Avidin conjugated to the fluorescent dyes rhodamine or fluorescein binds to mast cell granules in rodent and human skin. Sequential staining of tissue mast cells first with conjugated avidin, and then with a metachromatic stain demonstrated that both techniques ide ntify the same mast cell granules. Specificity for mast cells was confirmed by the absence of avidin-positive cells in the skin of mast cell-deficient (W/Wv) mice. Binding of conjugated avidin to mast cells was inhibited by

more »

... g tissue specimens with unconjugated avidin but not by pretreating conjugated avidin with biotin, indicating that avidin does not bind to biotin or a biotin-like molecule. Within murine dermis, unique patterns of mast cell distributions were observed, with a prominent perivascular localization in ear skin, and a complete absence of mast cells underlying the scales in tail skin. In tissue sections of g uinea pig skin undergoing basophil hypersensitivity reactions and in murine and human ski n specimens infiltrated with eosinophils, conjugated avidin selectively stained only dermal mast cells , demonstrating specificity for mast cells in sites of inflammation. Conjugated avidin also readily stained rat peritoneal mast cells, demonstrating its utility for identifying extracutaneous mast cells. Unlike the metachromatic stains, avidin binding to mast cells in tissues is not limited by methods of fixation or special embedding and cutting procedures. Thus, mast cell identification with conjugated avidin is a reliable, specific, and simple method with important clinical and investigative applications. The egg white protein avidin and naturally occurring biotin interact in an unusual stochiometric way, forming strong, noncovalent bonds that ex hibi t a n exceptioni:)Jly small dissociation constant of approximately 10-1 " M [1]. The strength and specificity of t hi s association has permi tted t he use of avidin a nd biotin as mutually see king ligands for the detection and quant itation of t issue substrates. One commonly used strategy employs biotin-conjugated molecules t hat bind to appropriate t issue substrates [2] [3] [4] . T he location of biotin conju gates are detected by avidin labeled with a n enzyme or a fluoresce nt dye . These techniques have been employed successfully using avidin conjugated to the dyes fluorescein isothiocyanate (FITC) and tetramethylrhodamine isothiocyanate (TRITC) or to the enzyme horseradish peroxidase (HRP). During the course of work with each of these reagents, we observed that conjugated avidin in the absence of exogenous biotin specifically identified t he granules of mast cells in cutaneous tissues and in isolated cell prepa rations. In this report we present data that characterize this unique specificity of avidin binding to rodent and human mast cells. MATERIALS AND METHODS Animals BALB/cCum mice 8-12 weeks old were obtained from Cumberla nd View Fa rms, Clinton, Tennessee. Mast cell-deficient W ; wv mice and their sex-and age-matched+ / + controls (WBB6F, / J-++) were purchased from Jackso n Laboratory, Bar Harbor, Ma ine. Sprague Dawley rats, barrier-derived and speci fic pathogen-free, weighing 150-200 g were obtained from Camm Laboratories, Wayne, New J ersey. Inbred fema le strain 2 guinea pigs, weighing 300-400 g, were obtained from Crest Caviary, Raymond, California. Reagents Avidin, biotin, and ovalbumin were purchased from Sigma Chemical Co., St. Louis, Missouri. Incomplete Freund's adjuvant (IFA) was purchased from DIFCO Labs, Detroit, Michigan. Giemsa stain and clophonium were obtai ned from Fisher Scientific Co., Dall as, Texas. Fluorescein isothiocya nate conjugated avidin (FITC-avidin ), tetramethylrhodamine isothi ocyanate-labeled avidin (TRlTC-avidin), and biot inylated monoclonal anti -Thy 1.2 antibody were obtained from Becton Dickinso n FACS Systems, Sunnyvale, California. Tissue Preparation Cutaneous spec imens from normal mice, rats, and guinea pigs were removed with a surgical blade after ge neral anesthesia; fixed in 10% buffered formalin ove rnight; was hed, and embedded in paraffin . Four mi cron -thick sections were prepared from all specimens, and deparaflinized with xylene and graded alcohols. Whole moun ts of dermis were obtained from surgically excised specimens by removing the underlying fat with blunt dissection, and incubating in 20 mM EDTA as reported prev iously [5] . After 4 h, the epidermis was separated mec ha nically from the dermis a nd discarded. The dermal specimens were fixed in acetone for 30 min at 23'C, divided in to 4 x 4 mm squares, and washed 3 times wi th 6.7 mM phosphate-buffered saline (PBS) pH 7.4. Immunization and Shin Testing of Guinea Pigs for Cutaneous Basophil Hypersensity (CBH) A gu inea pig was immunized for CHB with ovalbumin (OA), by the injection of 0.1 ml of a 1:1 saline emulsion of OA (250 1-'g/ ml) and IFA into each footpad. One week late r, the a nimal was skin tested by the intradermal injection into flank skin of 0.05 ml saline containing 50 1-'g OA. A similarly challenged nonimmune guinea pig served as the negative co ntrol for the CBH reaction. Ski n responses were measured 24 h later, and, as expected, t he CBH site ex hibited a characteristic reaction of erythema (16 mm in diamete r) with minimal induration [6, 7] while the negative control site had a barely perceptible, nonindurated, red macul e (5 mm). 214 Six-m illimeter punch biopsies were obtained from t he center of each skin test site and were divided in half. One portion was fixed in 10% neutral buffered formalin; the other in Helly's solution (Zenker-formol, pH 4.7) according to methods of Askenase et al [8] . The fixed specimens were then embedded in paraffin, and 4-J.Lm sections prepared. To identify infiltrating basophils optimall y, the Helly's-fixed tissue sec-