incubated overnight at 4°C. The plate was washed, 50 µL of sheep anti-human IgG conjugated to horseradish peroxidase (Binding Site, San Diego, CA) diluted 1:1,000 in blocking agent was added to individual wells, and the plate incubated for two hours at room temperature. The plates were washed and antibody binding was visualized by the addition of 100 µL of TMB prepared as described above. The reaction was stopped by adding 20 µL of 2.5 M sulfuric acid, and the absorbance of each well was measured at 450 nm. All assays were carried out in duplicate. Enzyme-linked immunosorbent (ELISA) values are expressed as the absorbance at 450 nm after the subtraction of a negative control value.
Western blotting. N. americanus ES products (10 µg/lane) were separated, under reducing conditions, by sodium dodecyl sulfate–polyacrylamide gel electrophoresis on a 12% gel22 and transferred onto a nitrocellulose membrane.23 Western blots were blocked for one hour in 5% (w/v) skimmed milk powder in Tris-buffered saline (TBS) at room temperature. Marmoset serum (diluted 1:200 in 5% [w/v] skimmed milk powder/TBS) was added to the blots and incubated overnight at 4°C. Blots were washed with TBS/0.05% (v/v) Tween 20 and incubated in sheep anti-human IgG (Binding Site) diluted 1:1,000 in 5% (w/v) skimmed milk powder/TBS for two hours at room temperature. After washing, blots were developed in 4-chloro-1-napthol (10 mg/mL) containing 0.02% hydrogen peroxide. The validity of these anti-human reagents in ELISA and Western blotting has previously been confirmed.24
Whole blood basophil histamine release. Whole blood was collected from marmosets into heparinized tubes. One hundred microliters of blood was added to give a total volume of 500 µL in PIPES buffer (0.01 M Piperazine-N'N-bis[2-ethaneculfonic acid], 0.14 M sodium acetate, 5 mM potassium acetate, 0.1% glucose, 1 mM CaCl2, and 0.03% human serum albumin, pH 7.4). Spontaneous histamine release was assessed after incubation for one hour at 37°C, and total histamine release was assessed when 50 µL of whole blood in 450 µL of double-distilled water was freeze-thawed three times. Standard histamine calibrators of 0, 10, 25, 50, 100, and 250 ng/mL (Hycor Biomedical Ltd., Penicuik, United Kingdom) were included with each set of whole blood challenges, mediated by anti-IgE, ES products, or recombinant calreticulin. Histamine released in each whole blood challenge was detected using a Histamine Assay Kit (Hycor Biomedical Ltd.). Fifty microliters of challenged whole blood was added to histamine-coated 96-well plates followed by 50 µL of mouse anti-histamine monoclonal antibody conjugated to alkaline phosphatase. After incubating for one hour at room temperature, wells were washed three times with the provided wash solution. Antibody binding was visualized by the addition of 100 µL (1 mg/mL) of p-nitrophenyl phosphate substrate. The plates were developed for one hour at 37°C, and the absorbance was measured at 405 nm using an MRX absorbance microplate reader (Dynex Technologies). Histamine levels were determined against a standard curve.
Pathologic analysis. At the end of the study, animals were killed with an overdose of sodium pentabarbitone and intestinal and lung tissue were fixed for a minimum of 48 hours in 10% (v/v) neutral (phosphate)–buffered formaldehyde. Tissues were resected into standard infiltration cassettes. Intestinal tissue was first enveloped in Whatman (Maidstone, United Kingdom) no. 1 filter paper. Tissues in cassettes were processed in a Sakura (Torrance, CA) Tissue Tek VIP processor, incorporating a pressure/vacuum and agitation cycle throughout with a nominal solvent chamber temperature of 40°C. Paraffin wax was introduced at 60°C. Tissues were dehydrated through three changes each of graded alcohols (80%, 90%, and absolute ethanol) followed by chloroform (one hour), chloroform (two hours, two changes), chloroform and xylene (50:50; 15 minutes), xylene (10 minutes), and paraffin wax (Paramat pastillated at 58°C; two hours, three changes). Embedding was performed using a Sakura Tissue Tek embedder with reservoir set at 65°C. Wax-embedded tissues were cooled on a freezing plate prior to storage at 4°C while awaiting microtomy. Sections were cut at a nominal thickness of 5 µm by using a Leitz base sledge microtome (Leica, Wetzlar, Germany). Sections were floated out on water at 52–55°C, collected onto pre-cleaned glass microscope slides, and dried in an oven overnight at 40°C prior to staining. Hematoxylin and eosin stain was applied to tissue sections by using an automated Sakura Linear Stainer II with a routine staining method25 based on the instrument Technical Manual Code 1419/1423 supplied through Bayer Plc (Newbury, United Kingdom).
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In the pilot study, all animals in treatment groups 1 and 3 showed evidence of infection, as indicated by the appearance of eggs in feces. All animals reacted immunologically to infection but only the two animals in treatment group 3, which were exposed to one challenge with 300 or 600 larvae from the PNG fresh field isolate demonstrated infection-associated pathology, as shown by a dramatic reduction in hemoglobin 48–62 days post-infection (Figure 2C) and PCV (Table 1). The animal infected with 600 larvae showed a peak decrease in hemoglobin levels approximately 14 days earlier than the animal infected with 300 larvae. Treatment group 2, which received the laboratory isolate prior to the PNG fresh isolate, did not show either pathology (Figure 2B) or egg production (Table 1). Similarly, erythrocyte numbers were only reduced during infection in treatment group 3, which were infected with the fresh field isolate and reached their lowest levels 40–60 days post-infection, and corresponded to peak egg output (Figure 3C). This effect was not seen in treatment groups 1 and 2 (Figure 3A and B). In treatment group 3, where it was also possible to measure mean cell volume, microcytic anemia was observed.
|
|
|
). Antibodies recognized a typical antigen profile on Western blots, which included the major 33-kD antigen (Figure 4) in all animals except for the animal in treatment group 2.25 Although the response was stronger in some animals than others, the peak response on Western blots corresponded in most cases with peak egg output.
|
, led us to investigate the sensitization of basophils with hookworm-specific IgE. Results in Figure 5 showed that all treatment groups had significant histamine release to ES products and a recombinant hookworm allergen calreticulin. Those animals exposed to the fresh field PNG isolate showed increased histamine release (Figure 5B–D) than animals only exposed to the laboratory isolate (Figure 5A). To control for possible non-specific release of histamine by enzymes in ES products,26 some preparations were heat inactivated by boiling for 30 minutes to neutralize activity. Heat-inactivated ES products produced a similar level of release, which indicated the presence of non-proteolytic, allergenic material in heat-inactivated ES products. To control for the ability of basophils to release histamine, cells were also challenged with antibodies to IgG and IgE. Three of six animals (treatment groups 2 and 3) released histamine to antibodies to IgE to a greater extent than that induced by antibodies to IgG. The failure of treatment group 1 to release histamine suggested that the cells were not sufficiently sensitized with hookworm-specific IgE when the old laboratory isolate was used.
|
At post mortem, pulmonary pathology, characterized by peripheral alveolar hemorrhagic and fluid accumulation, was evident in animals that received the fresh PNG isolate, an example of which is shown in Figure 6A. Figure 6B shows a transverse section of the mouth end of a hookworm, located in the gut of the animal that had received 600 fresh N. americanus larvae approximately four months earlier. The buccal cavity, cutting plates, esophagus, and amphidial glands of the worm can be seen.
|
). All 10 animals demonstrated a reduction in PCV and hemoglobin levels (Table 1 and Figure 2D), corresponding approximately with peak egg output. However, the most significant reductions were observed in the four animals that harbored worms at autopsy (Figure 2D, inset). Animal 7 harbored 18 worms (12 males and 6 females), animal 12 harbored 1 worm (1 male), animal 13 harbored 29 worms (17 males and 12 females), and animal 15 harbored 9 worms (6 males and 3 females). Similarly, erythrocyte levels decreased in all animals (Figure 3D), with the most significant decreases again observed in the four animals with worms at autopsy (Figure 3D, inset). All animals showed an increase in total IgE and parasite-specific IgG post-infection (Table 1). In addition, histopathologic examination of the lungs confirmed the presence of fibrosis (alveolar thickening) associated with evidence of debris left by worms in transit. The animal with the greatest number of adults found in the intestine (animal 13) also showed the greatest preponderance of these localized changes. |
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our report describes pilot and follow-up studies in which patent and pathologic hookworm (N. americanus) infections were established in non-human primates. These studies had the aim of developing an improved and biologically relevant animal model in a species closely allied to humans, thus reducing the use of inappropriate animal species to study necatoriasis.
The pilot was successful in that patency was observed in most animals, whether given laboratory or PNG isolates of the parasite. However, it is significant that blood loss was only seen in treatment group 3 (infected with 300 and 600 PNG larvae), which indicated that attenuation may have occurred in the laboratory isolate. This attenuation may have been a result of repeated passage (approximately 460) through hamsters during the laboratory maintenance of this strain since 1983, in which they seem to have lost their full anti-hemostatic repertoire. The blood loss caused by the PNG isolate was demonstrated by a dramatic decrease in hemoglobin and PCV levels in treatment group 3, accompanied by evidence of a microcytic anemia, presumably as a direct result of blood loss in the lungs during transit9 by infective larvae, and feeding by adult worms in the gut.
Infection in all animals was associated with an immune response reminiscent of that seen in humans, where antibodies were readily detectable against hookworm secretions, accompanied by (with the exception of the animal in treatment group 2) a typical Western blot profile, with a 33-kDa antigen predominant.27 The animal in treatment group 2 received 300 larvae from the laboratory strain followed by 300 larvae form the PNG isolate. After secondary infection, this animal did not show the patency or pathology associated with infection with the PNG isolate. This finding may have been caused by a potential protective effect induced by infection with the laboratory strain. Basophil sensitization with parasite IgE was a salient feature of this model, which is indicative of activation of the T-helper 2 immunologic subsystem and deemed important in immunity to N. americanus.28 Basophil histamine release to multiple agonists was most consistent in animals exposed to the PNG isolate. This finding probably reflected more efficient loading of Fc
RI on basophils with parasite-specific IgE, which was not detectable serologically because of the low sensitivity of anti-human IgE reagents in the ELISA. Histamine release by calreticulin confirmed the allergenic properties of this molecule.29
The follow-up study, in which all animals were infected with the PNG isolate, confirmed our pilot study findings of patent infections with a relatively consistent onset of fecundity approximately 42 days post-infection. In addition, both the epg level and the day of peak epg are comparable with data obtained from humans.30 All 10 animals demonstrated a considerable immune response to infection and a decrease in hemoglobin levels and PCV, although this was more noticeable in some animals than others. However, worm recoveries were inconsistent, with only four animals still harboring worms at autopsy.
It can be concluded at this stage that further development of this model will add to our knowledge of the pathobiology of necatoriasis per se, and support the investigation of a number of related immunologic issues. For example, given the increasing availability of reagents for immunologic studies in the marmoset24 the primate model will also provide an opportunity to assess the relationship between hookworm infection and the development of allergic31 and autoimmune diseases and the impact of hookworm infection on co-current infection with malaria and simian immunodeficiency virus.
Finally, further development of the model may ultimately provide a unique opportunity to develop a hookworm vaccine in a system where vaccine safety, delivery, and efficacy can be assessed against worm establishment, parasite patency, infection-associated pathology, and cognitive development.32,33 Furthermore, the comparatively close evolutionary relationships between marmosets and humans, the strong degree of similarity between their MHC and T cell receptor genes, their immunologic characterization, and the availability of defined reagents combined with their suitability for cognitive studies, and their acceptance by the Food and Drug Administration as being important to preclinical evaluation of biotechnology derived pharmaceuticals,34 lend support to the choice of this species as a model for hookworm vaccination development, in preference to the less suitable mouse, hamster, and canine models.
Received December 1, 2005. Accepted for publication January 31, 2008.
Acknowledgments: We thank Neil Hughes, Bob Knight, and Jan Platt for support and advice in preparation and processing of tissue samples for pathologic analysis and photomicrographs; Arthur Baskerville for advice on interpretation of pulmonary pathologic changes; and Ed Gosden and Keith Male for hematologic analysis. Financial support: This work was supported by the Wellcome Trust and the Defence Science and Technology Laboratory.
* Address correspondence to Alan P. Brown, The Boots Science Building, School of Pharmacy, University of Nottingham, Nottingham NG7 2RD, United Kingdom. E-mail: alan.brown{at}nottingham.ac.uk 
Authors’ addresses: Gareth D. Griffiths, Peter C. Pearce, Rebecca J. Hornby, and Leah Scott, Defence Science and Technology Laboratory, Porton Down, Salisbury, Wiltshire SP4 0JQ, United Kingdom, Tel: 44-1980-613-367, Fax: 44-1980-613-741. Alan P. Brown, Doreen S. W. Hooi, and David I. Pritchard, The Boots Science Building, School of Pharmacy, University of Nottingham, NG7 2RD, Nottingham, United Kingdom, Tel: 44-115-951-6165/846-6292, Fax: 44-115-951-5122.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
