The field of toxicology has been criticized in recent years for using high-dose animal toxicology studies to make judgments regarding human health after exposure to materials and chemicals. Risk assessments have also been made with little attention given to mechanisms. The new field of molecular toxicology is making profound changes in the way data will be accumulated in order to understand the impact of exposures on human health. OST is increasing the use of these molecular methodologies and techniques to answer questions which may ultimately enhance CDRH risk assessments of medical device materials. The ultimate goal is to reduce the uncertainties in extrapolating from in vitro to in vivo, animals to humans, and high dose to low dose. These types of studies will allow OST scientists to remain at the forefront of the new field of molecular toxicology.
The primary focus of these OST research activities continues to be basic and applied toxicology analysis of medical device materials. The toxicology research and risk assessment program has been developed into two major components: (1) toxicology methods development, and (2) tissuematerials interactions. Each of these research programs are designed to directly address fundamental scientific and regulatory issues of the biological effects of medical device materials and systems.
OST toxicology activities have been successfully directed primarily towards two general goals: (1) developing more efficient or more sensitive toxicology methods, and (2) applied toxicology analysis of specific materials and medical devices.
Return to the Table of Contents.
Key words: HPLC, latex, leachables, accelerators
An analytical method has been developed to quantify leachable latex accelerators. It has been verified that the multi-residue method works in mixtures of standards. The three types of latex accelerators are MBT, TMTD, TETD and ZDEC. This simultaneous quantitative approach is evaluated by using photo diode array (PDA) and electro-chemical (LCEC) detectors to obtain quantitative results. Linear least-squares slope and intercept of determined vs. prepared concentrations for the four-components were 1.00±0.01 and the intercepts passed origin. The interferences and instability of the accelerators were overcome by this method. High-precision of less than 2% of rsd values were achieved with major agreement from both detectors. This method allows level detection limits of parts per billion (ppb) to be reached. The ASTM D11.40 Committee has modified their protocol due to OST research impact and suggestion.
The four-component mixtures can be stored at room temperature for 24 hours with the relative error coefficient (rec) less than of 0.3%/hour and 0.5%/hour for the most unstable compound, ZDEC. The degree of dependency to which the analyte concentration in single-component and in mixtures affects the results was assessed via rec values. The differences of the rec values between the single and the mixtures for the four accelerators were among 1%/mg.(mL)-1.
The four-component recoveries of the spiked extractions from latex gloves were very satisfying, demonstrating the feasibility of OST analytical methods under a truly applicable condition. [Stds]
Return to the Table of Contents.
Key words: blood-contacting medical devices
During the past 3 years, OST has extensively reviewed published reports on the hazards of complement activation by blood-contacting medical devices. This information was organized into a number of subtopics relating to medical device type and relevant biological systems. This database will be used to design a research program addressing key regulatory questions not dealt with by industry or academia. [ProA]
Return to the Table of Contents.
Key words: cytotoxicity, biomaterials, genotoxicity
Few studies exist on the effects of long-term incubation of implant materials under simulated endogenous conditions, i.e. after contact with biological fluids such as blood. A collaborative study involving two OST branches, HSB and MBB, examined the cytotoxic and genotoxic potential of medical grade polyester urethane (PU) and polyethylene (PE) after incubation for different lengths of time up to 8 weeks under several different conditions designed to model physiological conditions. Contemporary technologies and simplified assays were used to test for cytotoxic and genotoxic effects. While PU degraded into cytotoxic material, none of the extracts exhibited a measurable genotoxic effect, using spot tests for mutation induction in Salmonella and induction of the SOS response in E. coli. The experiments provide a model for testing of implantderived extracts, and indicate that detectable levels of genotoxic materials were not generated during a relatively extended incubation period. [ProA]
Return to the Table of Contents.
Key words: nitric oxide, genotoxicity
Nitric oxide (NO) is a bioregulator and toxin induced in the immune, cardiovascular, neuronal and other systems. It is the endogenous mediator of toxic shock resulting from certain bacterial infections (e.g. related to tampon use). Current therapeutic uses include the control of blood pressure (drugs) and the treatment of respiratory insufficiency (devices instilling NO gas directly into the lungs). Experimental applications related to treatment of autoimmune disease, arthritis, male impotence, and other conditions make it likely that use will increase in the future. NOdelivering devices are currently under consideration for reclassification from class III to class II in CDRH, and new devices are under review. Studies in OST, in collaboration with scientists at NCI, have established that NO gas and drugs that generate it, including nitroglycerine, are mutagenic. Therefore it is necessary to understand the genetic toxicology of NO in order to assess the benefits and risks of NO-based therapies. Experiments in OST seek to determine the cellular requirements for NO mutagenesis and the characteristics of NO-induced genetic damage. Our results in two related species of bacteria have shown that very different mechanisms are operating in the generation of mutations. In Salmonella, 80% or more of the mutations are GC to AT transitions, while reversion of the same sequence in E. coli is more than 80% at AT base pairs. The predicted pathway for mutation generation, via cytosine deamination, may be occurring in Salmonella, but is not the mechanism in E. coli. Unlike most genotoxins, NO-action is not showing mechanistic conservation within many systems and cannot readily be predicted in humans. These studies provide the experimental methods for evaluating potential genotoxic activity of bioregulators as part of the regulatory review process. [ProA]
Return to the Table of Contents.
Key words: immunotoxicity, guidance
OST, with ODE, prepared an overview of the current role of immunotoxicity testing in regulatory review of medical device applications. Based on this survey, OST and ODE formed a committee to develop an "Immunotoxicity Testing Framework" that will provide reviewers and manufacturers with a coherent process for optimizing immunotoxicity testing. A draft framework has been prepared and made available for comment both within and outside FDA. [PreME, ProA]
Return to the Table of Contents.
Key words: HPLC, standards, latex, methods
ASTM Committee D11.40 Chemical Sensitivity Task Group met five times during the 1996 fiscal year. The standards for sodium salt of diethyldithiocarbamic acid, 2-mercaptobezothiazole, tetramethylthiuram disulfide and sulphenamide had been evaluated by HPLC method with various combinations of mobile phase, detecting UV wave length, and the stability of the standards. The task group also studied the extraction of specially prepared latex samples using the HPLC method to determine those sensitive chemical compounds in the latex samples.
From these studies, the task group discovered the carbanmate standard is not very stable in a working solution for longer than 2 hours. Different buffer solutions with various pH values were also studied. The HPLC result showed three different peaks in the chromatogram of carbamate standard which complicates the analytical result. The analytical procedures for the other standards were straight forward.
In the mean time, the task group had conducted a colorimetric method study by using copper sulfide to evaluate the alternate method for the carbamate determination.
OST/DMMS had participated in most of the meetings to discuss both round robin study results and laboratory studies. DMMS was also involved in rewriting the round robin protocol for the HPLC method and the colorimetric method. [Stds]
Return to the Table of Contents.
Key words: pharmacokinetic modeling, TDA, biomaterial degradation, computer simulation, carcinogenicity
Physiologically-based pharmacokinetic modeling, also known as PBPK, has been widely used in the study of drugs. But to date, little progress has been made in applying this approach to assess the hazard of medical devices. A known mutagen and rodent carcinogen, 2,4-Toluenediamine (2,4-TDA), was found as a degradation product of the polyester urethane cover of the Meme breast implants in vivo. The carcinogenicity of 2,4-TDA was studied in mice and rats and was found to be dose-dependent and vary with the route and species tested. PBPK modeling is used in this study to describe the fate of 2,4-TDA in the rat. The model was calibrated with data derived from studies of 2,4-TDA related to the biodegradation of the PU foam. Advantages and disadvantages of using this computer-simulated PBPK model as a tool to predict chemical exposure from biomaterial degradation are discussed. This model of ordinary differential equations was solved using Mathcad PLUS 6.0 (MathSoft, Cambridge MA). [ProA]
Return to the Table of Contents.
Key words: leachables, biocompatibility, polymers
A search was initiated on 14 polymers mostly used in medical device. The search was conducted to identify and quantify leachable substances from polymers and their biological effects. The search is limited to chemical and trade names, CAS number, manufacturers, chemistry, manufacturing and processing, sterilization, leachables and toxicology. This survey will provide CDRH a better estimate of the chemical exposure in order to formulate policies regarding the biocompatibility of polymers. [PreME, Stds]
Return to the Table of Contents.
Key words: particles, cytokines, wear and degradation, macrophage
The tissue/material program is designed to investigate the influence of particulate prosthetic materials on the immunological functions of macrophages including antigen recognition, direct tumoricidal activity and production of regulatory cytokines essential for the generation and activation of other immune effector cells.
Corrosion and wear of implanted medical devices (e.g., orthopedic prostheses) may produce particulate debris, leading to acute and chronic inflammatory responses in the host. Additionally, polymeric particles such as polytetrafluoroethylene (PTFE) may be injected directly into patients for clinical indications. In the presence of the biomaterial and/or wear particles, host monocytes/macrophages are activated and synthesize or secrete mediators of inflammation. In order to understand the mechanisms underlying the host immune response to particulates and device-associated infections, OST investigators have focused on the impact of medical device particles on macrophage function, because these cells play a pivotal role in the body's response to foreign bodies and their interaction with other cellular components of the immune system. In order to evaluate the effects of particles and chemicals of medical device materials on functional activities of macrophages, OST developed a test system (cultured murine macrophages; RAW 264.7) and used it to evaluate the inflammatory potential of particles and chemicals prepared from medical device materials, such as PTFE, titanium oxide, hydroxyapatite (HA), polymethylmethacrylate, and cadmium (both soluble and particulate forms), SiO2, and formed silica, diamond particles, and particulate dental materials (Bis MGA).
The murine macrophage cell line has two features that make it advantageous for this work: (1) macrophages are important cells in the immune system, and (2) macrophages phagocytize particles and secrete biological products (cytokines) that recruit other cells within the immune system. Thus, macrophages represent a good model for the study of biomarkers of immunotoxicologic effects of medical device materials implanted in the body.
OST scientists have assessed the effects of several particles and chemicals used in implanted medical devices (such as particles from metals and polymers) on macrophage antibacterial activity. Macrophage cells were exposed to particles or chemicals with and without lipopolysaccharide (LPS), and were evaluated for cytotoxicity, and production of nitric oxide and TNF-a, an inflammatory cytokine.
Our results indicate that TNF-a is induced by PTFE, PMMA, TiO2, HA, and SiO2 particles and by LPS and LTA. Nitric oxide (NO) production is not observed by the presence of only the particles to the murine macrophages. Addition of LPS or LTA to the particles increased TNF-a and NO production several-fold greater than the addition of any inducer alone. Addition of LPS and PmmA stimulated NO production in a dose-dependent response; LPS and increasing HA concentrations stimulated NO production but not in a dose-dependent response, and increasing TiO2 concentrations inhibited NO production. The production of NO by RAW cells in response to LPS is affected differently by these materials. This in vitro system is being considered for incorporation into developing ASTM standard on Biological Response to Particles (FOY.16.01). This demonstrates that a minute amount of LPS, such as would be found associated with a bacterial infection at the site of implanted biomaterials, induces a significant release of both TNF-a and NO by macrophages. Data also suggest that immunocompromised patients, such as AIDS patients, are at greater risk of inflammation and device-associated infections.
Another area of fruitful research involves a series of studies evaluating the effects of particles of titanium (Ti), cobalt-chrome (Co-Cr), stainless steel (SS), polyethylene, and hydroxyapatite (HA) on macrophage and lymphocyte function in vitro. Studies with Ti, SS, Co-Cr and HA have been initiated. Ti particles and TiO2 are phagocytized by macrophages. Dark deposits of Ti are clearly evident in the cultures. These results may help to explain why the titanium debris in tissue is black even though it is expected to be TiO2. Assays for NO revealed that this is being produced in response to Ti particles studied so far. NO is not being produced; thus there is no oxidative burst in these cells with the other particles. This is consistent with previous work demonstrating that titanium has an interaction with the oxidative burst different from other particles.
The response to particulate debris at the site of an implant is often a granulomatous response with abundant giant cells. Studies are being conducted in order to evaluate the role of giant cells and granulomas in the damage of host tissue or organs. An initial study obtaining appropriate cells from the peritoneal cavity has been done. Implants made of agar, with and without particles of HA, SS, and TiO2 were implanted into mice. The implants were removed 2 weeks later and set up in tissue culture. The implants had a mix of fibroblasts, macrophages, and lymphocytes on the surface. These cells grew in culture for more than a week. Thus, this will be an in vitro simulation of the response to an implanted material.
Estrogenic Effects of Medical Device Materials
Key words: endocrine disruption, estrogen, biomaterials, transfection assay, phthalates, acrylates
Some of the materials found in medical devices have been found to have estrogenic effects (DEHP, silicone). Estrogens are powerful hormonal growth regulators active in very small amounts; thus very small amounts of estrogen-like compounds from medical devices may have undesirable effects in patients and may need to be considered in CDRH device evaluation and risk assessments. The issue of estrogen exposure should concern not only CDRH but also the National Institute of Dental Research at NIH and the Environmental Protection Agency, as they have expressed concern about health effects of estrogens in foods and the environment. Studies on the effects of estrogens on implant-associated materials have taken two forms: OST scientists have examined the estrogenic effects of compounds using an assay in a whole animal system and an in vitro molecular marker system; and they have begun a study to determine if a more fundamental assay at a molecular level can be developed.
OST scientists examined the estrogenicity of a series of phthalates (esters of phthalic acid that are used as plasticizers). Phthalates were chosen to start a series of studies on implant materials because one of the group (diethylhexylphthalate) is reported in the literature to have estrogenic activity in cell culture systems. Phthalates are found in some cardiac catheters and other medical device plastics.
OST scientists have employed a whole animal assay involving young female mice. Mice were treated with the test material by injection in the nape of the neck (in a corn oil suspension medium for solids; directly without dilution for liquids). OST scientists sought to obtain the maximum dose possible. Mice were dosed daily for 4 days, then sacrificed on the fifth day. The uteri were removed, blotted, and weighed. Uterus-to-body weight ratio was taken as the measure of estrogenic or anti-estrogenic effect, as the uterus has a high concentration of estrogen receptor and is one of the most sensitive tissues to estrogen. The uterus demonstrates two basic responses to estrogen: it swells by influx of water into the lumen of the uterus, and it increases in weight by the growth of the endometrial layer of the uterus. The technically easiest feature to measure is the weight of the uterus after it is blotted. This is compared to the whole animal body weight to compensate for differences in size of the animals.
Preliminary studies in intact mice suggested that some of the phthalates might be estrogenic, although the variability was too high. However, because of this suggested activity, OST examined these compounds in the mice whose ovaries were removed. Scientists examined diethylhexylphthalate, poly diallyl isophthalate, terephthalate, and dicyclohexyl phthalate in this assay and did not detect significant estrogenic activity from these phthalates (significance was obtained for the positive control, estradiol).
OST scientists did not demonstrate estrogenic/anti-estrogenic effects from any of the compounds that they tested, despite literature reports of positive results from cell tests for at least one of the phthalates tested. This demonstrates the need for whole animal tests, as opposed to just cell culture tests in assaying the potential for pharmacological effect from implanted materials. The different results may have been due to the effects of the whole animal physiology on the endpoint. Another factor may have been solubility: the phthalates dosed into the animals via the subcutaneous route may not have been well absorbed and may largely have remained at the dose site. This route of dosing may be the most applicable to the manner in which humans might be exposed to phthalates via devices.
OST scientists examined several phthalates and methacrylates in a transient transfection assay in MCF-7 cells in collaboration with the Center for Veterinary Medicine (CVM). This assay links a receptor gene to an estrogen responsive element in MCF-A cells, a breast cancer cell line that is estrogen responsive. While positive controls were active, none of the phthalates or methacrylates tested were, thus indicating that they do not interact with the estrogen receptor.
An assay that would combine the use of whole animals with a more direct cellular response to estrogens would combine the best of both the uterine swelling assay and the cell culture techniques. OST scientists have been using some of the tissues from the test animals to begin to develop such an assay. Estrogen binds to its specific cellular receptor and then the estrogenestrogen receptor complex binds DNA at specific "estrogen responsive elements" on the DNA to initiate the molecular events leading to the eventual expression of its physiological effects. The estrogen receptor is available in the cell as a complex with "chaperon" proteins. These are thought to keep the receptor protein folded in such a way that it is receptive to binding estrogen but unable to bind DNA. When estrogen binds the receptor, at least some of these chaperon proteins are released from the complex, and the receptor changes the way it is folded, so that now it can bind with the DNA at the estrogen responsive elements. OST scientists are trying to develop an assay that detects an increase in the level of one or more of these chaperon proteins in response to estrogen treatment of the whole animal.
These chaperon proteins turn out to be "heat shock proteins" (hsp), originally discovered because they are released when cells are shocked by immersion in hot water, as a stress response to defend the cell from a potentially lethal stimulus. In particular, hsp72 and hsp90 are involved with the estrogen receptor. These can be detected by immunoblotting techniques. To date, the assay consists of treating animals with test compounds or control materials for 4 days, sacrificing the animal on day 5 and removing and weighing the uterus. The uterus is frozen for later use, eventually thawed and homogenized, solubilized, and proteins separated on electrophoretic gels. The gels are treated with a Western blot procedure and probed with antibodies to the hsps.
OST scientists have been able to demonstrate a clear association between significant increased levels of hsp72 and estrogen treatment of the animal. They are also studying with similar means changes in the level of estrogen receptor proteins (which are regulated by the amount of estrogen present) to see if this is another viable marker. However, the regulation of the estrogen receptor may prove too complex to make a reliable assay. A protocol has just been approved to continue these preliminary studies into the development of an assay that is fundamental, specific, and sensitive. [ProA]
Return to the Table of Contents.
Target Tissue Biomarkers
Key words: biomarkers, mercury, stress proteins, whole embryo culture, avian embryos, gene expression, protein synthesis.
An important part of FDA's mission is to facilitate the development, refinement, and validation of more sensitive and predictive preclinical methods in toxicology. The goal of the toxicology test methods development program is to develop methods to assess 1) exposure to insults early in the intoxication process, and 2) the potential hazards associated with exposures to these materials. One approach is to develop methodologies which define biomarkers of exposure and toxicity. Identification of biomarkers and understanding their role in cell responses and damage can provide information that will be essential for predicting, and therefore preventing, adverse effects associated with the use of chemicals and materials used in medical devices.
OST investigators are focusing on developing markers at the molecular level because these changes are usually the first responses affected by toxicants. Current studies are focusing on developing the "stress" protein response as a more sensitive assay to detect potentially toxic exposures to medical device materials and on predicting the potential adverse effects of device materials more reliably . Changes in expression of particular gene products could serve as a biomarker which may provide information on the potential toxicity of device materials, and consistent patterns of changes could be used to define "fingerprints" that are specific for a toxicant or class of toxicants. The patterns of change may also provide insight into the mechanisms of toxicity of these materials. The significance of this work is that improvements in molecular methods and integrating them into routine toxicology testing regimes will enable FDA and manufacturers to enhance identification of toxic materials, and improve assessments of risk associated with material use in medical devices.
One OST project has focused on in vivo administration of liver and kidney toxicants. These two organs, which have major excretory functions, are reversibly or irreversibly damaged by a variety of chemicals and materials, including metals, while attempting to remove toxic materials from the body. Previous studies in our lab demonstrated that the expression of inducible gene products, the stress proteins, in response to cadmium (hepatotoxicant) and mercury (nephrotoxicant) is targettissue specific.
In an effort to further develop this method as a biomarker of exposure and toxicity this past year, OST is attempting to determine if these molecular responses are target cell-type specific after exposure to toxic metals, including mercury, which constitutes 50% of dental amalgam. Immunohistochemical (IHC) localization of stress proteins using monoclonal antibodies is being undertaken in tissues where the stress response has been studied previously, i.e., liver and kidney. The advantage of this system is that it is more sensitive than gel electrophoresis analysis of whole tissue lysates after radiolabeling of proteins. Most importantly, this IHC technique will allow OST scientists to expand present research examining tissue-specific induction of stress proteins by making it possible to assess the cell-type specificity of the stress protein response within a given target tissue. This increase in specificity should enhance the use of the stress response as a biomarker of cellular perturbation which results from exposure to hazardous materials. The parallel question to be answered: Are cells which produce the stress proteins targets for the inducing agent?
Immunoblotting (Western blot) procedures demonstrated that inducible hsp72 reactivity is present in the renal cortex and outer stripe of the medulla. This is the location of many of the S3 proximal tubule cellsknown target sites for mercury. The expectation was that IHC procedures would show that hsp72 is induced in these target cells. Contrary to this hypothesis, enhanced hsp72 was expressed in undamaged distal tubules and proximal convoluted tubules, which are not affected by mercury. There was minimal hsp72 expression in the proximal segments containing the S3 cells. Results have also confirmed a known activity of hsp72 under stress, i.e., translocation to the nucleus to protect the DNA/RNA/protein synthesis machinery. These studies are significant in confirming the literature that stress proteins play a role in cellular defense; and conversely, cells that are unable to express hsp72 in response to stress may be more susceptible to injury.
In the developmental toxicology biomarkers program, OST staff, in collaboration with EPA personnel, are developing a short-term in vitro test (utilizing the rat whole embryo system) to evaluate the utility of stress proteins as biomarkers of exposure or toxicity. The stress protein response is recognized as a broad cellular response indicative of cell stress or adaptation to stress. The test examines the temporal relationships between altered gene expression (as assessed by the enhanced synthesis of stress proteins) and the production of malformations. The ultimate focus of the project is to determine if stress protein synthesis induced by heat, a teratogenic insult, can be used as biological markers of exposure to and/or effect of developmental toxicants. Our studies have indicated that, both in vivo and in vitro, the temperature at the target site (embryo) is critical. A primary defect of brief hyperthermia on gestation day 10 involves altered somite development within the first 24 hours and ultimately altered skeletal development. Stress protein synthesis was induced prior to the development of defective somites. The affected somites are not yet formed at the time of heat stress.
Current studies, both in vivo and in vitro, have been performed to confirm the presence or absence of heat shock band on Western blots at 41oC, 42oC, and 37oC, using HSP antibody 72 and HSP 94. Both in vitro and in vivo experiments have been assayed with four different antibodies on Western blots (HSP 90 family). In addition, experiments to delineate de novo synthesis of heat shock proteins are currently being performed using 84kD and 86kD heat shock antibodies from a different source. Further in vitro whole embryo culture experiments have been performed for developmental morphology of somite development.
Latex-associated Allergies
Key words: latex, allergy, protein measurement
Natural latex in medical devices frequently induces a Type 1 allergy which may be life threatening in individuals highly sensitized to latex proteins. Although awareness of allergy to latex proteins has increased in the last 5 years, the prevalence of latex sensitization is increasing in the general population. One of the most prominently affected groups, due to the frequent and extended exposure to latex gloves, is health care professionals. The prevalence of latex allergy in this group is estimated to be up to 15%. Due to the severity of reactions, sensitized individuals either avoid using latex gloves or are forced to change the profession. The potential for reduced usage of protective latex products (gloves and condoms) by increasing numbers of sensitized individuals may have a detrimental effect on the further spreading of HIV and other infections.
General research efforts are focused on identifying allergenic proteins and reducing protein levels on finished latex products. Present laboratory and clinical studies include several in vitro and in vivo approaches for evaluation of allergenic potential of latex products and identification of latex sensitized individuals. However, the clinical relevance of various in vitro methods has not been established, and the identity of specific allergen(s) had not been determined yet.
The aim of this project is to evaluate the specificity of the anti-latex IgE antibodies in human sera reacting with latex proteins from various sources. These findings will be correlated with the medical history of test subjects and specific exposure profile. This study should reveal either: (a) existence of major allergenic protein(s) which is present in any latex product, or (b) that the specificity of allergenic response depends on the type of product and the pattern of exposure.
Studies are in progress to evaluate the relationship between the total protein content on latex products and potential allergenicity. Latex protein extracts from three major sources (nonammoniated and ammoniated raw latex and latex products) were comparably evaluated for the total protein content, the level of allergenic proteins using pool of human sera, and the intensity of skin reactions in sensitized individuals. When the skin testing was performed with three extracts where dosing was normalized based on the total protein levels, the intensity of skin reactions was the same for all extracts. These findings indicated that the total protein level may be a reliable measure of the potential allergenicity of latex products. Correlation of the total protein measurement and a RAST inhibition test, as an in vitro measure of allergenic potential, is in progress. To confirm this finding, OST is extending the study further to make similar comparisons of two in vitro tests on the wide range of extracts from finished latex products. Protein extracts from surgical and examination gloves and other latex products (15 total) will be evaluated for total protein levels and allergenic protein levels. Selected samples from this group will be evaluated by skin testing also.
In efforts to develop a more sensitive method for reliable measurement of protein levels on the medical devices, OST investigators performed studies to generate pilot data for enzymatic assay of latex proteins. Several attempts to develop an ELISA for latex proteins raised question of appropriate source of latex protein to represent standard reference antigen. They evaluated a series of sera from rabbits immunized with latex proteins extracted from ammoniated and nonammoniated latex and from latex glove. Capacity and selectivity of protein recognition by those sera was evaluated by immunoblotting. The results demonstrated that the largest number of protein molecules in extracts from various sources were recognized by serum from NAL-immunized rabbit. Both AL- and GL- extract immunized rabbit sera reacted with a lesser number of proteins. Among proteins not recognized by these two sera were some that were identified as allergenic proteins by a pool of human immune sera. The results suggested that NAL should be the most complete source of latex proteins representing all potential allergens. [ProA]