Evaluation of various substances to prevent adsorption of tuberculin purified protein derivative (PPD) to glass surfaces*

Bull. Org. mond. Sant 1970, 43, 91-106Wil. Wld Hith Org. )17,4,9-0 Evaluation of Various Substances to Prevent Adsorption ofTubercu linPurified Protein Derivative (PPD) to Glass Surfaces* S. LANDI, H. R. HELD & M. C. TSENG It is well known that a dilute tuberculin PPD solution (1 IU or S IU per dose) very rapidly loses its potency owing to adsorption oftuberculoprotein to the wall of the container into which it is dispensed. The amount oftuberculoprotein adsorbedper cm2 ofglass surface has been measuredfor phosphate-buffered saline over a wide pH range (pH I to pH 10). The maximum adsorption was found at pH 4 (0.31 pg/cm') and the least at between pH 6 and pH 10 (0.15 ,ug/cm2). The rate of adsorption of tuberculoprotein to glass was not changed when the phosphate-buffered saline was replaced by borate-buffered saline. Tuberculin PPD prepared by the ammonium sulfate precipitation method, by the trichloro- acetic acid precipitation method and by a combination of both methods adsorbed equally well to glass andno difference in the rate ofadsorptionfor these tuberculoproteins wasfound. Forty-two substances in addition to Tween 80 were testedfor their property to prevent adsorption of tuberculoprotein to glass in dilute tuberculin PPD solutions (50 lU/ml of 14C-labelled PPD). The most efficient anti-adsorption agents were found to be nonionic surfactants, some ionic surfactants and some colloidal substances; polypeptides and non- surface-active substances of low molecular weight showed little or no anti-adsorption property. The labelling ofPPD with 4C has proved to be a valuable tool, particularly for long- term adsorption studies andfor screening substances to be used as efficient anti-adsorption agents. These studies have permitted the selection ofagents which could be added to dilute solutions of tuberculin PPD (10 IU/ml to 500 IU/ml or 0.2 pg/ml to 10 pg/ml respectively) in order to avoid loss ofpotency due to adsorption. Parish & O'Brien as early as 1935 demonstrated that tuberculin is adsorbed to glass and this observa- tion has since been confirmed by other investigators. Magnusson et al. (1958) carried out an extensive study in guinea-pigs to determine whether dilute solutions of tuberculin PPD could be stabilized by using a substance that would prevent the adsorption of PPD to the wall of the container. They concluded that-adsorption could be prevented by adding 0.005% Tween 80 (50 ppm) to the buffer used to prepare dilute tuberculin solutions. More recently, Marks (1964) measured the adsorption of PPD to glass by using radioactive tuberculin PPD labelled with iodine (1311). In recent years radioactive tuberculin PPD labelled with 14C * From the Connaught Medical Research Laboratories, University of Toronto, Toronto, Canada. was prepared in the Connaught Medical Research Laboratories (Landi et al., 1967) and used to study the adsorption of tuberculin PPD to glass and plastic surfaces. It was found that the addition of 0.0005% Tween 80 (5 ppm) instead of 0.005% to tuberculin PPD preparations protected them effect- ively against losses of tuberculoprotein due to adsorption (Landi et al., 1966). The present work was carried out firstly to assess the effect of the following factors on the adsorption of tuberculin PPD to glass: pH and type of diluent used, concentration and type of PPD employed (prepared from the same filtrate by three different methods of precipitation); and secondly to evaluate the efficacy of various substances as possible anti- adsorption agents and to compare them with Tween 80. 2546 91 - 92 S. LANDI, H. R. HELD & M. C. TSENG MATERIALS AND METHODS Reagents The reagents used and their. source of supply are listed in Table 1. Phosphate-buffered saline, pH 7.38 Isotonic phosphate-buffered solution (Magnusson et al., 1958; Landi, 1963) at pH 7.38 was prepared by dissolving 1.45 g of KH2PO4, 6.10 g of Na2HPO,, and 4.80 g of NaCl in 1.05 litres of distilled water and autoclaving for 30 minutes at 121°C and 15 lbf/in' (1.05 kgf/cm'). Phosphate-buffered saline, pH S to pH 10 Buffer solutions of pH 5, 6, 7, 8, 9 and 10 were prepared by dissolving 4.80 g of NaCl and the same total amount of anhydrous phosphate (7.55 g) in 1 litre of distilled water. The pH was adjusted only by varying the ratio between primary, secondary or tertiary phosphate (NasPO,). These solutions were sterilized by filtration through a Millipore filter (HAWP 04700, pore size 0.45 i). Phosphate-buffered saline, pH I to pH 4 Buffer solutions of pH 1, 2, 3 and 4 were prepared by dissolving 4.80 g NaCl and 7.55 g KH2PO4 in 1 litre of distilled water and adjusting the acidity to pH 1, 2, 3 or 4 by the addition of concentrated hydrochloric acid. These solutions were sterilized by filtration through a Millipore filter. Borate-buffered saline, pH 7.2 Borate-buffered saline of pH 7.2 (without and with 0.04% acacia) was prepared according to the method of Gottschall & Bunney (1938). Preparation and purification of tuberculin PPD-14C Tuberculin PPD-14C was produced by adding 1 mCi of a L-amino acid-"4C mixture (New England Nuclear Corporation) filtered through a Millipore filter (pore size 0.45 ,) to 1000 ml of sterile Long's synthetic medium (Long & Seibert, 1926). The medium was dispensed in a 5-litre Povitsky bottle and seeded with a pellicle ofMycobac- terium tuberculosis var. hominis, Johnston strain, and incubated at 37°C for 6 weeks. After the contents had been tested for purity, the bottle was steamed for 3 hours in a Wilmot-Castle autoclave (100°C) and cooled to room temperature. The bacterial growth was separated from the medium by centrifugation and filtration. Tuberculin PPD-14C was prepared from the culture filtrate by precipitation with trichloroacetic acid (4% final concentration), redissolved in 0.07 M Na,HPO, solution and reprecipitated with an equal volume of a neutralized saturated solution of ammonium sulfate, using essentially the method described by Landi et al. (1967). This particular preparation of PPD-14C had a specific radioactivity of 0.25 ,uCi per mg and a relative potency of 0.88 (95% confidence limits 0.618-1.231) by comparison with the US Standard for PPD1 which is identical to the International Standard for PPD. The estimate of potency was made by the method of Long et al. (1954). The preparation also contained 0.7% nucleic acid by weight, which accounted for only 0.16% of the radioactivity of the preparation. This percentage was considered negligible and therefore no correction for the radioactivity due to nucleic acid was necessary when the radioactivity measured was expressed in terms of weight of PPD. Since the relative potency of the PPD-14C (0.88) used was not significantly different from the relative potency of Standard PPD (1.00) 0.00002 mg of PPD-14C has been taken as equivalent to 1 IU of Standard PPD.' Preparation and dispensing of tuberculin PPD-'4C solutions Solutions of tuberculin PPD-14C (50 IU/ml), with and without the substance to be tested for its anti- adsorption property, were prepared in the buffer solution of pH 7.38. Phenol (0.3%.) was added as a preservative unless otherwise stated. Some tuberculin PPD solutions of different pH and/or different tuberculin PPD-14C concentrations were also pre- pared. Aliquots (2.5 ml) of these solutions were dispensed into 10-ml glass ampoules which were then flame-sealed and stored at 5°C. Two ampoules were withdrawn after various times of storage and tested for their radioactivity in solution. Ampoules Ampoules (designation N51A break 12012 fi 2) were supplied by Owens-Illinois Inter-America Corp., Toledo, Ohio, USA. They had a volume of 10.1 cm" and an inner glass surface of 24.2 cm'. Their chemical 1 Obtained from the Division of Biologics Standards, Public Health Service, US Department of Health, Education, and Welfare, Bethesda, Maryland 20014, USA. I1 International Unit (IU) for tuberculin=0.000028 mg of the International Standard for PPD, consisting of 0.00002 mg PPD plus 0.000008 mg salts. EVALUATION OF SUBSTANCES TO PREVENT ADSORPTION OF PPD TO GLASS TABLE I SUBSTANCES TESTED FOR THEIR ANTI-ADSORPTION PROPERTY Group Type Supplier Sodium laurylsulfate, USP Fisher Scientiflc Co., Fair Lawn, N.J., USA Anionic Sodium deoxycholate Fisher Scientiflc Co., Fair Lawn, N.J., USA Benzalkonium chloride a Canadian Laboratory Supplies, Toronto, Canada Cationic Cetrimonium bromide Fisher Scientiflc Co., Fair Lawn, N.J., USA Cetylpyridinium chloride Canadian Laboratory Supplies, Toronto, Canada Surfac- Tween 20 b Atlas Chemical Industries, Branfford, Ontario, Canada tants Tween 40 c Atlas Chemical Industries, Branfford, Ontario, Canada Tween 60 c Atlas Chemical Industries, Branfford, Ontario, Canada Tween 806e Atlas Chemical Industries, Brantford, Ontario, Canada Nonionic Triton WR 1339f Ruger Chemical Co. Inc., Irvington on Hudson, N.Y., USA Saponin La Motte Chemical Products Co., Baltimore, Md., USA Brij 35 g Atlas Chemical Industries, Branfford, Ontario, Canada Triton X 100 h Packard Instrument Co. Inc., Downers Grove, Ill., USA Corexit 7654 Enjay Chemical Co., Houston, Texas, USA White sheets Fisher Scientiflc Co., Fair Lawn, N.J., USA Purified calfskin Fisher Scientlflc Co., Fair Lawn, N.J., USA Bacto Difco Laboratories, Detroit, Mich., USA Gelatins USP granular Fisher Scientlflc Co., Fair Lawn, N.J., USA Pro- Purifled pigskin Fisher Scientiflc Co., Fair Lawn, N.J., USA teins Pharmagel A, porkskin American Agricultural Chemical Co., Detroit, Mich., USA Davis B3S grade Davis Gelatine (Canada) Ltd, Scarborough, Ontario, Canada Human serum albumin Connaught Medical Research Laboratories, Toronto, Canada Peptone, USP bacteriological Matheson Coleman & Bell, Norwood, Cincinnati, Ohio, USA Polypeptides Bacto-Peptone Difco Laboratories, Detroit, Mich., USA Alcohol Glycerol, Reagent Canadian Laboratory Supplies, Toronto, Canada Amino acid Glycine Fisher Scientiflc Co., Fair Lawn, N.J., USA Phenol, Analar British Drug Houses Ltd, Poole, EnglandPhenols Chinosoli Eastman Organic Chemicals, Rochester, N.Y., USA a Alkylbenzyldimethylammonium chlorides (40% Cit, 50% Cu, 10% Ci%).b Polyoxyethylene (20) sorbitan monolaurate. c Polyoxyethylene (20) sorbitan monopalmitate. d Polyoxyethylene (20) sorbitan monostearate. e Polyoxyethylene (20) sorbitan monooleate. f p-iso octylpolyoxyethylene phenol polymer.U Polyoxyethylene (23) laurylether. h Octylphenoxypolyethoxyethanol. A biodegradable nonionic oil-silck dispersant. 8-Quinolinol sulfate. 7 93 94 S. LANDI, H. R. HELD & M. C. TSENG TABLE I (concluded) Group J Type Supplier Tris k Fisher Scientlflc Co., Fair Lawn, N.J., USA Dyes Evans blue Allied Chemical Corp., New York, N.Y., USA Starch, soluble, Lintner Fisher Scientiflc Co., Fair Lawn, N.J., USA Dextran (mol. wt 60 000-0 000) Nutritional Biochemistry Corp., Cleveland, Ohio, USA Polysaccharides Acacia (Gum arabic) USP Fisher Scientific Co., Fair Lawn, N.J., USA Agar Bacto Difco Laboratories, Detroit, Mich., USA Jaguar LV m Stein-Hall Ltd, Toronto, Canada Bacterial polysaccharide n Prepared in the author's laboratory Old Tuberculin (OT) Connaught Medical Research Laboratories, Toronto, Canada PVP K28-28 (mol. wt 34 000) 0 Irwin Dyestuff Division, Toronto, Canada PVP K90 (mol. wt 360 000) 0 Canadian Laboratory Supplies, Toronto, Canada Synthetic polymers Polyethylene glycol (mol. wt 400) Canadian Laboratory Supplies, Toronto, Canada Polyethylene glycol (mol. wt 3 350) Canadian Laboratory Supplies, Toronto, Canada Polyethylene glycol (mol. wt 17 500) Canadian Laboratory Supplies, Toronto, Canada Polyethylene oxide P British Drug Houses Ltd., Poole, England k Tris (hydroxymethyl) aminomethane. I C.l. No. 23860. m A gum preparation from the guar plant. nCrude polysaccharide from culture flitrate of Myco. tuberculosis. o Polyvinylpyrrolidone (PVP). V Product No. 29740. composition and physical properties were given previously (Landi et al., 1966). Assays for radioactivity From two ampoules, aliquots of 1 ml were plated in quadruplicate on stainless steel ringed planchets1 (31 mm diameter) and counted with a Geiger-Muller gas-flow counter fitted with a Micromil (Nuclear- Chicago Corp.) window. The arithmetic mean of the four countings was used to calculate the radioactivity in solution. The results given below in Tables 3 and 5-14 are expressed directly in percentage radioactivity in solution; the original radioactivity (at zero time) was taken as 100% and the coefficient of variation (Sokal & Rohlf, 1969) was 3.54%. When 1 ml of a 50 IU/ml solution of PPD-1lC in phosphate-buffered saline was plated, a radioactivity of approximately 110 counts/min above background was measured. 1 Planchets Inc., Chelsea, Michigan 48118. USA. The loss of radioactivity in tuberculin PPD-14C solution is attributed to adsorption of PPD-1lC on glass since the lost radioactivity was found on the inner glass surface (Landi et al., 1966). RESULTS AND DISCUSSION Adsorption of tuberculoprotein to glass at variouspH values It is known that proteins can be adsorbed on glass surfaces. In fact this phenomenon had been reported by Bull in 1956 and 1957 for bovine serum albumin and egg albumin, by Bingle & Czerkawski in 1964 for insulin and by Hummel & Anderson in 1965 for ribonuclease. It is also known that the pH of the diluent influences the amount of protein which can be adsorbed on glass surfaces, as demon- strated by Bull for serum albumin (1956) and egg albumin (1957) and by Hummel & Anderson (1965) EVALUATION OF SUBSTANCES TO PREVENT ADSORPTION OF PPD TO GLASS 95 TABLE 2 COMPARISON BETWEEN THE pH VALUES FOR MAXIMUM ADSORPTION OF PROTEINS TO GLASS AND THEIR ISOELECTRIC POINT lsoelectric point Maxlmum adsorption of protein to glass Protein pH Reference pH IAgIcm | Reference Bovine serum albumin 4.4-4.8 Neurath & Bull (1938) 3.94.6 0.33 Bull (1956) Egg albumin 4.5-4.9 Neurath & Bull (1938) 4.5 0.26 Bull (1957) Rlbonuclease 8.0 Merck Index (1968) 8.0 0.25 Hummell & Anderson (1965) TuberculoproteIn 3.84.7 Seibert et al. (1938) 4.0 0.31 Present report Seibert (1949) ADSORPTION OF TUBERCULIN PPD-14C TO GLASS IN RELATION TO THE pH OF THE SOLUTION a 04- 0.3- "" 0.2 \ 0.1 I 2 3 4 5 6 7 8 9 10 pH a 2.5 ml tuberculin PPD-14C solution (250 IU/mI; phosphate- buffered saline, pH I to pH 10) was dispensed In 10-ml glass ampoules and stored at 50C. The adsorption Is expressed in pg of tuberculoprotein per cm' of the total Inner glass surface of the ampoule. for ribonuclease. Therefore, it was of importance to determine whether the adsorption oftuberculoprotein to glass would be also influenced by the pH of the diluent. Thus the following experiment was carried out. Ten sets of 10-ml ampoules were filled with 2.5 ml of an isotonic buffered saline containing tuberculin PPD-14C (250 IU/ml). Each set had a different pH-namely, 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10. After various times of storage at 5°C and at 37°C, 2 ampoules from each set were withdrawn and tested for their radioactivity in solution. After 2 weeks of storage a constant level of tuberculin PPD (PPD, tuberculoprotein) 1 in the solution had been reached. From the loss of radioactivity in solution the amount of tuberculoprotein adsorbed to the glass surface was calculated. 1Since the tuberculin PPD (PPD) contained at least 95% tuberculoprotein, the three terms tuberculoprotein, tuberculin PPD and PPD have been used interchangeably throughout this work. The accompanying figure shows that at 5°C adsorption of tuberculoprotein took place over the whole range ofpH values tested from 1 to 10 and that the adsorption at pH 6 to pH 10 was at its minimum level (0.15 ,ug/cm'), while its maximum was at pH 4 (0.31 jug/cm2), which is in the range of the isoelectric point of tuberculoprotein (Seibert et al., 1938; Seibert, 1949). This correlation seems to exist for other proteins as well (Table 2) and it is also in agreement with the fact that the spreading of proteins is at a minimum at their isoelectric point (Manegold, 1953). At 37°C, the degree of adsorption of tuberculoprotein to glass was very similar to that at 5°C. Adsorption of tuberculoprotein to glass at pH 7.38 Since Connaught tuberculin PPD solutions at the strengths of 10 IU/ml, 50 IU/ml and 2500 IU/ml are prepared in a diluent of pH 7.38, it was of interest to determine the maximum amount of tuberculo- protein that can be bound to glass by adsorption at that particular pH and at various concentrations of PPD. Thus the following experiment was carried out. Several solutions of tuberculin PPD-14C (pH 7.38; 0.3% phenol) were prepared, each having one of the following strengths of PPD-14C: 10 IU/ml, 50 IU/ml, 100 IU/ml, 200 IU/ml, 500 IU/ml, 1000 IU/ml and 2500 IU/ml. A volume of 2.5 ml was dispensed in 10-ml ampoules (the inner glass surface of the am- poule was 24.2 cm'). After various times of storage at 5°C, 2 ampoules of each strength were withdrawn and their contents tested for their residual radio- activity. Table 3 shows that practically all the tuberculoprotein is lost by adsorption in tuberculin PPD solutions containing 50 IU/ml or less, while in those containing 500 IU/ml or more the loss became S. LANDI, H. R. HELD & M. C. TSENG TABLE 3 LOSS OF RADIOACTIVITY DUE TO ADSORPTION OF TUBERCULOPROTEIN TO THE GLASS SURFACE OF AMPOULES, IN RELATION TO THE STRENGTH OF THE TUBERCULIN PPD-'4C IN SOLUTION a Radioactivity In solution (%) Storage time 10 IU/ml | 50 lU/ml 100 lU/ml 200 IU/mI 500 lU/ml 1 000 lU/mi 2 500 lU/mi 8 hours 56 58 70 67 87 93 96 24 hours 35 39 53 69 90 93 g6 I week 9 9 44 60 90 94 95 2 weeks 7 6 41 60 90 94 98 1 month 6 10 23 60 81 90 97 3 months 10 6 33 55 84 88 97 6 months 7 7 35 56 87 91 98 Omonths 8 9 22 61 83 91 97 12 months 8 9 23 52 87 94 97 18 months 8 7 30 56 87 92 97 24 months 4 6 27 79 84 97 96 Average;, -24 months 7.0 7.6 27.4 60.8 85.6 93.0 97.0 a 2.5 ml tuberculin PPD- 4C solution (10 IU/mI to 2500 IU/mI; phosphate-buffered saline, pH 7.38; 0.30/0 phenol, no stabilizer) was dispensed in a 10-mi glass ampoule and stored at 5°C. TABLE 4 CAPACITY OF ADSORPTION OF THE SURFACE OF GLASS AMPOULES FOR TUBERCULOPROTEIN AT pH 7.38 Tuberculoprotein in solution at start of experiment 10 lU/ml 501U/ml L IU/mlUm 2001U/ml 5001U/ml I|000IU/ml|2500IU/ml Loss of radioactivity in solution (%) a 93 92.4 72.6 39.2 14.4 7.0 3.0 Tuberculoprotein in solution at start of 0.2 1 2 4 10 20 50 experiment (Gg/ml) Loss of tuberculoprotein in solution 0.186 0.924 1.452 1.568 1.440 1.400 1.500(Ag/ml) a Tuberculoprotein adsorbed on the glass 0.019 0.095 0.150 0.162 0.149 0.145 0.155 surface b (pg/cm') Average (100 IU/mI to 2500 lU/mi) 0.15 a Average loss from 6 to 24 months of storage (Table 3). b Total inner glass surface of ampoules was 242 cm'; flling volume was 2.5 ml. The capacity of adsorption (0.15 pg PPD/cm' of glass surface) was arrived at by averaging the tuberculoprotein adsorbed at 100 lU/mi to 2500 lU/mi. At 10 lU/mi to 50 lU/mi there was not enough tuberculoprotein Inthe ampouleto saturatethe capacity of adsorption of the glass surface. 1- IU- 0.00002 mg PPD. 96 EVALUATION OF SUBSTANCES TO PREVENT ADSORPTION OF PPD TO GLASS negligible. Table 4 shows that if enough tuberculin PPD is present in the solution to saturate the glass surface of the container, then the maximum amount of tuberculoprotein which can be adsorbed per cm" is constant at about 0.15 ug and therefore is inde- pendent of the concentration ofthe tuberculin PPD in solution. However, if the amount of tuberculin PPD in the solution is not sufficient, the tuberculoprotein adsorbed per cm2 will depend on the concentration of tuberculoprotein in the solution. In fact Table 4 shows that at the strengths of 10 IU/ml (0.0002 mg/ml) or 50 IU/ml (0.001 mg/ml) most of the tuberculo- protein was adsorbed to glass (93% and 92.4% respectively). However, 100% adsorption was not achieved and this may indicate that a small fraction of a non-adsorbable substance may be present in the tuberculin PPD-14C used. It should be noted that the adsorption oftuberculoprotein to glass is not peculiar to tuberculin PPD prepared from the culture filtrate of Myco. tuberculosis var. hominis since adsorption takes place also for PPD produced from other species of mycobacteria (Landi et al., unpublished data). Adsorption to glass of PPD prepared by different methods ofpurification The tuberculin PPDs which are used most exten- sively at the present time have been purified by either the trichloroacetic acid precipitation method, such as RT23 (Magnusson & Bentzon, 1958), Weybridge PPD (Green, 1946) and Connaught PPD (Landi, 1963) or by the ammonium sulfate precipita- tion method, such as PPD-S (Seibert & Glenn, 1941). It has been reported by Toman et al. (1968) that PPS-S which was prepared by ammonium sulfate precipitation is much less readily adsorbed than trichloroacetic-acid-precipitated PPD such as RT23. To verify this claim three '4C-labelled tuberculin PPDs were prepared from the same culture filtrate, one by the ammonium sulfate precipitation method, another by the trichloroacetic acid (TCA) precipita- tion method, and finally one prepared by a combina- tion of both methods. Table 5 shows that all three PPDs adsorbed quite readily to glass and that there was no significant difference in their rate of adsorp- tion. The tuberculoprotein content of each of these three preparations was at least 95% and the specific radioactivities were 0.20 ,uCi/mg, 0.18 uCi/mg and 0.17 ,uCi/mg for the tuberculin PPD-14C prepared by the ammonium sulfate precipitation method, by the TCA-precipitation method and by the combination of both methods, respectively. The relative potency of each preparation was not signifcantly different from the US standard for PPD. We therefore venture to say that, ifdifferences in keeping properties between PPDs produced by the trichloroacetic acid method and the ammonium sulfate method exist, they cannot be attributed to a difference in their surface activity. Effectofvarious diluents on the adsorption oftuberculin PPD to glass It has been reported that dilutions of PPD prepared with saline give smaller reactions than the corresponding dilutions prepared with phosphate buffer (Jensen et al., 1938; Magnusson et al., 1958). Therefore it was felt that different diluents might influence the rate of adsorption of tuberculin PPD to glass. For this reason an investigation of the stabiliz- ing properties of three widely used buffered diluents was undertaken-namely, the diluents devised by Gottschall & Bunney (1938), by Seibert & Dufour (1954), and by Magnusson et al. (1958) as modified by Landi et al. (1966, 1968). Table 6 shows that none of the diluents tested had any influence on the TABLE 5 ADSORPTION TO GLASS a OF PPD PREPARED BY DIFFERENT METHODS OF PURIFICATION EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Radioactivity in solution (%) PPD prepared byStorage PPD PPD prepared by TCA b (1st pre- time prepared by ammonium cipitation) and TCA b-pre- sulfate ammonium cipitation precipitation sulfate (2nd precipitation) 8 hours 62 52 51 1 day 48 43 48 7 days 14 18 21 I month 5 12 15 3 months 7 16 18 6 months 5 8 10 9 months 4 12 6 12 months 4 8 11 18 months 5 8 9 24 months 5 9 10 a 2.5 ml tuberculin PPD-'4C solution (50 lU/ml; phosphate-buffered saline, pH 7.38; 0.3% phenol) dispensed In a 10-mi glass ampoule and stored at 5°C. b TCA = trichloroacetic acid. 97, S. LANDI, H. R. HELD & M. C. TSENG TABLE 6 EFFECT OF VARIOUS DILUENTS ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-'4C TO GLASS EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Radioactivity In solution (%) Storage Borate-buffered Phosphate- Phosphate- time saline a buffered salineb buffered salinec (0.5% phenol, (0.5% phenol, (0.3% phenol, pH 7.2) pH 7.3) pH 7.38) 8 hours 52 51 51 1 day 49 47 48 7 days 22 19 21 1 month | 12 12 15 3 months 17 4 18 6 months 15 3 10 9 months 9 3 6 12 months 11 3 11 18 months 6 4 9 24 months 8 3 10 a 2.5 ml tuberculin PPD-'4C solution (50 IU/mI; borate-buffered saline; 0.5% phenol (Gottschall & Bunney, 1938)) dispensed in a 10-ml glass ampoule and stored at 50C. b2.5 ml tuberculin PPD-'4C solution (50 lU/mI; phosphate- buffered saline; 0.5% phenol (Selbert& Dufour, 1954)) dispensed in a 10-mi glass ampoule and stored at 50C. c2.5 ml tuberculin PPD-'4C solution (50 lU/ml; phosphate- buffered saline (Magnusson et al., 1958); 0.3% phenol) dispensed in a 10-mI glass ampoule and stored at 50C. rate of adsorption of tuberculin PPD to glass when their stabilizers, acacia (0.04%) (Gottschall & Bunney, 1938), Tween 80 (50 ppm) (Magnusson et al., 1958) and Tween 80 (5 ppm) (Landi et al., 1966, 1968), were omitted. Effect of Tween 80 and gelatin on the tuberculin skin reaction It is well known that a dilute solution of tuberculin PPD (10 IU/ml or 50 IU/ml-respectively, 0.0002 mg/ml and 0.001 mg/ml) is unstable and that the principal cause of instability is the adsorption of tuberculoprotein to the glass wall of the container into which it is dispensed. To prevent the adsorption Tween 80 has been extensively used for many years at the concentration of 50 ppm (Magnusson et al., 1958) or 5 ppm (Landi et al., 1966) in dilute tuberculin PPD preparations. However, somewhat conflicting findings have been reported in the literature concern- ing the effect of Tween 80 on the tuberculin skin reaction in comparison with reactions to tuberculin without Tween 80 or tuberculin containing gelatin. In fact, Magnusson et al. (1958) concluded that Tween 80 does not increase the size ofthe tuberculin reaction in BCG-vaccinated guinea-pigs. Landi et al. (1966) also established that a 100-fold increase in the concentration of Tween 80 (from 0.0005% to 0.05 %, or 5 ppm and 500 ppm respectively), did not affect the size of the skin reaction in BCG-vaccinated guinea-pigs. However, other investigators have found that Tween 80 modified the tuberculin skin reaction. In fact Asami &Kataoka (1966) reported that a differ- ence was observed in the effect ofTween 80 in guinea- pigs when different lots were used and that the older the Tween 80, the more marked was the modification of the tuberculin reaction. Guld & Roelsgaard (1965) found in man that Tween 80 renders the skin reaction somewhat softer and also alters its size. In the continued search for a perfect anti-adsorp- tion agent, gelatin was advocated by Toman et al. (1965, 1968). They showed that gelatin did not alter the tuberculin reaction in man. However, Bleiker & Griep (1965) found that gelatin had some drawbacks as compared to Tween 80, since solutions of PPD stabilized with gelatin increased the size of the reactions to a greater extent in persons with low- grade sensitivity than in those with strong sensitivity. It is obvious from these results that further studies are needed to find a better anti-adsorption agent. Therefore, 42 substances in addition to Tween 80 were tested for their anti-adsorption properties in the hope that among these would be found some that, besides having good anti-adsorption properties, would not alter the tuberculin skin reaction in man. Effect of various substances on the rate of adsorption of tuberculin PPD to glass atpH 7.38 This study was carried out in a phosphate- buffered saline (pH 7.38) because this buffer is used in Connaught tuberculin preparations and also because at this hydrogen-ion concentration we have shown that the least amount of tuberculoprotein is adsorbed to glass. The tuberculin PPD-14C (tubercu- loprotein) used was prepared by precipitation with trichloroacetic acid and reprecipitation byammonium sulfate since it has been shown that the adsorption property of the tuberculoprotein is independent of the method of precipitation used (Table 5). The ampoules were stored at 5°C for a period of 12-24 months. Nonionic surfactants. When polysorbates (Tween 20, Tween 40, Tween 60 or Tween 80) were added in 98 EVALUATION OF SUBSTANCES TO PREVENT ADSORPTION OF PPD TO GLASS TABLE 7 EFFECT OF NONIONIC SURFACTANTS ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-14C a TO GLASS EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Substance tested Radioactivity In solution (%) T Concentration After | After After After | After 1 After AfterType W( I day 1 1 week '1 month months 1l2months '18 monthsb '24 monthsb Control - 47 25 18 12 18 11 12 Tween 20 0.0005 92 98 98 102 100 99 99 Tween 40 0.0005 103 95 99 103 104 104 100 Tween 60 0.0005 98 103 100 104 104 105 99 Tween 80 0.0005 102 100 98 104 104 105 103 Tween 80 0.00005 70 65 58 60 55 54 ND Triton WR 1339 0.1 101 101 101 104 101 101 102 Triton WR 1339 0.01 100 100 100 96 94 94 98 Triton WR 1339 0.001 101 103 100 102 100 103 103 Triton WR 1339 0.0005 93 103 103 104 104 102 104 Triton WR 1339 0.0001 63 56 50 49 55 55 56 Saponin 0.01 94 100 92 100 104 104 103 Saponin 0.001 74 72 76 86 75 66 65 Brij 35 0.005 93 87 83 89 82 ND ND Brij 35 0.0005 91 90 79 80 77 ND ND Triton X100 0.005 93 85 93 90 86 ND ND Corexit 7664 0.005 98 95 95 91 82 ND ND Corexit 7664 0.0005 J 54 47 49 37 32 ND ND a 2.5 ml tuberculin PPD-1"C solution (50 IU/ml; phosphate-buffered saline, pH 7.38; 0.3% phenol) was dispensed In a 10-ml glass ampoule and stored at 5°C. b ND = Not determined. 0.0005% concentration (5 ppm) to a tuberculin PPD solution, it was found that each of these Tweens prevented tuberculoprotein from adsorbing to glass (Table 7). Another nonionic surfactant, Triton WR 1339, also at the concentration of 5 ppm, gave protection against adsorption of tuberculoprotein (Table 7). Stronger concentrations of Triton WR 1339 or ofthe Tweens were equally effective, while 1 ppm of Triton WR 1339 or 0.5 ppm of Tween 80 gave only partial protection (Table 7). A nonionic plant surfactant, saponin, at a higher concentration (0.01 %) also gave protection against adsorption of tuberculoprotein (Table 7) while 0.001 % saponin gave only partial protection. Three other nonionic surfactants were tested- namely, Brij 35 (Augier et al., 1967), Corexit and Triton X100. After 12 months' storage, Brij 35, Corexit and Triton X100 at a concentration of 0.005% were effective as anti-adsorption agents. However, while Brij 35 was also effective at the concentration of 0.0005 %, Corexit at this concentra- tion gave only partial protection. Anionic surfactants. When an anionic surfactant, sodium laurylsulfate, was added in different amounts (0.002%, 0.001 % and 0.0001 Y) to a tuberculin PPD 99 S. LANDI, H. R. HELD & M C.- TSEN TABLE 8 EFFECT OF ANIONIC AND CATIONIC SURFACTANTS ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-'4C TO GLASS EXPRESSED IN PERCEIN&TAGE RADIOACTIVITY IN SOLUTION Substance tested Radioactivity in solution (%) Type Concentration After | After After After After After a AfterypeI| (%) I day I week I month Omonths 12 months 18 monthsa 24 months a Control b 0 49 18 13 11 13 14 10 Sodium laurylsulfate b 0.002 102 100 102 83 85 76 79 0.001 82 77 80 72 66 75 65 0.0001 53 35 32 30 28 26 29 Sodium deoxycholate b 0.10 87 84 80 75 69 ND ND 0.01 64, 55 46 41 25 ND ND Controlc 0 38 17 15 9 15 13 10 Benzalkonium chloride c 0.050 72 72 73 68 72 69 72 0.005 92 85 78 83 71 69 59 0.0005 72 69 70 76 69 74 74 Cetrimonium bromide c 0.1 102 98 96 89 88 78 85 0.01 96 83 77 69 63 68 61 0.001 71 69 70 63 64 55 65 0.0001 89 87 82 78 83 76 76 Cetylpyridinium chloridec,(d 0.050 77 101 93 83 100 92 92 0.005 75 95 94 88 68 71 71 0.0005 38 30 41 36 41 31 23 a ND = Not determined. b2.5 ml tuberculin PPD-"4C solution (50 lU/mi; phosphate-buffered saline, pH 7.38; 0.3% pheflol) was dispensed in a 10-mi glass ampoule and stored at 50C. c Same as footnote b, except that no phenol was added. d Ampoules containing 0.005% or 0.05% cetylpyrldinium chloride showed some precipitation of that substance atS°C which redissolved when the ampoules were brought to room temperature. solution, it was found that only at the concentration of 0.002% was good protection against adsorption of tuberculoprotein to glass obtained (Table 8). Sodium deoxycholate showed appreciable protection only at a higher concentration (0.1 %). Cationic surfactants. Benzalkonium chloride (0.05%, 0.005% and 0.0005 %) and cetrimonium bromide (0.1 %, 0.01 %, 0.001 % and 0.0001 %) and cetylpyridinium chloride (CPC) (0.05 %, 0.005% and 0.0005%.) gave a fairly good protection against adsorption of tuberculoprotein to glass (Table 8). Since these cationic surfactants are also effective antimicrobial agents, no phenol was added in these experiments. However, while in the case of nonionic surfactants (e.g., Triton WR 1339; Table 7) there was a minimum concentration above which good protection against adsorption of tuberculoprotein was obtained there seemed to exist, in the case of benzalkonium chloride and cetrimonium bromide, a very wide range ofconcentrations where fairly good anti-adsorption effect was obtained. In other words, a 100-fold or even a 1000-fold increase in concentration of these two cationic surfactants did not seem effectively to increase their anti-adsorption effect. 100 EVALUATION OF SUBSTANCES TO PREVENT ADSORPTION.OF PPD TO GLASS TABLE 9 EFFECT OF VARIOUS PROTEINS-AND POLYPEPTIDES ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-'4C a TO GLASS EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Substance tested Radioactivity In solution (%) T Concentration After After After After After After AfterType | %)I( day I week I month months 12 months 18 months 24 months Control 53 37 31 24 25 28 White sheets gelatin 0.001 68 62 48 57 49 49 White sheets gelatin 0.01 79 67 66 68 61 64 White sheets gelatin 0.1 94 83 100 98 98 92 Calfskin gelatin 0.1 81 77 74 78 72 73 Bacto gelatin 0.1 89 89 96 96 89 89 Granular USP gelatin 0.1 94 95 102 101 101 97 Pigskin gelatin 0.1 92 90 100 94 94 89 Pharmagel A gelatin 0.1 92 96 101 98 103 91 Davis gelatin 0.1 82 79 78 84 75 79 Human serum albumin Peptone Bacto-Peptone 0.1 95 94 0.1 0.1 79 71 75 53 95 65 42 90 92 57 37 53 38 88 57 41 24 45 64 93 72 84 89 88 90 70 78 53 41 a2.5 ml tuberculin PPD-'4C solution (50 IU/ml; phosphate-buffered saline, pH 7.38; 0.3% phenol) was dispensed in a 10-ml glass ampoule and stored at 50C. This could possibly be attributed to a concentration- quenching phenomenon (Albert, 1965). Proteins. Table 9 shows that all of 7 different gela- tins (0.1 % concentration) protected against adsorp- tion of tuberculoprotein to glass but that some gela- tins protected better than others. At lower concentra- tions (0.01 % and 0.001 % gelatin) protection against adsorption of tuberculoprotein became insufficient. Human serum albumin (0.1 Y.) gave about equal protection against adsorption of tuberculoprotein as some of the gelatins at the same concentration (Table 9). Polypeptides. Peptone or Bacto-Peptone (0.1%) offered little protection against adsorption of tuber- culoprotein to glass (Table 9). Tween 80 in the presence ofgelatin. The combined effect of a mixture of gelatin and Tween 80 on the adsorption of tuberculin PPD to glass was tested, as suggested by Guld & Roelsgaard (1965). Table 10 shows that a mixture of gelatin (0.1 %) and Tween 80 (0.0005 %) prevented the tuberculoprotein-l4C from adsorbing to glass no matter whether the gelatin used was by itself a good anti-adsorption agent (white sheet gelatin), or less efficient one (calfskin gelatin) (Table 9). Polysaccharides. The presence of carbohydrates of high purity (0.1 % soluble starch or 0.1 % dextran) gave no protection against adsorption of tuberculo- protein to glass (Table 11). However, polysaccharides of lesser purity such as acacia (0.1 % and 0.04%) and agar (0.1 %) gave some protection against adsorption of tuberculoprotein to glass while 0.1 % Jaguar LV gave good protection (Table 11). The addition of a polysaccharide prepared from a culture filtrate of Myco. tuberculosis var. hominis also gave good protection against adsorption of tuberculo- protein to glass at the concentration of 0.1% (Table 11). Old Tuberculin. There is a widespread belief that Old Tuberculin (OT) solutions are remarkably stable compared with tuberculin PPD solutions. In fact OT contains among other metabolic products an appre- 101 102 S. LANDI, H. R. HELD & M. C. TSENG TABLE 10 COMBINED EFFECT OF TWEEN 80 AND GELATIN ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-"4C a TO GLASS EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Radioactivity in solution (%) Storage Tween 80 Tween 80 (0.0005%) Tween 80 (0.0005%) time Control (0.0005%) without with white sheets with calfskin gelatin gelatin (0.1%) gelatin (0.1%) 8 hours 69 103 101 99 I day 46 104 102 102 7 days 33 103 102 100 14 days 24 101 95 103 1 month 16 104 94 98 3 months 14 98 99 102 6 months 15 104 96 103 9 months 14 100 97 97 12 months 14 104 96 100 18 months 13 102 96 102 24 months 10 102 96 98 a 2.5 ml tuberculin PPD-'<C solution (50 IU/ml; phosphate-buffered saline, pH 7.38; 0.3% phenol) was dispensed In a 10-mI glass ampoule and stored at 50C. ciable amount of polysaccharide which was shown to possess anti-adsorption properties at the concentra- tion of 0.1 % (Table 11). Therefore, it was thought of interest to find out whether OT mixed with PPD, both products at the concentration used in tuberculin skin testing, would stabilize PPD. To verify this assump- tion 2.5-ml aliquots of a borate-buffered saline, without acacia and containing 50 IU/ml (1: 2000 dilution) of OT 1 and 50 IU/ml (0.001 mg/ml) of PPD-14C, were stored in 10-ml glass ampoules. Borate-buffered saline without acacia, containing 50 IU/ml or 100 IU/ml of PPD-14C, was used as con- trol. Furthermore, since OT is usually prepared for commercial use in a borate-buffered saline containing 0.04% acacia (Gottschall & Bunney, 1938), the same set ofsolutions as above (OT + PPD-14C, and PPD- 1'C alone) was also prepared with this Gottschall- Bunney buffer and dispensed in ampoules. Table 12 shows some stabilizing effect of the acacia, confirming what was already shown previously 1 The relative potency of Connaught OT was 1.11 (0.832- 1.506) as compared with the Third International Standard for Old Tuberculin. The estimate of potency was made by the method of Long et al. (1954). (Table 11); it also shows that, no matter whether acacia is present or not, the solutions containing 50 IU/ml of PPD-14C plus 50 IU/ml of OT had a larger percentage loss of radioactivity than the corresponding solutions containing 100 IU/ml of PPD-14C only. This would indicate that the metabolic by-products present in 50 IU/ml of OT, such as polysaccharides, lipids, nucleic acid and glycerol and salts, do not prevent adsorption of tuberculoprotein to glass at this concentration. Synthetic polymers. Polyvinylpyrrolidone (PVP) (0.1 %) of molecular weight 360 000 or 0.1 % polyethylene glycol of molecular weight 17 500 gave good protection against adsorption of tuberculo- protein to glass (Table 13). However, when homo- logues of lower molecular weight of these two polymers were used (PVP 34 000 and polyethylene glycol 3350 and 400), the anti-adsorption effect decreased with the decreasing of their molecular weight (Table 13). However, a different polymer, polyethylene oxide, of a greater molecular weight (4 x 106), at a concentration of 0.02%, gave little protection against adsorption of tuberculoprotein to glass (Table 13). EVALUATION OF SUBSTANCES TO PREVENT ADSORPTION OF PPD TO GLASS TABLE 11 EFFECT OF POLYSACCHARIDES ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-'4C TO GLASS EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Substancetested Radioactivity in solution (%) Concentration After After After After After After AfterType (%) I day 1 week 1 month 6 months 12 months 18 months 24 months Control a 0 49 18 13 11 13 15 10 Dextran a, b 0.1 53 44 38 16 13 15 10 Starch, soluble a 0.1 40 13 16 11 9 13 16 Acacia a 0.1 58 57 43 46 47 42 47 Acacia a 0.04 59 48 43 37 37 33 36 Agar a 0.1 69 54 42 31 37 28 35 Jaguar LV a 0.1 87 85 91 95 96 97 93 Crude polysaccharide from Myco. tuberculosis a 0.1 97 91 89 91 98 89 90 Control c 0 49 22 12 15 11 8 Acacia c 0.04 59 31 37 27 25 25 26 a 2.5 ml tuberculin PPD-"4C solution (50 lU/ml; phosphate-buffered saline, pH 7.38 (Magnusson et al., 1958); 0.3% phenol) was dispensed In a 10-ml glass ampoule and stored at 50C. b Molecular weight of dextran: 60 000-0 000. c 2.5 ml tuberculln PPD-'4C solution (50 lU/mI; borate-buffered saline, pH 7.2; 0.5% phenol (Gottschall & Bunney, 1938)) was dispensed In a 10-mi glass ampoule and stored at 5°C. Substances of low molecular weight. Glycerol had practically no anti-adsorption effect in concentration up to 12% by weight (Table 14). Glycine (1 %), Tris (0.1 %), Evans blue (0.001 %), and preservatives such as phenol (2.0%, 1 %, and 0.3%Y.) or chinosol (0.01 Y.) also gave no protection against adsorption of tuberculoprotein to glass (Table 14). CONCLUSIONS The pH of the diluent influences the rate of adsorp- tion of tuberculoprotein to glass (being twice as much at pH 4 as at pH 7.38). This influence is of little practical interest since no commercial tuberculin solutions are prepared at pH values other than 7-7.5. At pH 7.38 the loss of tuberculoprotein due to adsorption (10-ml ampoules; 50 IU/ml) is approxim- ately 90%. The residual radioactivity is presumably due to a small fraction of non-adsorbable substances which may be present in the tuberculin PPD prepara- tion. At concentrations of 500 IU/ml or more, the percentage loss of tuberculoprotein in solution due to adsorption becomes negligible. However, since the tuberculin concentration used in epidemiological surveys as well as in the individual patient is usually 50 IU/ml or less, the percentage loss of tuberculin PPD will be appreciable in unstabilized preparations, thus giving unreliable skin test reactions. Tuberculin skin tests can be meaningful only if the tuberculin PPD concentration is known to be constant. There is no doubt, therefore, that an anti-adsorption agent should be added to a dilute tuberculin solution if the declared potency of the tuberculin solution is to be maintained during storage until its " shelf-life" expires. 103 S. LANDI, H. R. HIELD & M. C. TSENG TABLE 12 EFFECT OF-OLD TUBERCULIN (OT) ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-14C a TO GLASS EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Radioactivity in solution (%) Borate-buffered saline (no acacia) Borate-buffered saline (0.04% acacia)Storage time .- PPD-'_4C PPD-"C PPD-1C (50 lU/mi) PPD- 4C PPD-14C PPD-'4C (50 lU/mi)(50IUP mI) (100 lU/ml) with OT (50 lU/mi, (50 lU/mI) (100 lU/mi) w1:2 000 dii.) 1 :2000 dil.) 2______di 8 hours 58 73 67 56 73 77 1 day 42 56 51 55 64 58 7 days 28 50 39 44 65 61 14 days 22 57 45 41 61 49 I month 19 36 32 38 57 46 3 months 20 51 39 40 66 50 6 months 16 48 28 39 59 44 9 months 17 47 31 43 60 43 12 months 22 50 38 39 60 42 18 months 20 42 28 39 65 39 24 months 19 50 31 38 61 43 Average; 1-24 months 19.0 46.3 32.4 39.4 61.1 43.9 a 2.5 ml solution (borate-buffered saline with and without acacia, pH 7.2; 0.3% phenol) was dispensed in a 10-mi glass ampoule and stored at 50C. TABLE 13 EFFECT OF SYNTHETIC POLYMERS ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-14C TO GLASS EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Substance tested Radioactivity in solution (%) Molecular Concen- After After After After After After After Type weight tration 1 day 1 week 1 month 6 months 12 months 18 months a 24 monthsa(overage) (%) Control b - 0 49 22 21 16 15 15 14 PVP b,c 360 000 0.1 87 80 84 78 77 77 76 pVp b, d 34 000 0.1 51 37 32 33 16 18 20 Polyethylene glycol 17 500 0.1 92 90 86 87 84 88 80 Polyethylene glycol 3 350 0.1 59 34 31 37 32 ND ND Polyethylene glycol 400 0.1 44 22 10 8 7 ND ND Polyethylene oxide 4 000 000 0.02 67 67 41 40 37 ND ND a ND = Not determined. b2.5 ml tuberculin PPD-'4C solution (50 lU/mi; phosphate-buffered saline, pH 7.38; 0.3% phenol) was dispensed in a 10-mi glass ampoule and stored at 50C. c Polyvinylpyrrolidone K90. d PolyvinyIpyrrolidone K26-28. 104 EVALUATION OF SUBSTANCES TO PREVENT ADSORPTION OF PPD TO GLASS TABLE 14 EFFECT OF VARIOUS SUBSTANCES OF LOW MOLECULAR WEIGHT ON THE RATE OF ADSORPTION OF TUBERCULIN PPD-'4C TO GLASS EXPRESSED IN PERCENTAGE RADIOACTIVITY IN SOLUTION Substance tested Radioactivity In solution (%) Type Molecular Concen- After After After After After After | Afterype_________ - weight tration I day I week I month 6 months 112 months 18 months 24 months a Controlb | 0 49 22 21 16 15 15 14 Glycerol, 92.09 12.0 70 37 50 25 19 ND ND Glycineb 75.07 1.0 50 33 18 21 22 18 19 Evans blueb 0.001 49 17 14 11 9 ND ND Tris b | 121.14 0.1 46 12 7 7 6 ND ND Control c - 0 47 18 10 7 11 11 10 Chinosol c 388.39 0.01 40 34 19 24 15 12 10 Phenol 94.11 0.3 46 20 19 20 9 10 7 Phenol 94.11 1.0 51 30 15 15 9 ND ND Phenol 94.11 2.0 66 38 24 11 10 ND ND a ND = Not determined. b 2.5 ml tuberculin PPD-'4C solution (50 lU/ml; phosphate-buffered saline, pH 7.38; 0.3% phenol) was dispensed In a 10-mlglass ampoule and stored at 50C. c Same as footnote b, except that no phenol was added. Tween 80 (50 ppm and 5 ppm, or 0.005% and 0.0005% respectively), which is widely used in tuber- culin dilutions, gives protection against absorption of tuberculoprotein to glass. In addition to Tween 80, each one of the following substances can be con- sidered as an efficient anti-adsorption agent for tuberculin PPD-namely, nonionic surfactants: Tween 20 (0.0005°/), Tween 40 (0.0005 %), Tween 60 (0.0005%Y.), Triton WR 1339 (0.0005%), saponin (0.01 %), Brij 35 (0.0005%"O), Triton X100 (0.005%/O), Corexit (0.005%/); ionic surfactants: sodium lauryl- sulfate (0.002%), sodium deoxycholate (0.1 %), benzalkorium chloride (0.05 %), cetrimonium bro- mide (0.0001 %), cetylpyridinium chloride (0.05 %); colloidal substances: gelatins (0.1 %), human serum albumin (0.1 %), PVP (mol. wt 360 000) (0.1 Y), polyethylene glycol (mol. wt 17 500) (0.1 %); Jaguar LV (0.1 %) and a polysaccharide obtained from Myco. tuberculosis (0.1 %). However, before considering substituting any one of these substances for Tween 80 they will have to be thoroughly investigated as to their effects on the tuberculin skin reaction in man. ACKNOWLEDGEMENTS We wish to thank Mrs S. Ober and Mrs S. L. Jailos for their valuable technical assistance throughout this work. 105 106 S. LANDI, H. R. HELD & M. C. TSENG RItSUMIt ESSAI DE DIVERSES SUBSTANCES SUSCEPTIBLES D'EMPECHER L'ADSORPTION DU DtRIVt PROTEINIQUE PURIFIE (PPD) SUR LES SURFACES DE VERRE On sait que les solutions dilu6es de tuberculine (1 ou 5 UI par dose) perdent tres rapidement de leur activit6 par suite de l'adsorption de la tuberculoproteine sur la paroi des ampoules. On a utilise une solution de PPD (50 UI/ml) marquee au 14(C pour mesurer l'intensite du ph6nomene dans diverses conditions experimentales et dvaluer l'effet anti-adsorption d'une sdrie d'adjuvants. La quantit6 de PPD (en solution dans du solut6 salin tamponne au phosphate) adsorbee par cm' de surface est maximale (0,31 jig/cm2) a pH 4, dans la zone du point iso6lectrique du PPD, et minimale (0,15 ,ug/cm') entre pH 6 et pH 10. Le taux d'adsorption n'est pas modifie lorsqu'on remplace le phosphate par du borate. De meme, la methode utilisde pour preparer le PPD (precipitation par le sulfate d'ammonium, precipitation par l'acide trichloracetique, ou methode mixte) n'a aucune influence sur l'intensite du phenomene. A pH 7,38, la perte par adsorption de 2,5 ml d'une solution de PPD, a 50 UI/ml, contenus dans une ampoule de 10 ml est d'environ 50% apres un jour et d'environ 90% apres un mois, ce qui montre a suffisance la n6cessit6 d'employer un adjuvant anti-adsorption. Le Tween 80, ai la concentration de 0,005% ou de 0,0005 %, pr6vient efficacement l'adsorption, mais on I'a accuse de modifier l'aspect et la taille des reactions tuberculiniques. Ce produit, et 42 autres substances, ont fait l'objet d'essais d'efficacit6. Parmi les agents qui assurent une pr6vention de bonne qualite figurent des surfactifs non ioniques, certains surfactifs ioniques et certaines substances colloidales. Par contre, les polypep- tides et les substances non tensio-actives de faible poids mol6culaire n'ont qu'un effet anti-adsorption faible ou nul. Avant d'envisager de substituer au Tween 80 l'une ou l'autre des substances actives, il est toutefois indispen- sable de procdder Liune etude complete de leurs effets sur les caracteres de la reaction tuberculinique. REFERENCES Albert, A. (1965) Selective toxicity, London, Methuen, p. 322 Asami, N. & Kataoka, T. (1966) Kekkaku, 41, 7-11 Augier. J., Lajudie, J. & Augier-Gibory, S. (1967) Ann. Inst. Pasteur, 112, 358-368 Bingle, J. P. & Czerkawaski, J. W. (1964) Biochem. J., 91, 400-408 Bleiker, M. A. & Griep, W. A (1965) Bull. Wld Hlth Org., 33, 375-383 Bull, H. B. (1956) Biochim. Biophys. Acta. (Amst.), 19, 464-471 Bull, H. B. (1957) Arch. Biochem., 68, 102-111 Gottschall, R. & Bunney, W. E. (1938) J. Immunol., 34, 103-115 Green, H. H. (1946) Brit. vet. J., 102, 267-278 Guld, J. & Roelsgaard, E. (1965) Bull. Wld Hlth Org., 33, 345-363 Hummel, J. P. & Anderson, B. S. (1965) Arch. Biochem., 112, 443-447 Jensen, K. A., Bindslev, G., Moller, S., Hansen, A. & Lind, P. (1938) Tubercle (Lond.), 19, 388-397, 433-447 Landi, S. (1963) Appl. Microbiol., 11, 408-412 Landi, S., Held, H. R., Hauschild, A. H. W. & Hilsheimer, R. (1966) Bull. Wld Hlth Org., 35, 593-602 Landi, S., Held, H. R., Hauschild, A. H. W. & Hilshei- mer, R. (1967) Appl. Microbiol., 15, 839-843 Landi, S., Held, H. R. & Pivnick, H. (1968) Bull. Wld Hlth Org., 39, 809-820 Long, D. A., Miles, A. A. & Perry, W. L. M. (1954) Bull. Wld Hith Org., 10, 989-1002 Long, E. R. & Seibert, F. B. (1926) Amer. Rev. Tuberc., 13, 393-397 Magnusson, M. & Bentzon, M. W. (1958) Bull. WldHith Org., 19, 829-843 Magnusson, M., Guld, J., Magnus, K. & Waaler, H. (1958) Bull. Wld. Hlth. Org., 19, 799-828 Manegold, E. (1953) Schaum, Heidelberg, Strassenbau, Chemie und Technik, G.m.b.H., p. 189 Marks, J. (1964) Tubercie (Lond.), 45, 62-63 Merck Index, 1968, 8th ed., Rahway, N. J., Merck & Co., p. 919 Neurath, H. & Bull, H. B. (1938) Chem. Rev., 23, 391-435 Parish, H. J. & O'Brien, R. A. (1935) Brit. med. J., 1, 1018 Seibert, F. B. (1949) Amer. Rev. Tuberc., 59, 86-101 Seibert, F. B. & Dufour, E. H. (1954) Amer. Rev. Tuberc., 69, 585-594 Seibert, F. B. & Glenn, J. T. (1941) Amer. Rev. Tuberc., 44, 9-25 Seibert, F. B., Pedersen, K. 0. & Tiselius, A. (1938) J. exp. Med., 68, 413-438 Sokal, R. R. & Rohlf, F. J. (1969) Biometry, San Fran- cisco, W. H. Freeman & Co., p. 62 Toman, K., Hejdovd, E., Polknsk9, F., Wterbovd, E. & Guld, J. (1968) Bull. Wid. Hlth. Org., 39, 801-808 Toman, K., Poldnsk9, F., Hejdovd, E. & WtFrbovd, E. (1965) Bull. Wld Hlth Org., 33, 365-373