Short communication: Effect of 3 phytoceutical products on elimination of bacteria in experimentally induced Streptococcus uberis clinical mastitis.

Our objective was to assess the ability of 3 herbal products to eliminate experimentally induced Streptococcus uberis mastitis. These herbal products, also known as phytoceuticals, are used in organically managed dairy cattle to maintain or promote udder health. The products tested were an intramammary product, a topical product, and a product applied to the vulvar area. These products are not approved by the US Food and Drug Administration for treatment of mastitis but they are sold to enhance milk quality or for maintenance or improvement of udder health. Each of the products contains at least one component shown to have antibacterial activity. In this study, we successfully challenge-inoculated 25 lactating dairy cows maintained under organic conditions with an isolate of S. uberis. All challenged cows were positive for S. uberis by milk culture after challenge. When cows met predefined criteria indicating the presence of clinical mastitis, treatment with 1 of the 3 products was initiated based upon a predetermined random allocation. Culture of aseptically collected quarter milk samples was performed before, during, and following challenge with S. uberis. Eight, 8, and 9 cows received the intravulvar, intramammary, and topical treatments, respectively. Milk from all cows that were treated with phytoceuticals were culture-positive for S. uberis at every time point following treatment through 168 h following the last phytoceutical treatment. Based upon the presence of clinical signs and for humane reasons, 2 intravulvar-treated cows, 1 topical-treated, and 4 intramammary-treated cows received intramammary antibiotic therapy. We concluded that the phytoceuticals tested, as dosed and used in this trial, did not produce bacterial cures in S. uberis-induced mastitis.

Short communication: Use of an ultrafiltration device in gland cistern for continuous sampling of healthy and mastitic quarters of lactating cattle for pharmacokinetic modeling.

Pharmacokinetic studies of the drugs in the milk are often limited due to infrequent sampling associated with milking. Alternatively, frequent sample collection with repeated milking may increase drug elimination. The objective of this study was to determine the feasibility of continuously sampling the udder using ultrafiltration. An ultrafiltration probe was placed into the gland cisterns through mammary parenchyma of normal and mastitic quarters of 6 mature mid-lactation Jersey cows with naturally occurring subclinical mastitis. An ultrafiltration probe was secured to the caudal or lateral aspect of the udder depending on the quarter being sampled. The timed interval samples were collected at 0, 2, 4, 6, 8, 12, 18, 24, 28, 32, 36, 48, 60, 72, 84, and 96 h after drug administration. Plasma samples were collected at the same time points. Each cow received 2.2 mg/kg of flunixin intravenously before milking at time 0. All cows were routinely milked by machine every 12 h. Flunixin concentrations in plasma, whole milk, and milk ultrafiltrates were analyzed by use of ultra-high-performance liquid chromatography with mass spectrometric detection. We found no significant effects on the appearance of the milk or the ability to milk the cows after implantation of the ultrafiltration probes. The concentration of flunixin collected from the ultrafiltration probes in the mastitic quarters tended to be greater than that of the healthy quarters. We concluded that collection of ultrafiltration samples from the mammary gland of cows provides a viable means to continuously assess drug concentrations in the milk while continuing to milk the cow normally. This study demonstrates the utility of continuous sampling of milk via ultrafiltration for future pharmacokinetic studies in cattle.

Pharmacokinetics and distribution in interstitial and pulmonary epithelial lining fluid of danofloxacin in ruminant and preruminant calves.

The objective of this study was to compare active drug concentrations in the plasma vs. different effector compartments including interstitial fluid (ISF) and pulmonary epithelial lining fluid (PELF) of healthy preruminating (3-week-old) and ruminating (6-month-old) calves. Eight calves in each age group were given a single subcutaneous (s.c.) dose (8 mg/kg) of danofloxacin. Plasma, ISF, and bronchoalveolar lavage (BAL) fluid were collected over 96 h and analyzed by high-pressure liquid chromatography. PELF concentrations were calculated by a urea dilution assay of the BAL fluids. Plasma protein binding was measured using a microcentrifugation system. For most preruminant and ruminant calves, the concentration-time profile of the central compartment was best described by a two-compartment open body model. For some calves, a third compartment was also observed. The time to maximum concentration in the plasma was longer in preruminating calves (3.1 h) vs. ruminating calves (1.4 h). Clearance (CL/F) was 385.15 and 535.11 mL/h/kg in preruminant and ruminant calves, respectively. Ruminant calves maintained higher ISF/plasma concentration ratios throughout the study period compared to that observed in preruminant calves. Potential reasons for age-related differences in plasma concentration-time profiles and partitioning of the drug to lungs and ISF as a function of age are explored.

The effect of breed and sex on sulfamethazine, enrofloxacin, fenbendazole and flunixin meglumine pharmacokinetic parameters in swine.

Drug use in livestock has received increased attention due to welfare concerns and food safety. Characterizing heterogeneity in the way swine populations respond to drugs could allow for group-specific dose or drug recommendations. Our objective was to determine whether drug clearance differs across genetic backgrounds and sex for sulfamethazine, enrofloxacin, fenbendazole and flunixin meglumine. Two sires from each of four breeds were mated to a common sow population. The nursery pigs generated (n = 114) were utilized in a random crossover design. Drugs were administered intravenously and blood collected a minimum of 10 times over 48 h. A non-compartmental analysis of drug and metabolite plasma concentration vs. time profiles was performed. Within-drug and metabolite analysis of pharmacokinetic parameters included fixed effects of drug administration date, sex and breed of sire. Breed differences existed for flunixin meglumine (P-value<0.05; Cl, Vdss ) and oxfendazole (P-value<0.05, AUC0→∞ ). Sex differences existed for oxfendazole (P-value < 0.05; Tmax ) and sulfamethazine (P-value < 0.05, Cl). Differences in drug clearance were seen, and future work will determine the degree of additive genetic variation utilizing a larger population.

Use of population pharmacokinetic modeling and Monte Carlo simulation to capture individual animal variability in the prediction of flunixin withdrawal times in cattle.

The objective of this study was to develop a population pharmacokinetic (PK) model and predict tissue residues and the withdrawal interval (WDI) of flunixin in cattle. Data were pooled from published PK studies in which flunixin was administered through various dosage regimens to diverse populations of cattle. A set of liver data used to establish the regulatory label withdrawal time (WDT) also were used in this study. Compartmental models with first-order absorption and elimination were fitted to plasma and liver concentrations by a population PK modeling approach. Monte Carlo simulations were performed with the population mean and variabilities of PK parameters to predict liver concentrations of flunixin. The PK of flunixin was described best by a 3-compartment model with an extra liver compartment. The WDI estimated in this study with liver data only was the same as the label WDT. However, a longer WDI was estimated when both plasma and liver data were included in the population PK model. This study questions the use of small groups of healthy animals to determine WDTs for drugs intended for administration to large diverse populations. This may warrant a reevaluation of the current procedure for establishing WDT to prevent violative residues of flunixin.

Plasma pharmacokinetics and milk residues of flunixin and 5-hydroxy flunixin following different routes of administration in dairy cattle.

The objective of this study was to determine if the plasma pharmacokinetics and milk elimination of flunixin (FLU) and 5-hydroxy flunixin (5OH) differ following intramuscular and subcutaneous injection of FLU compared with intravenous injection. Twelve lactating Holstein cows were used in a randomized crossover design study. Cows were organized into 2 groups based on milk production (<20 or >30 kg of milk/d). All cattle were administered 2 doses of 1.1mg of FLU/kg at 12-h intervals by intravenous, intramuscular, and subcutaneous injections. The washout period between routes of administration was 7d. Blood samples were collected from the jugular vein before FLU administration and at various time points up to 36 h after the first dose of FLU. Composite milk samples were collected before FLU administration and twice daily for 5d after the first dose of FLU. Samples were analyzed by ultra-HPLC with mass spectrometric detection. For FLU plasma samples, a difference in terminal half-life was observed among routes of administration. Harmonic mean terminal half-lives for FLU were 3.42, 4.48, and 5.39 h for intravenous, intramuscular, and subcutaneous injection, respectively. The mean bioavailability following intramuscular and subcutaneous dosing was 84.5 and 104.2%, respectively. The decrease in 5OH milk concentration versus time after last dose was analyzed with the nonlinear mixed effects modeling approach and indicated that both the route of administration and rate of milk production were significant covariates. The number of milk samples greater than the tolerance limit for each route of administration was also compared at each time point for statistical significance. Forty-eight hours after the first dose, 5OH milk concentrations were undetectable in all intravenously injected cows; however, one intramuscularly injected and one subcutaneously injected cow had measurable concentrations. These cows had 5OH concentrations above the tolerance limit at the 36-h withdrawal time. The high number of FLU residues identified in cull dairy cows by the United States Department of Agriculture Food Safety Inspection Service is likely related to administration of the drug by an unapproved route. Cattle that received FLU by the approved (intravenous) route consistently eliminated the drug before the approved withdrawal times; however, residues can persist beyond these approved times following intramuscular or subcutaneous administration. Cows producing less than 20 kg of milk/d had altered FLU milk clearance, which may also contribute to violative FLU residues.

Challenges obtaining a biowaiver for topical veterinary dosage forms.

Obtaining a biowaiver for topical drugs used in veterinary species faces many of the same challenges associated with human topicals. However, the skin of domestic animals varies anatomically and biochemically and experimental approaches to assess bioequivalence (BE) in veterinary species have challenges that are not often encountered with human skin. This is especially the situation with locally acting drugs. The focus of this paper is to address several of the challenges associated with (i) determining the BE of these locally acting drugs and (ii) critically examine the current technological advances that can act as a surrogate for clinical trials.

Tulathromycin assay validation and tissue residues after single and multiple subcutaneous injections in domestic goats (Capra aegagrus hircus).

Tulathromycin is a macrolide antimicrobial labeled for treatment of bacterial pneumonia in cattle and swine. The purpose of the present research was to evaluate tissue concentrations of tulathromycin in the caprine species. A tandem mass spectrometry regulatory analytical method that detects the common fragment of tulathromycin in cattle and swine was validated with goat tissues. The method was used to study tulathromycin depletion in goat tissues (liver, kidney, muscle, fat, injection site, and lung) over time. In two different studies, six juvenile and 25 market-age goats received a single injection of 2.5 mg/kg of tulathromycin subcutaneously; in a third study, 18 juvenile goats were treated with 2.5, 7.5, or 12.5 mg/kg tulathromycin weekly with three subcutaneous injections. Mean tulathromycin tissue concentrations were highest at injection site samples in all studies and all doses. Lung tissue concentrations were greatest at day 5 in market-age goats while in the multi-dose animals concentrations demonstrated dose-dependent increases. Concentrations were below limit of quantification in injection site and lung by day 18 and in liver, kidney, muscle, and fat at all time points. This study demonstrated that tissue levels in goats are very similar to those seen in swine and cattle.

Development of a physiologically based pharmacokinetic model to predict tulathromycin distribution in goats.

Physiologically based pharmacokinetic (PBPK) models, which incorporate species- and chemical-specific parameters, could be useful tools for extrapolating withdrawal times for drugs across species and doses. The objective of this research was to develop a PBPK model for goats to simulate the pharmacokinetics of tulathromycin, a macrolide antibiotic effective for treating respiratory infections. Model compartments included plasma, lung, liver, muscle, adipose tissue, kidney, and remaining poorly and richly perfused tissues. Tulathromycin was assumed to be 50% protein bound in plasma with first-order clearance. Literature values were compiled for physiological parameters, partition coefficients were estimated from tissue:plasma ratios of AUC, and the remaining model parameters were estimated by comparison against the experimental data. Three separate model structures were compared with plasma and tissue concentrations of tulathromycin in market age goats administered 2.5 mg/kg tulathromycin subcutaneously. The best simulation was achieved with a diffusion-limited PBPK model and absorption from a two-compartment injection site, which allowed for low persistent concentrations at the injection site and slower depletion in the tissues than the plasma as observed with the experimental data. The model with age-appropriate physiological parameters also predicted plasma concentrations in juvenile goats administered tulathromycin subcutaneously. The developed model and compilation of physiological parameters for goats provide initial tools that can be used as a basis for predicting withdrawal times of drugs in this minor species.

Milk and plasma disposition of thymol following intramammary administration of a phytoceutical mastitis treatment.

Despite the recent growth of the organic dairy industry, organic producers and veterinarians have limited information when choosing mastitis treatments for animals in organic dairy production. Organic producers commonly administer homeopathic or other plant-based products without having research evaluating the efficacy of these products and using estimated or no withholding times to treat mastitis and other health problems in their herds. In this pilot study, we attempted to identify several active ingredients of Phyto-Mast (Penn Dutch Cow Care, Narvon, PA), a plant-based mastitis treatment used on organic dairy farms, and to quantify the product residue in milk and plasma after intramammary administration. We developed an assay to quantify thymol (one of the active ingredients in Phyto-Mast) in milk and plasma using gas chromatography and mass spectrometry (GC-MS). Thymol is a volatile aromatic compound with antiinflammatory properties. As a model for dairy cows, 5 healthy, lactating alpine dairy goats were given 5 mL of Phyto-Mast per udder half. For 10 d following treatment, we analyzed blood and milk samples for thymol residues using GC-MS. The GC-MS assay was very sensitive for thymol detection, to a concentration of 0.01 µg/mL in plasma. Using thymol as a marker, Phyto-Mast was detectable and quantifiable in plasma beginning with the 15-min posttreatment sample, but was no longer detectable in the 4-h posttreatment sample. Thymol residues were only detected in the 12-h posttreatment milk sample. An inflammatory response was not evident in the udder following phytoceutical administration. Although this study provides information about the elimination of thymol, the product contains several other active chemicals, which may have different pharmacokinetic behaviors. Further analysis and additional study animals will help to determine a milk withholding time for Phyto-Mast. Given the recent growth of the organic dairy industry, understanding the pharmacokinetics of therapeutics used in organic production and developing accurate withholding recommendations will help to ensure milk safety.

Pharmacokinetics of tulathromycin after single and multiple subcutaneous injections in domestic goats (Capra aegagrus hircus).

Tulathromycin, a novel triamilide in the macrolide class, is labeled for treatment of bacterial pneumonia in cattle and swine. This manuscript evaluates pharmacokinetics of tulathromycin in goats. In two different studies, six juvenile and ten market-age goats received a single injection of 2.5 mg/kg of tulathromycin subcutaneously; in a third study, 18 juvenile goats were treated with 2.5, 7.5, or 12.5 mg/kg tulathromycin weekly with three subcutaneous injections. Pharmacokinetic parameters estimated from the plasma concentrations from single injections were similar between the two groups of goats and to previously reported parameters in cattle and swine. Mean terminal half-lives were 59.1 ± 7.6 and 61.2 ± 8.7 h for juvenile and market-age goats, respectively. In the multi-dose study, pharmacokinetic parameters estimated from plasma concentrations demonstrated significant differences at P < 0.05 among repeated injections but not among doses. Overall, pharmacokinetic parameters in goats are similar to those reported in cattle and swine, and tulathromycin may prove a useful drug for treating respiratory disease in goats.

A physiologically based pharmacokinetic model linking plasma protein binding interactions with drug disposition.

Combination drug therapy increases the chance for an adverse drug reactions due to drug-drug interactions. Altered disposition for sulfamethazine (SMZ) when concurrently administered with flunixin meglumine (FLU) in swine could lead to increased tissue residues. There is a need for a pharmacokinetic modeling technique that can predict the consequences of possible drug interactions. A physiologically based pharmacokinetic model was developed that links plasma protein binding interactions to drug disposition for SMZ and FLU in swine. The model predicted a sustained decrease in total drug and a temporary increase in free drug concentration. An in vivo study confirmed the presence of a drug interaction. Neither the model nor the in vivo study revealed clinically significant changes that alter tissue disposition. This novel linkage approach has use in the prediction of the clinical impact of plasma protein binding interactions. Ultimately it could be used in the design of dosing regimens and in the protection of the food supply through prediction and minimization of tissue residues.

A system coefficient approach for quantitative assessment of the solvent effects on membrane absorption from chemical mixtures.

A system coefficient approach is proposed for quantitative assessment of the solvent effects on membrane absorption from chemical mixtures. The complicated molecular interactions are dissected into basic molecular interaction forces via Abraham's linear solvation energy relationship (LSER). The molecular interaction strengths of a chemical are represented by a set of solute descriptors, while those of a membrane/chemical mixture system are represented by a set of system coefficients. The system coefficients can be determined by using a set of probe compounds with known solute descriptors. Polydimethylsiloxane (PDMS) membrane-coated fibres and 32 probe compounds were used to demonstrate the proposed approach. When a solvent was added into the chemical mixture, the system coefficients were altered and detected by the system coefficient approach. The system coefficients of the PDMS/water system were (0.09, 0.49, -1.11, -2.36, -3.78, 3.50). When 25% ethanol was added into the PDMS/water system, the system coefficients were altered significantly (0.38, 0.41, -1.18, -2.07, -3.40, 2.81); and the solvent effect was quantitatively described by the changes in the system coefficients (0.29, -0.08, -0.07, 0.29, 0.38, -0.69). The LSER model adequately described the experimental data with a correlation coefficient (r(2)) of 0.995 and F-value of 1056 with p-value less than 0.0001.

A physiologically based pharmacokinetic model of organophosphate dermal absorption.

The rate and extent of dermal absorption are important in the analysis of risk from dermal exposure to toxic chemicals and for the development of topically applied drugs, barriers, insect repellents, and cosmetics. In vitro flow-through cells offer a convenient method for the study of dermal absorption that is relevant to the initial processes of dermal absorption. This study describes a physiologically based pharmacokinetic (PBPK) model developed to simulate the absorption of organophosphate pesticides, such as parathion, fenthion, and methyl parathion through porcine skin with flow-through cells. Parameters related to the structure of the stratum corneum and solvent evaporation rates were independently estimated. Three parameters were optimized based on experimental dermal absorption data, including solvent evaporation rate, diffusivity, and a mass transfer factor. Diffusion cell studies were conducted to validate the model under a variety of conditions, including different dose ranges (6.3-106.9 microg/cm2 for parathion; 0.8-23.6 microg/cm2 for fenthion; 1.6-39.3 microg/cm2 for methyl parathion), different solvents (ethanol, 2-propanol and acetone), different solvent volumes (5-120 microl for ethanol; 20-80 microl for 2-propanol and acetone), occlusion versus open to atmosphere dosing, and corneocyte removal by tape-stripping. The study demonstrated the utility of PBPK models for studying dermal absorption, which can be useful as explanatory and predictive tools that may be used for in silico hypotheses generation and limited hypotheses testing. The similarity between the overall shapes of the experimental and model-predicted flux/time curves and the successful simulation of altered system conditions for this series of small, lipophilic compounds indicated that the absorption processes that were described in the model successfully simulated important aspects of dermal absorption in flow-through cells. These data have direct relevance to topical organophosphate pesticide risk assessments.

Effect of in vivo jet fuel exposure on subsequent in vitro dermal absorption of individual aromatic and aliphatic hydrocarbon fuel constituents.

The percutaneous absorption of topically applied jet fuel hydrocarbons (HC) through skin previously exposed to jet fuel has not been investigated, although this exposure scenario is the occupational norm. Pigs were exposed to JP-8 jet fuel-soaked cotton fabrics for 1 and 4 d with repeated daily exposures. Preexposed and unexposed skin was then dermatomed and placed in flow-through in vitro diffusion cells. Five cells with exposed skin and four cells with unexposed skin were dosed with a mixture of 14 different HC consisting of nonane, decane, undecane, dodecane, tridecane, tetradecane, pentadecane, hexadecane, ethyl benzene, o-xylene, trimethyl benzene (TMB), cyclohexyl benzene (CHB), naphthalene, and dimethyl naphthalene (DMN) in water + ethanol (50:50) as diluent. Another five cells containing only JP-8-exposed skin were dosed solely with diluent in order to determine the skin retention of jet fuel HC. The absorption parameters of flux, diffusivity, and permeability were calculated for the studied HC. The data indicated that there was a two-fold and four-fold increase in absorption of specific aromatic HC like ethyl benzene, o-xylene, and TMB through 1- and 4-dJP-8 preexposed skin, respectively. Similarly, dodecane and tridecane were absorbed more in 4-d than 1-dJP-8 preexposed skin experiments. The absorption of naphthalene and DMN was 1.5 times greater than the controls in both 1- and 4-d preexposures. CHB, naphthalene, and DMN had significant persistent skin retention in 4-d preexposures as compared to 1-d exposures that might leave skin capable of further absorption several days postexposure. The possible mechanism of an increase in HC absorption in fuel preexposed skin may be via lipid extraction from the stratum corneum as indicated by Fourier transform infrared (FTIR) spectroscopy. This study suggests that the preexposure of skin to jet fuel enhances the subsequent in vitro percutaneous absorption of HC, so single-dose absorption data for jet fuel HC from naive skin may not be optimal to predict the toxic potential for repeated exposures. For certain compounds, persistent absorption may occur days after the initial exposure.

Transferrin receptors and transferrin iron uptake by cultured human blood monocytes.

Transferrin receptors have been previously found on human macrophages and it has also been shown that transferrin iron is taken up by these cells. It has therefore been inferred that the uptake is receptor mediated and involves an endocytic pathway. The subject was addressed directly in the present study in which the transferrin-iron-receptor interaction was characterized in cultured human blood monocytes. Specific, saturable diferric transferrin binding was demonstrated, with a kDa of 3.6 X 10(-8) M and a calculated receptor density of 1.25-2.5 X 10(5) receptors per cell. Incubation at 4 degrees C markedly reduced transferrin binding and completely inhibited iron uptake. Chase experiments confirmed progressive cellular loading of iron, with concomitant loss of transferrin. Inhibitors of endocytic vesicle acidification (ammonium chloride and 2,4-dinitrophenol) inhibited iron unloading from endocytosed diferric transferrin, while microtubular inhibitors (colchicine and vindesine) and a microfilament inhibitor (cytochalasin B) reduced diferric transferrin uptake but had little effect on the iron unloading pathway. A similar effect was noted with a calcium ion antagonist (verapamil) and with 2 calmodulin antagonists (chlorpromazine and imipramine). These latter findings suggest the importance of cytoskeleton-membrane interactions via a calcium, calmodulin and protein kinase C mediated system. Endocytosed iron accumulated progressively as ferritin within the cultured monocytes.

Transferrin iron interactions with cultured hepatocellular carcinoma cells (PLC/PRF/5).

Hepatocellular carcinoma cells of the PLC/PRF/5 cell line had 1.9 x 10(5) transferrin receptors per tumor cell with a Kd of 1.5 x 10(-8) M. At high concentrations of transferrin the binding was not saturable. Transferrin internalization by hepatoma cells was shown by time and temperature-dependent binding studies and by pronase experiments. Transferrin recycling was confirmed by the demonstration of a progressive increase in the cellular molar ratios of iron to transferrin and by chase experiments. Ammonium chloride interfered with iron unloading. The vinca alkaloid vincristine inhibited iron and transferrin uptake. The hepatocarcinoma cells appeared to lack asialoglycoprotein receptors and therefore internalized partially desialated transferrin by the regular route. Iron uptake from transferrin was markedly inhibited by the hydrophobic ferrous chelator 2,2' bipyridine but was relatively unaffected by the hydrophilic ferric chelator desferroxamine. The implication that ferrous iron was involved in postendocytic transvesicular membrane iron transport was supported by a study in which hepatoma cells were shown to take up large amounts of ferrous iron suspended in 270 mM sucrose at pH 5.5. The interaction at this pH between surface labeled hepatoma cell extracts and ferrous iron on a Sephacryl S-300 column suggested that the postendocytic transvesicular transport of iron through the membrane was in part protein mediated. The endocytosed iron in hepatoma cells was found in association with ferritin (33%), transferrin (31%) and a low molecular weight fraction (21%).

Inhibitory effect of nuts on iron absorption.

The effects on iron absorption of nuts, an important source of dietary protein in many developing countries, were measured in 137 Indian women. When the absorption from bread and nut meals (walnuts, almonds, peanuts, and hazelnuts) was compared with that from bread meals, the overall geometric mean absorption from the nut meals (1.8%) was significantly less than from the bread meals alone (6.6%, t = 9.8, p less than 0.0005). In contrast, coconut did not reduce absorption significantly. All the nuts tested contained significant amounts of two known inhibitors of Fe absorption (phytates and polyphenols) but the amounts in coconut were significantly less than in the other nuts. Fifty milligrams ascorbic acid overcame the inhibitory effects of two nuts that were tested (Brazil nuts and peanuts). This is different from that found previously for soy protein, another potent inhibitor of Fe absorption.

Effect of traditional oriental soy products on iron absorption.

The effect of a variety of traditional oriental unfermented and fermented soy products on iron absorption was evaluated in 242 Indian women. When compared with a soy-flour meal containing an equal amount of protein, iron absorption was found to be significantly improved with silken tofu, natto, tempeh, rice miso, barley miso, and soybean miso. This improvement could not be adequately explained except with reference to changes in the protein composition of the products. The protein fraction profiles of the soy products were obtained by size-exclusion high-performance liquid chromatography. An inverse relationship between food iron absorption and the high-molecular-weight fraction of the soy products was demonstrated (r = 0.66, p = 0.01). However, anomalous results obtained with three products (sufu, tempeh, and fully hydrolyzed isolated soy protein) did not make it possible to reach firm conclusions concerning the effect of the protein fraction of soy on iron bioavailability.