The Reactivity of the Sulfhydryl Groups in Normal Bovine Lens*

The Reactivity of the Sulfhydryl Groups in Normal Bovine Lens*

THE REACTIVITY OF THE NORMAL LORENZO O . MEROLA, B.S., AND JIN Boston, GROUPS IN LENS* H. KINOSHITA, PH.D. Massachusetts INTRODUCTION ...

643KB Sizes 0 Downloads 3 Views

THE

REACTIVITY

OF

THE

NORMAL LORENZO O .

MEROLA,

B.S.,

AND JIN

Boston,

GROUPS

IN

LENS* H.

KINOSHITA,

PH.D.

Massachusetts

INTRODUCTION

One of the most widely studied chemical aspects of the cataractous process is the change in the sulfhydryl groups (-SH).'"''' There seems to be general agreement that in experimental cataract formation, the amount of lens glutathione drops markedly before any sign of opacification. O n the other hand, as evidenced by irradiation cataract studies of Pirie and Van Heyningen,' the loss of protein - S H ( P - S H ) is observed at a much later stage in the formation of a cata­ ract. These results suggest that the reactivity of the - S H of glutathione ( G S H ) is differ­ ent from that of lens protein. The disappearance of the - S H groups .seems to involve an oxidation resulting in the formation of disulfide linkages as demon­ strated by Dische and Zil.* These investi­ gators also suggest that the oxidation of - S H occurs not only in the cataract process but also in the transformation of soluble to the insoluble protein. In view of the many studies on the change of - S H levels in the cataractous lens, it was felt that an investigation on the factors which influence the reactivity of - S H groups of normal lens would be worth while. In this study, the ability of the - S H groups to undergo oxidation was used as the meas­ ure of their reactivity. This index of reactiv­ ity seems particularly appropriate since the oxidation of - S H groups appears to play an important role in establishing the state of lens protein. METHODS P R E P A R A T I O N OF L E N S

SULFHYDRYL

BOVINE

HOMOGENATES

Lenses were removed from the eyes of freshly slaughtered cattle and were washed with glass redistilled water. After removal of excess water by blotting with filter paper. 326

the lenses were weighed and then homogen­ ized in a W a r i n g blender at 5 ° C . under nitrogen in isotonic Tris+ bunker at p H 7.4. The amount of buffer used for homogenization was to give a final dilution of one lens in 20 cc. of buffer. T h e insoluble material of the lens was removed by centrifugation at 10,000 X g for 15 minutes at 5 ° C . PROCEDURE

FOR

LENS

-SH

OXIDATION

STUDIES

In an Erlenmeyer flask, which was shaken in a W a r b u r g bath, 20 cc. of the homogenate containing the soluble lens protein was in­ cubated at 2 5 ° C . or 3 7 ° C . in the presence of cupric ion (2 X 10-=^M). T h e flask was flushed for 10 minutes with 100-percent oxygen and then clamped off. A n aliquot of the homogenate was removed every hour and the total - S H was determined by amperometric titration (described below). T h e G S H was measured on a protein free filtrate by the nitroprusside method.' In some cases, the amperometric titration method was used. In either case both methods gave the same values. The oxidized glutathione was deter­ mined by the electrolytic reduction method previously described.' T h e method of Roe** was employed for the determination of as­ corbic acid. AMPEROMETRIC

TITRATION

T h e lens P - S H groups were determined by the difference between the amount of total - S H obtained by amperometric titration and the G S H present. T h e amperometric titra­ tion employed was the method of Benesch * From the Howe Laboratory of Ophthalmology, Harvard Medical School. This work was supported in part by the United States Atomic Energy Com­ mission, and by a Graduate Training Grant P.H.S. No. 2-B-S096 from the National Institute of Neuro­ logical Diseases and Blindness, Public Health Serv­ ice. t Tris (hydroxymethyl) aminomethane.

SULFHYDRYL GROUPS IN BOVINE LENS et al.* with slight modifications. A n external potential was applied (—O.lOv.) and the reference electrode was a mercury pool. A more detailed description of this method with the modifications will be presented in a future publication. T h e titration was carried out in 100 cc. electrolytic beaker containing 30 cc. of a supporting electrolyte made up by 4.0 cc. of l.OM Tris, 3.4 cc. of l.OM HNO3, and 0.3 cc. of l.OM KNO3. T o this media a solution containing approximately 1 μΜ of - S H was added and titrated with 1 X 10-='M AgNOa solution. T o prevent possible oxida­ tion during the titration nitrogen was bub­ bled into the titrating mixture. However, when the galvanometer readings were made, the bubbling was stopped to prevent disturb­ ances of the diffusion current. During this short interim nitrogen was passed over the titrating mixture. The apparatus and pro­ cedure were checked every day by titrating standard amounts of G S H . T o demonstrate that certain substances of the lens homog­ enate did not interfere with this procedure, known quantities of G S H were added to the homogenate and titrated. A 99 to 100-percent recovery was obtained in all cases. RESULTS AND DISCUSSION

A s can be seen in Table 1, considerable quantities of - S H groups are found in bo­ vine lens. T h e total - S H content is 54.1 μΜ per gm. wet weight of lens. Of this amount 9.4 μΜ is G S H and 44.7 μΜ is P - S H . As mild as possible oxidizing conditions were employed to study the factors which influence the oxidation of - S H groups of TABLE I SuLFHYDRYL CONTENT OF BOVINE LENS

No. of Obser­ vations

Type of -SH groups

AiM-SH/gm. lens (wet)

15

Total -SH

54.1+3.6 S.D.

IS

Protein -SH

44.7+3.35 S.D.

IS

Glutathione (Reduced)

9.4 + 0.56S.D.

327

lens. If strong oxidizing agents such as hydrogen peroxide are used, the oxidation is so rapid that it becomes difficult to demon­ strate the diflierences in the reactivity of the various - S H groups. O n the other hand, if simply an atmosphere of oxygen is used then the - S H oxidation is so slow that it becomes impractical to study. In the experi­ ments to be reported here, the conditions employed were to catalyze the oxidation with 2 X lO-'M cupric ion and to use 100 percent O2 as the gas phase. F r o m the preliminary experiments it was apparent that in the intact lens the level of - S H groups of protein and G S H did not change significantly when lens was incubated at neutral reaction under the oxidizing con­ ditions described above. Therefore, it was necessary to use homogenized lens to study the factors which influence the oxidation of - S H groups in lens. In Figure 1, the extent of the disappearance of the - S H groups of lens homogenate is shown. In contrast to the findings with the intact lens, homogeniza­ tion of the lens and subsequent incubations under conditions favorable for oxidation, re­ sulted in a rapid disappearance of G S H . T h e effect of the presence of catalytic amounts of cupric ion is demonstrated in the oxidation of lens G S H . In the presence of cupric ion the disappearance of G S H is complete within one hour, but without added cupric ion five hours are required for complete oxidation. In both cases the disappearance of G S H seems to be due to its oxidation to the di­ sulfide form since most of it is recovered when the protein-free filtrate is subjected to electrolytic reduction. In a previous paper' from this laboratory, evidence was presented suggesting the pos­ sibility that G S H in the intact lens appears to be bound in some manner. F r o m the studies on the extractability of G S H it ap­ peared that this peptide was not freely diffusable, and from the effect of concentrated urea and guanidine solutions it was sus­ pected that binding of G S H involves hydro­ gen bonds. T h e fact that in the intact lens

LORENZO O. MEROLA AND JIN H. KINOSHITA

328

OXIDATION of L E N S - S H

45 25837 40

ζ

lU

i

in

15,

•s 1.

25 8 37

Fig. 1 (Merola and Kinoshita). Oxidation of lens sulfhydryl groups. Lens homogenates were in­ cubated with and without 2 χ lO^M cupric ion in the presence of 100-percent oxygen. only a slight amount of G S H is capable of being oxidized is consistent with the hy­ pothesis that G S H is bound in lens. F o r if it were freely diffusible, oxidation, which re­ quires the collision and interactions of two molecules of G S H , would take place. In the intact lens this does not occur to any signifi­ cant extent. W h e n the lens is homogenized, however, the G S H being a tripeptide not having many sites to establish strong hydro­ gen bonding is released from these weak binding sites and is able to undergo oxida­ tion. I n contrast to the ease with which G S H is oxidized, the P - S H groups, even in lens homogenates, are particularly resistant to the oxidizing conditions employed. A s is shown in Figure 1, the level of P - S H in the presence of cupric ion does not change significantly. In five hours at 2S°C. or 37°C. no more than seven percent of the P - S H groups disappears. It, therefore, appears that

the lens P - S H groups are not easily suscep­ tible to oxidation. A noteworthy fact emerges from this experiment. G S H and ascorbic acid have been considered by many as serving some type of protective action on the P - S H groups.^ It is believed that these reducing substances, found in such high concentra­ tions in lens, help maintain the - S H groups of lens protein in the reduced state. The evi­ dence from our experiments does not seem to support this idea. F r o m Figure 1, it can be seen that in the copper catalyzed homo­ genates, the level of P - S H is still maintained even after all the G S H is oxidized. T h e ex­ periments with resin-treated homogenates also indicate that G S H does not participate in maintaining the P - S H level. F o r this study, lens homogenates were passed through a column of Dowex 1 on a nitrate cycle. The G S H was removed completely by this treat­ ment but no loss in the P - S H was observed. Incubation of the resin-treated homogenates under the oxidizing conditions for five hours resulted only in a slight decrease in the amount of P - S H content (five percent). These studies are further supported by the experiments on dialyzed lens homogenates. I n these experiments, lens homogenates were dialyzed against 80 volumes of isotonic po­ tassium nitrate for 24 hours at 5°C. T h e dialysis was effective in removing all the G S H and ascorbic acid as checked by colori­ metric tests. Dialysis under these conditions did not change the amount of P - S H which is further indication of the stability of this group. The dialyzed homogenates when incu­ bated at 25°C. and 37°C. and subjected to the oxidizing conditions revealed a P - S H oxidation curve which was practically identi­ cal as the one observed in the experiment with undialyzed homogenates (fig. 1 ) . From these results it would appear that the P - S H level of dialyzed lens homogenates does not change for five hours when exposed to the oxidizing conditions of these experitnents. Therefore it seems that the absence of G S H or ascorbic acid does in no way influence the level of lens P - S H .

SULFHYDRYL GROUPS IN BOVINE LENS It would appear that the reason why the lens P - S H are so unreactive is that they are "masked." The "masking" is thought to be due to hydrogen bonding between the sulf­ hydryl groups and neighboring atoms. These bonds have been demonstrated to be broken by high concentrations of compounds which also have a strong tendency to form hydro­ gen bonds. T h e most effective of these re­ agents are urea and guanidine solutions.'"''' T o demonstrate the "masked" state of the lens P - S H , lens homogenate was treated with 8.0M urea solutions and subjected to the oxidizing conditions. T h e results of incu­ bation studies at 25°C. and 37°C. are shown in Figure 2. It appears that the presence of 8.0M urea has a marked effect on the re­ activity of the P - S H . At both temperatures there is an increase disappearance of P - S H . As might be expected, the rate of disappear­ ance at 37°C. is much faster than 25°C. W e believe that the disappearance of the - S H groups is due to an oxidative process prob­ ably involving the formation of disulfide bonds. T h e evidence for this is that the dis­ appearance of the P - S H and G S H is pre­ vented when O2 is replaced by N2. T h e de­ termination of O2 uptake by dialyzed lens homogenate is now being undertaken. A few preliminary studies have revealed that when 8.0M urea is added to the dialyzed homoge­ nate, there is an increase in O2 consumption concomitant with the disappearance of P-SH. T h e increase in reactivity of P - S H by urea treatment indicates that the disruption by hydrogen bonds has brought about the " u n ­ masking" of the - S H groups. Other investi­ gations appear in the literature which sup­ port the findings that lens P - S H are "masked." Nordmann et al.'^ have shown that the O2 consumption of boiled lens ho­ mogenate is much higher than unboiled ho­ mogenate. The O2 consumption is inhibited by lead acetate presumably by tying u p the "unmasked" - S H . Heat denaturation, al­ though not as effective as urea treatment, is another means by which lens P - S H may be "unmasked." Another investigation which

329

EFFECT of 8M UREA ON -SH OXIDATION

Fig. 2 (Merola and Kinoshita). Effect of .8.0 Μ urea on sulfydryl oxidation. Lens homogenates were incubated with 2 χ 10"°M cupric ion in the presence of 100-percent oxygen. may be cited is that by G l o s t e r " who studied the reduction of dehydroascorbic acid in lens extract. In attempting to show that the re­ duction of dehydroascorbic acid was not due to an enzymatic process, he presented evi­ dence that boiled lens extract reduced de­ hydroascorbic acid more effectively than did the untreated lens extract. H i s observation may be explained by the fact that heat de­ naturation has served to increase the reac­ tivity of the lens P - S H to reduce dehydro­ ascorbic acid. In our experiments, we were also able to demonstrate the reduction of dehydroascorbic acid by urea-treated lens homogenates. I n the incubation studies on the ureatreated homogenates, an observation on the G S H oxidation was made for which we have

330

LORENZO O. MEROLA AND TIN H. KTNOSHITA

no adequate explanation. T h e oxidation of G S H is markedly inhibited in the ureatreated lens homogenate (fig. 2 ) . In the un­ treated homogenate the oxidation of G S H is complete within 1 hour. However, in the urea-treated homogenates the oxidation of G S H is much slower, although that of the P - S H groups is rapid. Since the G S H oxida­ tion curve of the urea-treated homogenates resembles the uncatalyzed G S H oxidation curve in Figure 1, there existed a possibility that the urea was complexing the cupric ion. This does not seem to be the case since the oxidation of pure G S H in buffer, in the absence of lens protein, catalyzed by cupric ion occurs at the same rate with and without the presence of 8.0M urea. T h e r e are a num­ ber of other possible explanations, but as yet, no experimental support has been ob­ tained for any of these. It would appear from these experiments in lens homogenates that the P - S H groups are "masked" and not particularly reactive as tested by the conditions of our experiments. A rather drastic treatment of urea denatura­ tion is necessary to activate these - S H groups so that they can undergo oxidation. In contrast to these experimental observa­ tions, in cataract fomiation the conditions apparently become favorable so that the hy­ drogen bonds which " m a s k " the - S H groups are broken and the oxidation of these groups readily takes place. These facts suggest that in either cataracts or in lens of aged animals.

where an oxidation of sulf hydryl groups oc­ curs, there must first be a denaturation proc­ ess which unfolds the peptide chains of lens protein so as to facilitate the oxidation. SUMMARY

The level of P - S H groups in normal bo­ vine lens as determined by amperometric titration was found to be 44.7μΜ - S H per gm. lens (wet w e i g h t ) . This concentration of P - S H is approximately four times that of GSH.

In the intact lens both the P - S H and G S H appear to be unreactive toward oxidation. The difference in reactivity of the - S H groups of protein and glutathione becomes apparent in the study of lens homogenate. T h e G S H is readily oxidized to the disulfide form. O n the other hand the P - S H groups appear to be quite stable toward oxidation. Compounds such as ascorbic acid and G S H which are dialyzable or removed by ion exchange seem to be unnecessary to maintain the level of P - S H groups. The effect of concentrated urea solution on the reactivity of the - S H groups of pro­ tein seem to indicate that in the normal state these groups are marked. 243 Charles Street (14). ACKNOWLEDGMENT

We wish to express our gratitude to Dr. W. Morton Grant and Dr. Charles Rife of the Howe Laboratory, for their advice and counsel on the problems of electrochemistry.

REFERENCES

1. Bellows, J. G.: Cataracts and Anomalies of the Lens. St. Louis, Mosby, 1944. 2. Pirie, Α., and Van Heyningen, R.: Biochemistry of the Lens. New York, Oxford, 1956. 3. Pirie, .Λ., and Van Heyningen, R., and Boag, J. W.: Changes in lens during the formation of X-ray cataracts in rabbits. Biochem. J., 54:682-688, 1953. 4. Dische, Z., and Zil, H.: Studies on the oxidation of cysteine to cystine in lens protein during cataract formation. Am. J. Ophth., 34:104-113 (May, Pt. II) 1951. 5. Dische, Z., Borenf reund, E., and Zelmenis, G.: Changes in lens protein of rats during aging. Arch. Ophth., 55 :471-483, 1956. 6. Dische, Z., Borenfreund, E., Zelmenis, G.: Proteins and protein synthesis in rat lenses with galactose cataract. Arch. Ophth., 55 :633-642, 1956. 7. Kinoshita, I. H., and Masurat, T.: Studies on the glutathione in bovine lens. Arch. Ophth., 57:266274. 1957. 8. Roe, J. H.: Methods of Biochemical Analysis. New York, Interscience, 1954, v. 1, pp. 121-24. 9. Benesch, R. E., Lardy, H. Α., and Benesch, R.: The sulfhydryl groups of crvstalline proteins. J. Biol. Cliem., 216 :663, I9SS.

SULFHYDRYL GROUPS IN BOVINE LENS

331

10. Benesch, R., Benesch, R. E., and Rogers, W. I.: The Reactivity of the Sulfhydryl Group in Glutathione and Related Peptides. Symposittm on Glutathione, Edited by S. Colowick. New York, Aca­ demic Press, 1954, pp. 31-43. 11. Simpson, R. B., and Kauzman, W.: The kinetics of protein denaturation: I. The behavior of the optical rotation of ovalbumin urea solution. J. Am. Chem. Soc, 75:5139-5152, 1953. 12. Nordmann, J., Mandel, P., and Izraelewicz, D.: Enzymatic character of respiration of the lens. Arch. Ophth., 52:42-45, 1954. 13. Gloster, J.: Reaction between dehydroascorbic acid and dialyzed lens extract in vitro. Brit. J. Ophth., 40:536-544, 1956.

DISCUSSION DH. ZACHARI.AS D I S C H E (New York): I was very much interested in this impressive work which was carried out with such refined analytic proce­ dures, and I was particularly interested because for the last few years our laboratory was carrying on investigations on the oxidation of lens proteins. We tried to obtain evidence that the change in certain physical or chemical factors in the lens might be responsible for an increase in the oxidation of the proteins. We approached this problem in a somewhat dif­ ferent way than Dr. Merola and Dr. Kinoshita. They were apparently intent upon creating condi­ tions which could be regarded as physiologic, that means to obtain oxidation of sulfhydryl groups by a catalyst. Dr. Croisy, in our laboratory, tried something similar, although not on such a simplified system, by investigating the effect of phosphate on the oxidation of the sulfhydryl groups of rat lens proteins. Phosphate forms complexes with heavy metals, and phosphate solutions are a well-known oxidizing agent. Our results, which we presented about a year and a half ago at the eastern regional meeting of this association, agreed to a certain ex­ tent with the results of Dr. Merola and Dr. Kino­ shita, in so far as we obtained 10 to 15 percent oxidation of the total soluble protein of rat lenses with phosphate. One possible objection that could be made against this type of experiments in which catalysts are used, is that the catalyst could be bound by proteins (as is very probable in the case of copper), and of course in this case, if the cata­ lyst is bound by proteins, then we cannot be sure that the true availability of the sulfhydryl groups (if they are partly masked) can be determined, because what then determines the amount of oxi­ dation is not only the availability of the sulfhydryl groups but also the availability of the catalyst for oxidation. As we do not know the nature of the catalyst that is active in the lens, we may be completely misled. For this reason, in collaboration with H. Zil, we were also trying to determine the availability of the sulfyhydryl groups against oxidizing agents which were present in excess, a type of experiments which Dr. Merola and Dr. Kinoshita also appar­ ently tried, using hydrogen peroxide as the oxidiz­ ing agent in excess. In our experiments Potassium

ferricyanide and HjOj were used at pH 7.0 and above and Ferrichloride at pH 6.0 and below. When experiments are carried out, with an excess of oxidizing agents, the results are very different as far as lens protein of cattle are concerned from those presented by the authors, because if we oxi­ dize lens proteins by ferricyanide or even hydrogen peroxide, or heavy metals like ferric ions, then we obtain at physiological pH a much higher availa­ bility of SH groups, particularly at 37°C. Furthermore, we have evidence that this avail­ ability can change as the process of oxidation pro­ ceeds. This was particularly clearly shown by Dr. Zil, in our laboratory, who found that when the oxidation by ferric ions has proceeded to a certain extent, then it stops, but after a certain time inter­ val more SH groups become available to ferric ions and the oxidation process starts again. I do not think we can reject, a priori, such phe­ nomena as unphysiological, because we know that in the lens the oxidation of sulfhydryl group pro­ ceeds far beyond the few percentages which can be demonstrated by using catalysts. We must, there­ fore, rather assume that the availability of SH groups of lens proteins in vivo may be of the same order of magnitude as is observed against oxidiz­ ing agents present in excess of sulfhydryl groups. That means that secondary availability may be created by the oxidative process. We tried also to increase the availability of pro­ teins by the same procedures that the authors used, namely, by high concentration of urea and also some other agents, and we arrived at similar con­ clusions, although I must say that even at 25 °C. we obtained a much higher increase in availability by urea than appeared in experiments of the au­ thors. In my opinion this is probably due to very strong interactions between the catalyst and the lens protein on the one side, and the denaturing agent on the other side. We must consider the possibility of the forma­ tion of mixed chelates between the copper, the pro­ tein and the urea, which may completely change the course of oxidation. Phenomena of this kind may be responsible for some of the very interesting ob­ servations reported in this paper. I would like to express my great appreciation of the work done by the authors. It is very ele­ gant, clear, and conclusive. D R . L O R E N Z O O . M E R O L A (closing) : May I thank

332

LORENZO O. MEROLA AND JIN H. KINOSHITA

Dr. Dische for his very fine discussion. In answer to Dr. Dische's first question. We be­ lieve that in our system free cupric ion was present in sufficient quantities to have a catalytic effect. For if this were not the case, then we should not have obtained 100-percent oxidation of glutathione within one hour where cupric ion was used, while, without cupric ion present, it took five hours to obtain approximately the same amount of oxida­ tion. Oxidation of protein sulfhydryl groups was used to determine whether they were free or masked by hydrogen bonding. When strong oxidizing agents,

STUDIES

O N

PROTEIN

such as peroxide, were used the degree of proteinSH oxidation was dependent upon the concentra­ tion of the oxidizing agent. This suggested that this type of oxidizing agent appears capable of break­ ing hydrogen bonds, and, therefore, do not neces­ sarily give a true index of the availability of the normal protein sulfhydryl groups. We felt that by using mild oxidizing conditions, the protein-SH groups would not be oxidized unless they were free or until they were made available by treatment with urea. Again may we thank Dr. Dische for his fine discussion.

A N D

PROTEIN

D U R I N G T H E D E V E L O P M E N T O F GALACTOSE ZACHARIAS DISCHE,

M.D.,

GINEVRA ZELMENIS, M . D . ,

New

Previously,' it has been reported that when six-week-old rats are left for 21 days on a galactose diet inducing cataract, the total amount of soluble proteins as well as their concentration at the end of the feeding period is significantly lower than that of controls on normal diet, and about the same as it was at the beginning of the feeding period. This was interpreted as an inhibi­ tion of the synthesis of soluble proteins in rats on galactose diet. It has been, furthermore, shown that feed­ ing of galactose induces characteristic changes in the composition of the soluble proteins. These changes consist in an ac­ cumulation of a protein which, after oxida­ tion with ferricyanide a p H 7.4 at 37°C., becomes insoluble, and precipitates. T h i s protein which will be referred to as precipitable protein was also shown to be pres­ ent in smaller quantities in lenses of young rats on normal diet. Its amount in these cases ran parallel to the rate of protein syn­ thesis, so that it became insignificant at the end of the period of rapid growth of the lens. These observations suggested that this protein is an intermediate in the synthesis of lens proteins. * From the Department of Ophthalmology, Col­ lege of Physicians and Surgeons, Columbia Uni­ versity.

SYNTHESIS* CATARACT

AND JOSHUA YOULUS,

M.D.

York

A s there is evidence that the steady-state of soluble proteins in the lens is the result of the superimposition of a continuous breakdown of proteins on a process of syn­ thesis, the inhibition of the net synthesis could be due to an increase in the rate of the first, as well as to a decrease in the rate of the second process. In an attempt to reach an understanding of the mechanisms involved in the formation of galactose cataract, it seemed first of all necessary to determine whether the inhibi­ tion of the net protein synthesis and the accumulation of the abnormal protein ap­ pears immediately after the beginning of galactose feeding, and whether the degree of inhibition and accumulation changes sig­ nificantly during the period of feeding. It seemed, furthermore, of particular interest to obtain some indications as to the nature of the abnormal protein which precipitates after oxidation, and whether such an ab­ normal protein can also be found in lenses of other young animals. W e determined, therefore, the net synthe­ sis of soluble and insoluble proteins in lenses of rats kept for only six instead of 21 days on galactose diet and compared it with that of controls on normal diet. W e also de­ termined, the effect of oxidation by ferri­ cyanide on the soluble proteins of lenses of