Substantia. An International Journal of the History of Chemistry 7(1): 67-78, 2023 Firenze University Press www.fupress.com/substantia ISSN 2532-3997 (online) | DOI: 10.36253/Substantia-2040 Citation: Adams M. H. (2023) Surface Inactivation of Bacterial Viruses and of Proteins. Substantia 7(1): 67-78. doi: 10.36253/Substantia-2040 Copyright: reproduced from Surface Inactivation of Bacterial Viruses and of Proteins, published by Mark H. Adams on Journal of General Physiol- ogy, 5(31), 417-431, 1948. Copyright © 2023, by Mark H. Adams. Reproduced with permission of the publisher. Data Availability Statement: All rel- evant data are within the paper and its Supporting Information files. Competing Interests: The Author(s) declare(s) no conflict of interest. Feature Article Surface Inactivation of Bacterial Viruses and of Proteins Mark H. Adams Department of Bacteriology, New York University College of Medicine, New York Due to the Covid-19 pandemic an enourmous number of papers have appeared in the literature. Here we republish, with permission, the paper written by M.H. Adams in 1948 for the remarkable contribution in Physiology and in other fields. A short introduction by Barry W. Ninham precedes the paper. MISSED OPPORTUNITIES AND FASHIONABLE PURSUITS by Barry W. Ninham Department of Materials Physics, Australian National University, Canberra, Australia The extraordinary paper of Mark H. Adams: Surface Inactivation of Bac- terial Viruses and of Proteins appeared in a mainstream Journal of Physiol- ogy, in 1948.1 It was neglected and has been ever since. In retrospect this was and is a tragedy for science of the very first order.2 Adams was recognised in his time as a brilliant bacteriologist who, sadly, died young (1912-1956). His peers completed a partially finished book on his work in 1959.3 It was forgotten in the rush to join the DNA biological revolution. In that revolution the physical sciences, the physical chemistry of solutions, col- loid and surface science have played almost no serious conceptual role at all. The biological/medical and the physico-chemical sciences have diverged almost completely, to a point where their languages are mutually incompre- hensible.2 Of course, characterisation and diagnostics of disease by myriad new techniques has been essential to progress in medicine and biology. But the gap remains. We are missing something. 1 Adams, M. H. SURFACE INACTIVATION OF BACTERIAL VIRUSES AND OF PROTEINS. Journal of General Physiology 1948, 31 (5), 417–431. https://doi.org/10.1085/jgp.31.5.417. 2 Ninham, B. W. The Biological/Physical Sciences Divide, and the Age of Unreason. Substantia 2017, 1 (1), 7–24. https://doi.org/10.13128/Substantia-6. 3 Blair, J. E. Bacteriophages. Mark H. Adams, with Chapters by E. S. Anderson, J. S. Gots, F. Jacob and E. L. Wollman. Interscience Publishers, Inc., New York, 1959. Clinical Chemistry 1959, 5 (6), 634. https://doi.org/10.1093/clinchem/5.6.634. http://www.fupress.com/substantia 68 Mark H. Adams Adams paper circumscribes that something. To see why we remark that 1948 was the same year that Overbeek ’s landmark thesis on colloid sta- bility marked the basis for the DLVO theory of colloid stability.4,5 It introduced long ranged quantum mechani- cal, dispersion forces of interaction between particles and dominated thinking about forces until now. Its limi- tations were spelt out by Derjaguin and Overbeek. Many of these limitations appeared to be resolved by sophisti- cated further extensions that embraced many body forc- es via Lifshitz theory, on charge regulation, on effects due to solvent structure and molecular size. Direct force measurements between surfaces that confirmed theory, a challenge dating back before Newton, appeared to repre- sent a triumph.6 But there was and remained an uneasy juxtaposi- tion with the classical theories of the physical chemistry of electrolytes and electrochemistry, and colloid science. These theories, of pH, of activities, of pKas, of conduc- tivity, the electrical double layer, of zeta potentials that were developed before quantum mechanics, and ignored it. Dispersion forces between ions and ions and surfaces are the key to specific ion (Hofmeister) effects essential to biology. And hydration was unquantified. Add to that the fact that undefined anthropomorphic words like hydrophilic and hydrophobic figured prominently in the conversation and we have an unquantifiable mess. That is explicit in that standard measurements like pH and zeta potentials were based on inadequate theory. It got worse when it was realised that even the apparently impressive addition of Electromagnetic Quantum Field theory embodied in Lifshitz theory was 4 Verwey, E. J. W. Theory of the Stability of Lyophobic Colloids. J. Phys. Chem. 1947, 51 (3), 631–636. https://doi.org/10.1021/j150453a001. 5 Derjaguin, B. V., & Landau, L. (1941). Acta Physicochim. URSS. Jour- nal of Experimental and Theoretical Physics, 14, 635-649. Derjaguin, B. V., & Landau, L. (1941). Theory of stability of highly charged lyophobic soils and adhesion of highly charged particles in solutions of electrolytes. Zhunal Eksperimentalnoi; theoretischekoi Fisiki, 11, 801-818. Derjaguin and Landau publications in Russian were not easily accessible in the west during the 2nd world war. Landau was appallingly scathing of Sam Levine who nearly had the theory first. He did so because Sam replaced a nonlinear coupling constant integration in calculating the double layer free energy by a linear one. It is therefore ironic that Lan- dau’s students Dzyaloshinski, Lifshitz and Pitaevski made the same mis- take in their tour de force of Quantum Electrodynamic forces. See also: Ninham, B. W.; Brevik, I.; Boström, M. Equivalence of Electromagnetic Fluctuation and Nuclear (Yukawa) Forces: The Π₀ Meson, Its Mass and Lifetime. Substantia 2022. Just Accepted https://doi.org/10.36253/Sub- stantia-1807. 6 Ninham, B. W.; Lo Nostro, P. Molecular Forces and Self Assembly: In Colloid, Nano Sciences and Biology; Cambridge Molecular Science; Cambridge University Press: Cambridge, 2010. https://doi.org/10.1017/ CBO9780511811531. flawed too, and that the ansatz of additivity of electro- static and electrodynamic fluctuation (dispersion) forces, the one treated in nonlinear theory, the other linear, vio- lates the laws of physics.7,8 Since hardly anyone understood mathematics or physics, it did not matter. Matters reached a nadir in the 1980s when the leading speaker at a Nobel symposium, from a Swiss drug company began his lecture with slide: NMR. No, not nuclear magnetic resonance he said. It means NO MORE RESEARCH! By which, he explained, the game was over. Mod- ern computer simulation with 70,000 effective molecular interaction parameters without water, could design any required drug or protein or enzyme with ease. A laconic biochemist asked; What happens if you raise the temperature above body temperature of 37 degrees. Ha! Easy. We change the parameters! Such mind-blowing idiocy continues and marks the end of an era. Faraday would be turning in his grave. We can forgive the medical and biology people for ignoring what ought to be the enabling discipline of physical chemistry on the reasonable grounds that they have more than enough to do than be confused by the subtleties of the physical sciences; especially if even the quantum electrodynamicists produce a flawed theory. But there is much more, and here we come to Adams. The physical theories (and simulation) ignore any effects due to dissolved atmospheric gases. The ancient Greeks tell us that there were 4 elements. These are: fire (temperature), water, earth, and air. Take away air and hydrophobic interactions go away; emulsions become stable, chemical interactions cease, enzymes working. The simplest of experiments is bubble-bubble interac- tions in salt water. Bubbles do not fuse above a concen- tration of 0.17 molar for 1:1 salts like NaCl.9,10 Not coin- cidently, that corresponds exactly to the ionic strength of 7 Ninham, B. W. B. V. Derjaguin and J. Theo. G. Overbeek. Their Times, and Ours. Substantia 2019, 3 (2), 65–72. https://doi.org/10.13128/Sub- stantia-637. 8 Lo Nostro, P.; Ninham, B. W. After DLVO: Hans Lyklema and the Keepers of the Faith. Advances in Colloid and Interface Science 2020, 276, 102082. https://doi.org/10.1016/j.cis.2019.102082. 9 Craig, V. S. J.; Ninham, B. W.; Pashley, R. M. The Effect of Electrolytes on Bubble Coalescence in Water. J. Phys. Chem. 1993, 97 (39), 10192– 10197. https://doi.org/10.1021/j100141a047. 10 Bunkin, N. F.; Ninham, B. W.; Ignatiev, P. S.; Kozlov, V. A.; Shkirin, A. V.; Starosvetskij, A. V. Long-Living Nanobubbles of Dissolved Gas in Aqueous Solutions of Salts and Erythrocyte Suspensions. Journal of Bio- photonics 2011, 4 (3), 150–164. https://doi.org/10.1002/jbio.201000093. 69Surface Inactivation of Bacterial Viruses and of Proteins salts in the blood. The phenomenon occurs for a whole range of salts. But for other ion pairs, no such effect occurs. There are strict rules that govern which ion pairs “work ” and which do not. The effects occur at higher concentrations for sugars. They occur at much lower concentrations for amino acids.11 Classical physical and colloid chemistry are impo- tent in the face of such a challenge. The answer seems to lie in the fact that in any solid or liquid impurities aggregate, in much the same way that surfactants aggregate to form micelles. There is a natural critical salt concentration for sta- ble nanobubble formation. This shows up in conductiv- ity measurements in salt solutions with and without dis- solved gas.12 Such nanobubbles inhibit bubble-bubble fusion by known depletion forces. Nanobubbles that form spontaneously in enzyme- substrate interactions produce free radicals that drive catalysis. Cavitation produced by ship propellers, a very major economic transport problem, disappears if a jet of air deficient water is projected onto the propeller.13 The tensile strength of water is two orders of magni- tude higher when gas is removed, a fact known for more than a century, and explained immediately by Griffith’s theory of. the strength of solids if gas aggregates of nanobubbles impurities occur. Much more to the point stable nanobubbles of CO2 produced by metabolism produce a foam that forms the endothelial surface layer lining all cells and tissues. It is extremely effective in killing viruses like covid and other pathogens.13,14,15 Nanobubbles of oxygen/nitrogen are delivered by the lungs to capillaries, not molecularly and catalyse many reactions.16 11 Nafi, A. W.; Taseidifar, M.; Pashley, R. M.; Ninham, B. W. The Effect of Amino Acids on Bubble Coalescence in Aqueous Solution. Journal of Molecular Liquids 2023, 369, 120963. https://doi.org/10.1016/j.mol- liq.2022.120963. 12 Ninham, B. W.; Lo Nostro, P. Unexpected Properties of Degassed Solutions. J. Phys. Chem. B 2020, 124 (36), 7872–7878. https://doi. org/10.1021/acs.jpcb.0c05001. 13 Vol. 4 No. 2 Suppl. 1 (2020) – About Water: Novel Technologies for the New Millennium | Substantia. B.W Ninham and R.M Pashley eds. 14 Garrido Sanchis, A.; Pashley, R.; Ninham, B. Virus and Bacteria Inac- tivation by CO2 Bubbles in Solution. npj Clean Water 2019, 2 (1), 1–9. https://doi.org/10.1038/s41545-018-0027-5. 15 Reines, B. P.; Ninham, B. W. Structure and Function of the Endothe- lial Surface Layer: Unraveling the Nanoarchitecture of Biological Surfaces. Quarterly Reviews of Biophysics 2019, 52, e13. https://doi. org/10.1017/S0033583519000118. 16 Ninham, B.; Reines, B.; Battye, M.; Thomas, P. Pulmonary Surfactant and COVID-19: A New Synthesis. QRB Discovery 2022, 3, e6. https:// doi.org/10.1017/qrd.2022.1. The present state of affairs as it exists in Colloid Sci- ence can be seen in Ref. 17.17 It is seems clear that dissolved gas is a hidden vari- able that is essential to understanding the properties of condensed matter. The effects are dramatic. The Greeks told us so. Our theories have ignored dissolved gas. They compare gas-free models with real world soft matter that does contain gas. A simple example is the deviation from theory of the measured Debye length in a 1:1 elec- trolyte above the critical concentration for nanobubble formation. It is impossible to simulate these effects. Adams paper on effects on proteins of shaking test tubes containing physiological saline, or not, containing gas or not covered all this 75 years ago. A pity we missed it. There is nothing we can do about it except wonder at what might have been had he lived. There are appearing now a number of good papers that are following Adams lead.18 And finally we can note that the extinction of spe- cies after various ice ages makes sense if we allow dis- solved gas to do its job. During an ice age salt is precipi- tated out. After the ice age the melt water is salt depleted below the critical concentration for bubble-bubble (and nanobubble) fusion. All single celled creatures would die of the bends. Some few souls are beginning to shake things up and surprised at what they see. Barry W. Ninham 17 Ninham, B. W.; Pashley, R. M.; Nostro, P. L. Surface Forces: Changing Concepts and Complexity with Dissolved Gas, Bubbles, Salt and Heat. Current Opinion in Colloid & Interface Science 2017, 27, 25–32. https:// doi.org/10.1016/j.cocis.2016.09.003. 18 Bunkin, N. F.; Shkirin, A. V.; Ninham, B. W.; Chirikov, S. N.; Chai- kov, L. L.; Penkov, N. V.; Kozlov, V. A.; Gudkov, S. V. Shaking-Induced Aggregation and Flotation in Immunoglobulin Dispersions: Differences between Water and Water–Ethanol Mixtures. ACS Omega 2020, 5 (24), 14689–14701. https://doi.org/10.1021/acsomega.0c01444. 70 Mark H. Adams It has been noticed previously that certain viruses can be rapidly inactivated by shaking or by bubbling gases through the virus suspensions. Campbell-Rent- on (1) studied the effect of violent mechanical shaking on bacteriophages and found them to be fairly rapidly inactivated, at rates which were charac teristic for each phage. Grubb, Miesse, and Puetzer (2), while studying the effect of various vapors on influenza A virus, noted that bubbling air at the rate of 1 liter a minute through the virus suspension resulted in detectable reduction in infectivity in 10 minutes. In a somewhat more extensive study McLimans (3) found that both Eastern and West- ern strains of equine encephalitis virus were rapidly inactivated by shaking in buffered saline suspensions. The inac tivation also occurred when gases were bub- bled through suspensions of the virus. The rate of inac- tivation was the same whether oxygen or helium was the gas used, indicating that the inactivation was prob- ably a physical process, rather than the result of chemi- cal interaction between virus and gas. He also noted that the rate of inactivation increased markedly as the pH was reduced from 7 to 5, though control suspensions at rest suffered no inactivation. The inactivation of certain physiologically active proteins such as enzymes (4) and toxins (5) on shaking is a familiar phenomenon. Perhaps not quite so well- known is the fact that this kind of inactivation can be specifically pre vented by the presence in the diluent of very small amounts of proteins. It has been demonstrat- ed that the spreading of a protein at a gas-liquid inter- face results in the denaturation of the protein, since the spread protein becomes completely insoluble in water (6) Presumably the role of the shaking or bub bling in the inactivation of viruses and physiologically active pro- teins is simply that of enormously increasing the area of the gas-liquid interface, and hence increasing the chances of the susceptible protein being spread on that surface. This paper is devoted to the kinetics of the inac- tivation of bacteriophage by shaking and to the effect of environmental influences on the rate of inactivation. MATERIALS AND METHODS The group of seven coli-dysentery phages studied by Demerec and Fano (7) was used. The properties of this group of bacterial viruses have been summarized by De1bl-tick (8). These phages were grown on Escheri- chia coli, strain )3, in a chemically defined medium of the following composition: NH4C1 ........................................................................... 1.0 gm. MgSO4 ........................................................................... 0.1 gm. KH2PO4 3.5 gm. Lactic acid .................................................................... 9.0 gm. NaOH ............. about 4.0 gm. or to a final pH of about 6.5 H2O ........................................................................... 1,000 ml. Since T5 is not produced in the absence of calcium ion, calcium choride to a concen tration of 0.001 M was added when preparing stocks of this phage. All phage stocks used contained about 1010 plaque forming parti- cles per ml. All phage assays were made on strain B of E. coli using the agar layer technique of Gratia as modified by Hershey (9). The saline buffer diluent used in the inactivation experiments contained 0.15 M NaCI, 0.001 M MgSO4, 0.01 M buffer, and other additions as noted. Most exper- iments were performed using phosphate buffer at pH 6.5. Inorganic chemicals were reagent grade; the gelatin was Eastman ash-free calfskin gelatin; the bovine serum albumin was Armour’s fraction V; yeast nucleic acid was a purified specimen from Eimer and Amend; the thy- mus nucleic acid was a highly viscous Hammarsten type preparation. In the bubbling experiments nitrogen was passed through a coarse grade Corning sintered glass filter at the rate of 1 liter per minute producing a vigorous effer- vescence in the virus suspension held in the filter. The gas was saturated with water vapor and the gas stream as well as the filter and its contents was in a constant temperature bath. The shaking experiments were carried out in test tubes 15 mm. × 100 mm. with a capacity of 16 cc. These tubes as well as dilution tubes were cleaned with hot acid dichromate, well rinsed, and twice boiled with dis- tilled water. Pipettes were simi larly acid-cleaned and rinsed with hot distilled water The test tubes were closed with rubber stoppers which were boiled with sodium hydroxide, well rinsed, then twice boiled with distilled water before each use. The most meticulous cleanliness was essential in obtaining consistent results. The shaking machine had a horizontal reciprocating motion of 320 cycles per minute and the carriage traversed a distance of 7 cm. The test tubes were shaken with the long axis parallel with the direction of motion of the carriage. EXPERIMENTAL Kinetics of the Inactivation Reaction. — Bacterio- phage T7 at an initial concen tration of 6 × 109 plaque- forming particles per ml. was diluted in the saline-buffer diluent to a concentration of about 104/cc. The condi- tions of the experi ment were: phosphate buffer of pH 6.5, temperature 26°C., volume of phage suspension 5 71Surface Inactivation of Bacterial Viruses and of Proteins cc., shaker stopped every 2 minutes for sampling. The log per cent survivors proved to be a linear function of time indicating that the rate of phage destruction was proportional to the concen- tration of surviving phage; or K=1/t lnP0/Pt The data of this sample experiment are given in Table I. The first order velocity constant for the inacti- vation of T7 under the stated conditions was 0.28 min-1. There was no recovery of activity on standing in buffer diluent or broth, and inactivation occurred at a signifi- cant rate only during the periods of shaking (Table I). The velocity constants for the inactivation of each of the seven coil phages and of two of their mutants at pH 6.5 and 26°C. are given in Table II. From the data in Table II it may be noted that the small phages T1, T2, and T7 are inactivated more rap- idly than the larger phages. Phage T4r+ and its rapid lys- ing mutant (10) T4r are much more stable than the other phages. With both T2 and T4 phages there was no signifi- cant difference between the stabilities of wild type and rapid lysing mutant. Also in mixtures of wild type and mutant forms, the proportion of the two types remained constant during the inactivation. The volume of phage suspension was varied from 4 cc. to 7 cc. per 16 cc. tube without affecting the veloc- ity of the inactivation. However, if the tube is filled with virus suspension so that no air space is left, there is no perceptible inactivation of the phage during 40 min- utes’ shaking, even when half a dozen glass beads are added to the tube. Because of the possibility of inactiva- tion of phage through adsorption to the glass walls of the tube or to the rubber stopper, both of these surfaces were coated with melted paraffin. In the paraffin-coated tube the rate of inactivation of phage was the same as in un coated tubes even though the paraffin surfaces were not wetted by the suspen sion of phage. If loss of activity were due to adsorption, the virus must adsorb equally well to glass and to paraffin. From these experiments it would appear that the shaking or agitation of the fluid suspension in contact with glass sur faces is not the cause of the inactivation of virus, but rather that the inactiva- tion occurs at the gas-liquid interface which is present in enormous area when tubes half filled with liquid are vio- lently shaken. The variation of velocity constant of inactivation as a function of pH is shown in the curves of Fig. 1. From these curves it is evident that the rate of inactivation by shaking is minimal between pH 5 and 8 but increases rapidly outside this range. The small phage T7 is much more rapidly inactivated at all pH values than are the larger phages, and repeated assays at a given pH are less reproducible with the small Table I. The Inactivation of TT Bacteriophage by Shaking in Saline-Buffer Diluent at 26°C. and pH 6.5 Time Sample Count Survivors per 0.1 ml. Pe/Pt In Po/Pg K 0 0 ml. 0.02 0.02 136 165 Av. 753 2 0.05 225 450 1.67 0.51 0.26 4 0.05 104 208 3.6 1.28 0.32 6 0.05 68 136 5.5 1.7 0.28 8 0.05 45 90 8.3 2.1 0.27 10 0.1 52 52 14.4 2.7 0.27 15 0.1 13 13 58 4.1 0.27 35 0.1 1 1 753 6.6 (0.19) Av. 0.28 Table II . The Average Velocity Constants for the Shaking Inactiva- tion of coli Phages at 26°C. and pH 6.5. Phage Velocity constant T1 0.59 min.-1 T2r+ 0.24 T2r 0.23 T3 1.2 T4r+ 0.05 T4r 0.07 T5 0.24 T6 0.20 T7 0.48 72 Mark H. Adams phages, resulting in a more erratic looking curve. All phages were markedly unstable in the absence of shak- ing at pH 3 except T4. At pH 4 the phages were more stable than at pH 3 but in a few cases there was a 10 to 50 per cent loss in activity on standing at room temper- ature for 1 hour. At the higher pH values studied there was no detectable loss in activity in un shaken controls during the course of the experiments. Phage T7 which is very rapidly inactivated by shaking at pH values above 7 is not detectably inacti vated in unshaken control tubes after 1 hour at pH 8.7. The effect of temperature on the velocity constants of inactivation was determined for phages T1 and T7. The averages for a number of determinations at 0°, 25°, and 38°C. are given in Table III. From these values, using the Arrhenius equation, the Arrhenius constant for the shaking inactivation of phages T1 and T7 appears to be about 10,000 cal./mol. This value is high- er than the reported values for the heat of activa tion of denaturation of proteins by urea and by shaking (11) and far less than the values for heat denaturation of proteins. In the presence of 1 mg./ml. of gelatin all phages were stable on shaking for 1 hour at room temperature at pH 6.5. Therefore, the protective effect of various con- centrations of gelatin on phage T5 was determined. The results are summarized in Fig. 2. It will be seen from Fig. 2 that as little as 0.01 γ per ml. of added gelatin has a definite protective effect on phage T5 while 1 γ/ml. gave complete protection for 14 minutes. However, after 20 minutes of shaking with 1 γ/ml. of gelatin the phage activity began to decrease. It would appear that the duration of the protective effect of gelatin is a function of the concentration of gelatin, and that the gelatin also appears to be inactivated by shak- ing. If the survival time of the gelatin is taken as the time when the inactivation curve becomes parallel to the inactivation curve in the absence of gelatin, then it becomes possible to estimate the rate of disappearance of the gelatin. The disappearance of the gelatin under these conditions appears to follow the kinetics of a first order reaction with a half-life of about 2 minutes. This relationship does not hold for concentrations of gelatin Figure 1. Velocity constants, K minute-1, as a function of pH for the phages T2, T4, T5, and T7 at about 26°C. Table III. The Average Velocity Constants for Inactivation of T1 and T7 at 0°, 25°, and 38°C. Temperature Velocity constants for T1 T7 cc. 0 0.31 0.09 25 0.59 0.48 38 0.97 0.70 Figure 2. The inactivation of phage T6, shaken in the presence of various amounts of gelatin. 73Surface Inactivation of Bacterial Viruses and of Proteins above about 0.5 γ/ml. since as the protein concen tration becomes higher the kinetics change from those of a first order reaction to those of a zero-order reaction in which the rate of inactivation is determined by the available surface rather than by the concentration of protein in solution (11). If this supposition is correct, preshaking of the gela- tin solutions before adding the phage should destroy the protective effects of the gelatin. The experiment illustrated in Fig. 3 is identical with the previous experiment except that the dilutions of gelatin in saline-buffer diluent were preshaken for 15 minutes before addition of phage. Then after phage addition the tubes were shaken and samples withdrawn at intervals for assay. It may be seen from Fig. 3 that the protective effects of all quantities of gelatin through 0.33 γ/ml. are destroyed by shaking for 15 min- utes so that the resultant inactivation curves are identi- cal with the curve with no added protein. There is little appreciable diminution in the protective effect of 1 γ/ml. of gelatin after 15 minutes of shaking. Similar experiments have been carried out using various concentrations of gelatin with T7 and T2r, with very similar results. Since gum arabic is a colloidal substance with reput- ed protective effect against inactivation of tuberculin (12) it was tested for its effect on the shaking inactiva- tion of phage T5. As may be seen from Fig. 4 gum arabic gives a family of curves similar to those given with gela- tin, except that about 100 times as much gum arabic is required to equal the effect of gelatin. It is probable that the protecting effect of the gum arabic is due to con- tamination with about 1 per cent of protein. This agrees with previously made quantitative estimates of the pro- tective effect of gum arabic against the surface inactiva- tion of tyrosinase (13). In a similar manner yeast nucleic acid and thymus nucleic acid prepared according to Hammarsten were tested for possible protective effect. Both of these sub- stances had a protective effect equivalent to about 1 per cent of their weight of gelatin. Since no amino acid anal- Figure 3. The inactivation of phage T5, shaken in the presence of various amounts of gelatin which had already been shaken for 15 minutes before addition of phage. The symbols correspond to the same gelatin concentrations as in Fig. 2. The solid line is the curve for no added gelatin and the dotted line is an average curve for the tubes containing added gelatin. Figure. 4. The inactivation of phage T5, shaken in the presence of various amounts of gum arabic. 74 Mark H. Adams yses were available for these samples of nucleic acid we cannot say whether the protective effect is due to con- tamination with protein or is an inherent property of nucleic acids. A commercial sample of bovine serum albumin (Armour fraction V) prepared by alcohol fractionation was tested for its protective effect with the results shown in Fig. 5. By a comparison of the curves of Fig. 5 with the curves with gelatin in Fig. 2 it may be seen that serum albumin is about one-tenth as active in protecting the virus from inactivation as is gelatin. This observation is in accord with experi ments of Berger, Slein, Colow- ick, and Cori (14) on the inactivation of hexo kinase in which the protective effect of serum albumin was about one-tenth that of insulin or rabbit muscle protein. It also agrees qualitatively with reported effects on the stability of diphtheria toxin diluted for the Schick test. Edsall and Wyman (15) reported that 500 γ/ml. of human serum albumin gave in complete protection while 1 mg./ml. gave excellent protection. Moloney and Taylor (16) using similar test conditions reported that 12.5 γ/ml. of gela- tin gave considerable protection while 25 γ/ml, of gelatin gave complete protection for 6 months. If the inactivation of bacteriophages is due to some change occurring at the surface of gas bubbles produced in the fluid by shaking, this same kind of in activation should occur when an inert gas is bubbled through a suspension of the virus. Accordingly 25 ml. of buffer- diluent at pH 6.5 containing phage T2r at a concentration of 2.5×104 infectious particles per ml. were placed in a Corning sintered glass filter of coarse grade. This was held in a water bath at 30°C. and nitrogen gas was bub- bled through the filter at the rate of 1 liter per minute. Samples were withdrawn at 5 minute intervals for an hour without interrupting the gas flow. The inactivation followed the kinetics of a first order reaction through- out this time with a velocity constant of 0.047 minute-1, as compared with 0.23 minute-1 for shaking with air at 26°C. Similar results were obtained with T7 although the rate was somewhat faster with this phage. DISCUSSION The denaturation of proteins probably involves the unfolding of a highly specific globular structure into a relatively unspecific polypeptide chain. This change exposes hitherto hidden -SH and phenolic groups to the action of chemi cal reagents, and results in the loss of solubility and of the specific physiological activity of the protein. Denaturation may be brought about by the action of heat, of chemicals such as urea, of detergents, of excessive concentrations of H+ or OH’ ions, and by shak- ing. All of these denaturing agents will also bring about a destruction of the infectious properties of viruses. That vigorous shaking will cause the precipitation of proteins from solution has been known for a long time. The precipitation of egg albumin from 1 per cent solu- tions on vigorous shaking follows the course of a zero- order reaction since the high concentration of protein maintains the gas-liquid interface in a saturated condi- tion. The rate-limiting factors are the amount of surface available, and the rate at which the surface is renewed by agitation (11) With highly diluted proteins however, one might expect the kinetics of the reaction to be first order since the number of protein molecules arriving at the surface in unit time will be proportional to the pro- tein concentration. There is very little data on this point in the literature. Shaklee and Meltzer (4) in 1909 studied the effect of shaking on the stability of pepsin in HCl. From their data it can be calculated that the inactiva- tion of pepsin follows the course of a first order reaction with a velocity constant of 0.029 minute-1- at 33°C. Since no charac terization of the pepsin was made it is impos- sible to say how much pepsin was present or even how Figure 5. The inactivation of phage T5, shaken in the presence of various amounts of bovine serum albumin. 75Surface Inactivation of Bacterial Viruses and of Proteins much total protein was present in the shaking experi- ments. However, it is significant that the addition of pep- tone stabilized the pepsin, there being a loss of only 25 per cent of the pepsin activity on shaking for 24 hours at 33°C. in the presence of peptone. Shaklee and Meltzer made certain observations that agree closely with our own observations on the inactivation of bacteriophage by shaking, namely: 1. Presence of glass beads did not accelerate shaking inactivation. 2. No inactivation of shaking full bottles, with or with- out glass beads. 3. Results in paraffined bottles were same as in non- paraffined glass bottles. 4. Results in sealed glass tubes were same as in rubber- stoppered bottles. 5. Inactivation rate increased with increasing acidity. 6. Results were the same with air, CO2, or H2 as the gas phase. MacFarlane and Knight (17) in 1941 studied the α toxin of Cl. welchii which they demonstrated to be an enzyme, lecithinase. This enzyme when highly dilut- ed was rapidly inactivated by bubbling air or nitrogen through the enzyme solution. They did not follow the course of the inactivation over a sufficient range of activ- ities to make it possible to decide whether the kinetics are those of a zero order or a first order reaction. It has been observed repeatedly that physiologically active proteins on high dilution often show a spontane- ous loss of activity which may be prevented by carry- ing out the dilution procedure in the presence of other proteins. In Table IV is listed a number of examples of this phenomenon culled from the literature. Included are the concentration at which the activity of the protein in question is measured and at which the inactivation is observed, together with the concentration of added pro- tein which has been found to prevent this inactivation. It should be noted that in many cases the concentra- tion given is the lowest concen tration of protein tested for protective effect since no titration of the protecting protein was made. It may also perhaps be significant that many of the enzyme activities are assayed in a Warburg or similar manometric apparatus in which a vigorous shaking of a highly diluted enzyme preparation is part of the assay procedure. From Table IV it may be noted that the physiologi- cally active proteins with which this type of instability has been observed are all proteins in which the specif- ic activity is measured at a final protein concentration of 4 γ/ml. or less. Presumably proteins which must be assayed at higher concentrations do not show this phe- nomenon. Also it may be noted that where the protect- ing protein has been assayed, the amount required has varied from 1 γ/ml. of gelatin in the case of short dura- tion experiments with bacteriophage to 25 γ/ml. of Table IV. A summary of data from the literature concerning physiologically active proteins which are unstable when highly diluted, includ- ing the concentration at which the protein is usually assayed and at which its lack of stability is noted, and the concentration of protective protein employed to stabilize it. Physiologically active protein Concentration at which protein is markedly unstable Concentration of protective protein employed Reference Diptheria toxin in Schick test 0.02 to 0.2 γ/mi. 1 mg./ml. serum albumin 15 25 γ/ml. gelatin 16 Tetanus toxin M. L. D. is 3×10-4 γ protein 10 mg./ml. peptone* 18 α toxin of Cl. welchii or leci- M.L.D. is 0.2 to 0.5 y 5 mg./ml. gelatin* 19 thinase γ protein 10 mg. /ml. serum albumin* 20 Botulinus toxin M.L.D. is 10-4 γ pro tein 2 mg./ml. gelatin* 21 Invertase 2 to 4 γ/m1. gelatin 22 Tyrosinase 1 γ/ml. 10 γ/m1. gelatin 13 Ascorbic acid oxidase 1 γ/ml. 6 γ/m1. gelatin 23 Carbonic anhydrase 1.6 γ/mi. 33 γ /ml. peptone 24 Catalase <3 γ/ml. 25 Desoxyribonuclease 3 γ/ml. 100 γ/ml. gelatin* 26 Hexokinase 4 γ/ml. 6 γ/m1. insulin or 14 60 γ/nil. serum albumin α glycerophosphate dehydro genase 2.5 γ/ml. 1 mg./ml. gelatin* 27 Bacteriophage 104 particles/ml. 1 to 10 γ/m1. gelatin * Protective effect not titrated, concentration given is lowest one tested or only one given in reference cited. 76 Mark H. Adams gelatin needed to stabilize Schick toxin for 6 months at room temperature. Serum albumin when it has been compared with other proteins such as insulin or gela- tin has been much less effective as a protecting agent. It is highly significant that proteins present in solutions of less than a few γ/ml. concentration are highly unsta- ble, and that they are protected from inactivation by the presence in so lution of other proteins at a concen- tration higher than a few γ/ml. It has been shown that proteins will unfold at a gas-liquid interface to form a monomolecu lar film about 10Å thick and covering an area of about 10 cm.2/γ of protein (28). This protein film is insoluble in water, and once formed on a quiet sur- face will effectively prevent more protein molecules of the same or different type from reaching the surface. On stirring or agitation of the surface however, the pro- tein film will be folded upon itself to form an insoluble coagulum of denatured protein, leaving a fresh inter- face for the unfolding of additional protein. A physi- ologically active protein present at a concentration of 1 γ/ml. could then be completely spread and inactivated at a total interface corresponding to 10 cm.2/ml., an area readily obtainable with very little shaking. In the pres- ence of a second protein, the rate of inactivation of the physiologically active protein would be a function of the relative concentrations of the two proteins, of their respective diffusion constants, and of the relative ease with which they unfold once they reach the surface. A protective protein present at a concentration of 10 γ/ml. should effectively exclude from the surface a protein of similar proper ties present at a concentration of 1 γ/ml. Also if a physiologically active pro tein is present at a concentration of 10 γ/ml. or higher, the available surface will be saturated with an undetectably small fraction of this protein and hence no loss in activity will be noticed unless the shaking is more violent and prolonged than in the usual assay procedures in the Warburg apparatus for instance. Langmuir and Schaefer (29) derived an equation for the diffusion of solute molecules to the surface, assum- ing only that every molecule which reached the surface stayed at the surface. This is a reasonable assumption for protein molecules if every molecule which reaches the surface unfolds into a film. The equation is where n is the amount of protein reaching 1 cm.2 of surface in time t, no is the concentration of protein per cm.’ and D is the diffusion constant of the protein. For egg albumin at 20°C. and a concentration of 100 γ/m1., the surface should be saturated in 1 second, whereas at a concentration of 5 γ/ml. it would take 26 minutes to saturate the surface. Bull (30) measured the rate of fall of surface tension with time in solutions containing various concentrations of egg albumin. At albumin concentra- tions higher than 50 γ/m1., the major portion of the sur- face tension drop occurred in less than a minute, while at a concentration of 5 γ/ml. there was no noticeable drop for several minutes, then the major fall in surface tension occurred between 5 and 15 minutes, the surface tension ap proaching the equilibrium value in 30 minutes. At a concentration of egg albumin of 1 γ/m1., the albumin will reach the sur face in the quantity of 10-2 γ per cm.2 of surface in 100 seconds. If the surface to vol- ume ratio is increased by shaking or bubbling it is obvi- ous that a large proportion of the total protein would reach the surface in a fairly short time especially since Langmuir and Schaefer (29) point out that in stirred solutions the amount of solute reaching the surface is proportional to time rather than to the square root of time as it is in solutions at rest. Failure to realize that the concentration of pro- tein in solution was the critical factor in determining whether or not rapid spontaneous inactivation occurred on dilution, has resulted in the publication of probably erroneous conclusions. For instance Traub, Hollander, and Friedemann (31) concluded that broth and serum “potentiated” the lethal action of tetanus toxin. They considered the possibility that the added broth or serum prevented the inactivation of tetanus toxin but discarded this explanation, largely on the grounds that if the low titer of toxin in saline were due to inactivation it would have to occur with unreason able rapidity, and because “potentiation” occurred in the case of titrations in small animals such as mice but not in large animals such as rabbits. Exami nation of their data reveals that with tox- in lot 1556 the lethal dose in rabbits is 0.1 ml. of a 1/10 dilution in either broth or saline; whereas in the guin- ea pig the lethal dose is 0.1 ml. of a 1/2000 dilution in saline, and 0.1 ml. of a 1/128,000 dilution in serum. It seems not unreasonable to assume that culture filtrates containing tetanus toxin when diluted beyond 1/2000 in saline contain less than the critical 1 γ/ml. of protein; especially since in the titrations recorded in their paper, the potentiating effect of broth decreased to almost noth- ing when the broth was diluted 1/1000 in saline. The observations of Traub et al. on poten tiation can be satis- factorily explained on the assumption that tetanus toxin is markedly unstable when it is diluted beyond a limiting value for total protein concentration, and that dilution in the presence of small amounts of protein prevents this loss of activity. 77Surface Inactivation of Bacterial Viruses and of Proteins In the present paper we have discussed the inactiva- tion of viruses by shaking as a process quite analogous to the surface denaturation of proteins. We do not picture the inactivation of the virus as necessarily involving an unfolding of the entire virus particle into a protein layer 10Å thick. In fact Seastone has shown (32) that tobacco mosaic and vaccinia viruses do not readily unfold in the way that egg albumin does, but that never the less these viruses do form surface films. We merely suggest that once the virus reaches a gas-liquid inter- face it is subjected to such forces that it may very rapidly be deprived of the property of infectivity. This loss of infectivity may be prevented by saturating the gas-liquid interface with another protein, thereby denying the virus acess to the surface. In this respect the phenomenon is analogous to the surface denaturation of proteins. The prevention of surface denaturation is not the only protective role which may be played by proteins. Sumner (33) has demonstrated that dilute solu tions of crystalline urease are rapidly inactivated by traces by heavy metals. This type of inactivation can be pre- vented by the addition of proteins, as well as by gum arabic, hydrogen sulfide, amino acids, and many other substances. Urease can similarly be protected by pro- teins against inactivation by small amounts of oxidiz- ing agents. Proteins should play a similar role in the pro tection of viruses against the inactivating effects of heavy metals and oxidizing agents. It is probably a sum- mation of these various protective mechanisms which is responsible for the generally recognized fact that viruses are more stable when diluted in serum or broth than when diluted in salt solutions or distilled water. SUMMARY 1. The seven bacterial viruses of the T group active against E. coli, are rapidly inactivated at gas-liquid interfaces. 2. The kinetics of this inactivation whether brought about by shaking or by bubbling with nitrogen are those of a first order reaction. 3. This inactivation may be prevented by the addition of enough protein to maintain the gas-liquid inter- face in a saturated condition. 4. The analogy between this phenomenon and the sur- face denaturation of proteins is pointed out and dis- cussed. The author wishes to acknowledge his indebtedness to Miss Nancy J. Collins for technical assistance. Addendum. — Since submitting this manuscript, we have found a paper by J. Steinhardt (34) on “The stability of crystalline pepsin” in which the inactivation of pepsin by shaking is noted. At a pH of 6, temperature of 25°C., and pepsin con centration of about 30 to 60 micrograms per ml., pepsin is inactivated by shaking in accordance with the kinetics of a first order reaction. The velocity constant was independent of pH over the range of 4 to 6 but was somewhat dependent on the rate of shaking. The inactivated pepsin separated from solution as an insolu- ble suspension. REFERENCES 1. Campbell-Renton, M. L., J. Path. and Bact., 1937, 45, 237. 2. Grubb, T. G., Miesse, M. L., and Puetzer, B., J. Bact., 1947, 53, 61. 3. McLimans, W. F., J. Immunol., 1947, 56, 385. 4. Shaklee, A. 0., and Meltzer, S. J., Am. J. Physiol., 1909, 25, 81. 5. Glenny, A. T., Pope, C. G.,Waddington, H., and Wallace, V., J. Path. and Bad., 1925, 28, 471. 6. Langmuir, I., and Waugh, D. F., J. Gen. Physiol., 1938, 21, 745. 7. Demerec, M., and Fano, U., Genetics, 1945, 30, 119. 8. Delbriick, M., Biol. Rev., 1946, 21, 30. 9. Hershey, A. D., Kalmanson, G., and Bronfenbren- ner, J., J. Immunol., 1943, 46, 267. 10. Hershey, A. D., in Cold Spring Harbor Symposia on Quantitative Biology, Cold Spring Harbor, Long Island Biological Association, 1946, 11, 67. 11. Bull, H. B., in Cold Spring Harbor Symposia on Quantitative Biology, Cold Spring Harbor, Long Island Biological Association, 1938, 6, 140. 12. Gottschall, R., and Bunney, W. E., J. Immunol., 1938, 34,103. 13. Adams, M. H., and Nelson, J. M., J. Am. Chem. Soc., 1938, 60, 2472. 14. Berger, L., Slein, M. W., Colowick, S. P., and Cori, C. F., J. Gen. Physiol., 1946, 29, 379. 15. Edsall, G., and Wyman, L., Am. J. Pub. Health, 1944, 34, 365. 16. Moloney, P. J., and Taylor, E. M., Tr. Roy. Soc. Cana- da, Section V, 1931, 25, 149. 17. MacFarlane, M. G., and Knight, B. C. J. G., Bio- chem. J., 1941, 35, 884. 18. Pillemer, L., J. Immunol., 1946, 53, 237. 19. Adams, M. H., J. Immunol., 1947, 56, 323. 20. Zamecnic, P. C., Brewster, L. E., and Lipmann, F., J. Exp. Med., 1947, 85, 381. 21. Abrams, A., Kegeles, G., and Hottle, G. A., J. Biol. Chem., 1946, 164, 63. 78 Mark H. Adams 22. Saul, E. L., and Nelson, J. M., J. Biol. Chem., 1935, 111, 95. 23. Lovett-Janison, P. L., and Nelson, J. M., J. Am. Chem. Soc., 1940, 62, 1409. Powers, W. H., Lewis, S., and Dawson, C. R., J. Gen. Physiol., 1944, 27, 167, 181. 24. Scott, D. A., and Mendive, J. R., J. Biol. Chem., 1941, 139, 661. 25. Sumner, J. B., in Advances in Enzymology and Related Subjects of Biochemistry, (F. F. Nord, edi- tor), New York, Interscience Publishers, Inc., 1941, 1, 165. 26. McCarty, M., J. Gen. Physiol., 1945, 29, 123. 27. Racker, E., personal communication. 28. Gorter, E., in Chemistry of Amino Acids and Pro- teins, (C. L. A. Schmidt, editor), Baltimore, Charles C. Thomas, 1938, 428. 29. Langmuir, I., and Schaefer, V. J., J. Am. Chem. Soc., 1937, 59, 2400. 30. Bull, H. B., Physical Biochemistry, New York, John Wiley and Sons, Inc., 1943, 199. 31. Traub, F. B., Hollander, A., and Friedemann, U., J. Bact., 1946, 52, 169. 32. Seastone, C. V., J. Gen. Physiol., 1938, 21, 621. 33. Sumner, J. B., and Hand, D. B., J. Biol. Chem., 1928, 76, 149. 34. Steinhardt, J., K. Danske Vidensk. Selsk., Mat.-fys. Medd., 1937, 14, No. 11. Substantia An International Journal of the History of Chemistry Vol. 7, n. 1 – 2023 Firenze University Press Superbugged Pierandrea Lo Nostro Equivalence of Electromagnetic Fluctuation and Nuclear (Yukawa) Forces: the π0 Meson, its Mass and Lifetime Barry W. Ninham1, Iver Brevik2, Mathias Boström3,4 The Rate Constant – Reaction Free Energy Dependence for the Electron Transfer Reactions in Solutions. The Way to Interpret the Experimental Data Correctly Lev I. Krishtalik1,† Training of Future Chemistry Teachers by a Historical / STEAM Approach Starting from the Visit to an Historical Science Museum Valentina Domenici A New Response to Wray and an Attempt to Widen the Conversation Eric Scerri Boxing Partula: 25 Years After Stephen T. Hyde Surface Inactivation of Bacterial Viruses and of Proteins Mark H. Adams Johann Beckmann (1739-1811) and Modern Chemical Technology Juergen Heinrich Maar Kuroda Chika (1884-1968) – Pioneer Woman Chemist in Twentieth Century Japan Yona Siderer Review of A Cultural History of Chemistry. Peter J. T. Morris and Alan Rocke, eds., Bloomsbury Academic: London, 2022 Robert H. Crabtree1, Arthur Greenberg2, Seth C. Rasmussen3