- Email: [email protected]
An Investigation of High-Vacuum Freeze Drying as a Means of Drug Preservation. I*†
An Investigation of High-Vacuum Freeze Drying as a Means of Drug Preservation. I*J By MELVIN A. CHAMBERS$ and JOHN W. NELSON! Analo ous samples of belladonna leaves were oven fried and lyophilized. Various analytical determinations were made in an effort to determine what differences there are in the two samples as a result of the drying process. A lyophilizing apparatus on a laboratory scale has been developed and is shown in the photograph and in the diagram which has been drawn to scale for the convenience of those who would like to reproduce this for their own use.
reported that solanaceous leaves yield 2 per cent to 10 per cent more tropine alkaloids if the leaves are stabilized (4). It was further found that the temperature of drying had a significant influence on the degree of racemization (5). Fermentation of belladonna leaves (6) was carried out at 32" while preventing the loss of water, and upon an alkaloidal determination of the leaf, it was found to contain less alkaloid than the control. It was concluded that enzymes were the causative COMPARATIVE study of the effects produced by lyophilizing and oven drying belladonna agents for the loss in alkaloidal content. Thus leaves was made. The object was to determine if it can be seen that the temperature of drying the newer method of desiccation, lyophilization, affects the alkaloidal content. A method of drying without the aid of heat is is applicable to belladonna leaves, and to determine what changes, if any, other than desiccation lyophilization. With this method the moisture in the material is frozen and removed by means of are produced by drying at 50". The basis for the activities of plant cells is to be reduced pressure directly from the frozen state found in their enzymes. The rate at which an to the vapor state. It is sublimation of the enzymatic reaction proceeds is influenced not ice. This process has been used successfully with only by the temperature but also by the length the biologicals (7). It was possible to keep conof time the reaction mixture has already been valescent human sera for two or three years at that temperature. Within certain limits, an increase in the temperature increases the while retaining its full potency (8). Antisera and velocity of enzymatic reactions. If plant ma- antigens as well as certain related products, when terial is being dried within these limits, these dried in the frozen state and stored in vacuo, enzymatic reactions will continue for a period of suffered no deterioration in numerical values and time, thus altering the quantity of the constit- the dried products when reclaimed after a number of years showed no loss in potency. uents contained therein. Bacteria and yeasts have been preserved unThe effect of low temperatures (1) upon enchanged by lyophilization, and ip no instance zymic systems has been studied and it was found that the rate was retarded and there was no in- have the strains shown any changes even in activation of the enzymes which were studied. minor particulars (9). The apparatus that was used in this laboratory Attempts have been made to demonstrate hydrolis shown in the photograph while in operation and ysis in the solid state ( 2 ) . The rate of hydrolysis was found to be not more than 2 per cent in the accompanying diagram which was drawn of that occurringin supercooled liquid at the same to scale. This apparatus has proved to be very satisfactory. Pressures as low as 0.006 mm. of temperature. mercury were obtained which kept the material Upon application of these results to the desiccation of plant materials similar effects caused frozen while being dried. The drying chamber and moisture trap were by temperature would be expected. The effect 4l/1 and 4 inches in diameter, respectively, with of the temperature used for the drying of bella71/60 ground glass joints. Tubes of this size donna leaves has been noted by Koch (3). It is made it possible to lyophilize several hundred * Received October 29, 1949, from the College of Phar- grams of fresh material at one time. The mamacy Ohio State University Columbus Ohio t 6bstract of a thesis subkitted to th;?Graduate School in terial to be lyophilized was shell-frozen on the partial fulfillment of the requirements for the degree of sides of the drying chamber so as to expose a Doctor of Philosophy. X Fellow of the American Foundation for Pharmaceutical maximum of surface area to facilitate drying. Education: present address: College of Pharmacy, Universitv of Texas. Austin. The ground glass joints were sealed with Apiezon, $, Akociate Piofessor; College af Pharmacy, Ohio State a wax (obtained from James G. Biddle Co., University.
A
323
324
JOURNAL OF THE
AMERICAN PHARMACEUTICAL ASSOCIATION
Philadelphia, Pa.), which prevented freezing of the joints. The moisture trap was surrounded by a Dewar flask containing dry ice and acetone. The mercury trap, also surrounded by a Dewar flask containing dry ice and acetone, was included in the apparatus to prevent mercury from the diffusion pump entering the moisture trap and the drying chamber. In order to break the vacuum and to admit dry air into the system a drying tube containing anhydrous calcium sulfate was attached to a stopcock in the system. The McLeod gauge was calibrated and used to measure the pressure in the system. After making the pressure readings, the column of mercury in the gauge was pulled down by means of an aspirator.
EXPERIMENTAL The material used for the investigation was the leaves of Atropa belladonna L. Opposite leaves were gathered whenever possible to obtain leaves of the same age, size, metabolic processes, and which have been under the same environmental conditions. They were divided into two analogous samples to be oven dried and lyophilized; analyses could be made and the results compared on these samples. Sample IA was dried at 65" for twelve hours. It had a dark olive green color. Sample 1B was dried at room temperature under reduced pressure. It had essentially the same color as 1A. Sample 2A was dried at 45-50' for thirty-six hours and it also had a dark olive green color. Sample 2B was lyophilized. It was very fluffy and occupied a volume more than twice as large as its corresponding oven-dried sample. It had a bright green color. Samples 3A and 3B were dried in the same manner as-2A and 2B, respectively, and appeared the same in every respect.
-0
Fig. 1.-Lyophilizing
apparatus in operation.
The mercury diffusion pump was included in the system to insure obtaining a pressure low enough to effect lyophilization. A Cenco-Megavac pump was used as a forepump. The arms of the mercury pump were uniformly heated to prevent condensation by wrapping them with asbestos and nichrome wire as a resistor. The water intake and outlet are for the condenser in the pump.
Fig. 2.-To scale drawing of apparatus in Fig. 1. After drying, the samples were reduced to a No. The average results in per cent on a dry-weight basis are given in Table I for the various determinations. The moisture was determined by the U. S. P. XI11 method for drugs containing no constituents volatile at 100". It can be seen that lyophilization reduces the moisture content as low as oven drying at 50' for thirty-six hours. Ash determinations were made according to the U. S. P. XI11 method. The drug is well within 40 powder and the various analyses made.
SCIENTIFIC EDITION
325
TABLE I.-SUMMAR~ Moisture Total ash Acid-insoluble ash Total alkaloids Total alcoholic extractive Total nitrogen in extract calculated as protein Total nitrogen in residue calculated as protein Amino nitrogen Ammonia and amide nitrogen Ammonia nitrogen Amide nitrogen Starch Sugars Total chlorophyll Chlorophyll a Chlorophyll b % Chlorophyll a of total chlorophyll % Chlorophyll b of total chlorophyll Animal tests Q
'
1A
1B
2A
2B
5.51 15.84 1.46
6.24 15.59 1.69
22:33 0.62 4.50
22129 0.53 4.28
...
.. . ... ...
3.12 14.59 0.90 0.328 23.00 0.53 4.53 0.41
3.42 15.43 1.34 0.338 22.97 0.59 4.52 0.35 0.027 0.006 0.021 4.40 3.44 4.33 2.71 1.61 62.6 37.4
...
... ... ...
...
2.60 1.55 0.98 59.9 37.5
...
... ... ...
2.50 1.62 0.99 64.8 37.3
...
0.054 0.010 0.044 4.81 2.25 3.30 2.09 1.21 63.3 36.7
...
...
3A
3B
3.88 4.40 15.81 15.45 1.08 1.06 0.285 0.307 23.29 25.79 0.79 0.90 4.18 4.51 0.35 0.26 0.049 0.015 0.018 0.001 0.031 0.014 5.81 4.85 2.31 4.31 2.45 4.30 1.50 2.64 0.95 1.66 61.4 61.3 38.9 38.7 No difference
Average results of multiple determinations.
U. S. P. XI11 requirements for acid-insoluble ash. The leaves were assayed for alkaloids by the U. S. P. XI11 method. The data presented here do not indicate any difference, or at least no significant difference, between the lyophilized and oven-dried products. It does not appear, under the conditions of drying here, that the total alkaloidal content is affected by the temperature. It may be that oven drying was carried out so rapidly that no detectable change had taken place. Total alcoholic extractive was determined to see if heat had any coagulating or other effect. AIso because of the greater bulk, it was thought the lyophilized drug would allow greater penetrability resulting in more extractives and quicker complete extraction. The results showed no difference between the two products. Aliquot portions of the alcoholic extracts were evaporated and total nitrogen, amino nitrogen, and amide nitrogen determined for the purpose of detecting if any proteolysis had taken place. Total nitrogen in the extracts and residues were found to be the same, as well as the amino nitrogen; however, a difference was found in the amide nitrogen. The oven-dried products contained about twice as much as the lyophilized products. This indicates that some proteolysis had occurred during oven drying. For determining sugar and starch the samples were extracted in a Soxhlet extractor for twenty-four hours with 95% alcohol. The extract was then made up t o a 500-cc. volume with 95% alcohol. The residue after extraction was dried a t room temperature. The residue was placed in a beaker with 50 cc. of water and heated on a steam plate for fifteen minutes. After cooling, 20 cc. of a freshly prepared 0.5% solution of pancreatin was added. The mixture was digested in a n oven a t 55' for one hour and again placed on a steam plate for fifteen minutes. After cooling, the mixture was digested with an additional 20-cc. portion of the pancreatin solution at 55" for one hour. It was then filtered and washed. The filtrate was diluted to 200 cc. and hydrolyzed with 20 cc. of 36% hydrochloric acid in a 600-1~. Erlenmeyer flask on a steam plate for two and one-half hours.
The hydrolyzed extract was partially neutralized with 15 cc. of 30% sodium hydroxide, completely neutralized with solid anhydrous sodium carbonate, and then filtered, washed thoroughly, and diluted to 500 cc. Aliquot portions were analyzed according to the A. 0. A. C. methods of analysis (10). Duplicate results on the same samples treated in the same manner could not be obtained. It was believed the extractions were incomplete. New samples were run in exactly the same manner. The residue, after incubating with pancreatin and washing, was dried; the filter paper and residue were put into a Waring blendor with a small amount of water and broken up. It was then washed into a beaker and treated in exactly the same manner as previously. The second time each sample yielded from one-half t o the same amount as it yielded the first time. The combined results are given in Table I. The previous assumption that the extractions were incomplete was correct. It is possible that some of the cells never came into contact with the extracting medium. Since the material was dried, the tissues would become filled with air. In multicellular particles, it is possible that air would become trapped in the internal cells and thus prevent the extracting medium from entering those cells. This would prevent complete extraction. The alcoholic extracts were treated and analyzed according t o A. 0. A. C. methods (11). The carbohydrates were determined according to the Munson and Walker method and calculated as dextrose. There is a noticeable difference between the samples as a result of the method of drying. The lyophilized samples have a higher sugar content than the oven-dried products. A possible explanation for the difference was through the hydrolysis of starch, but a n examination of the results from the starch determinations will immediately eliminate this as a source of the difference. Of the other processes going on in a plant the one most likely t o account for the difference in the amount of sugar is respiration. It has been found that sugars are utilized in enzymatic and respiratory processes. In the sample in which these processes were inhibited, the sugar content was greater than in the sample in
326
JOURNAL OF THE
AMERICAN PHARMACEUTICAL ASSOCIATION
which these processes were not inhibited. Lyophilization then preserves the carbohydrates as compared’to oven drying a t 45-50”. Chlorophyll was determined according to A. 0. A. C. methods (12). I t can be seen in the table that the lyophilized product has a greater amount of total chlorophyll and that chlorophyll a and chlorophyll b occur in the same ratio in the samples which were dried by different methods. It appears that lyophilization prevents some process which has a destructive effect on chlorophyll. An aqueous extract of both samples was made, saturated with ammonium sulfate, and allowed to set for twenty-four hours. The lyophilized sample had considerably more flocculent precipitate than the oven-dried. The precipitate was filtered, washed into a dialyzing bag, and dialyzed for seventy-two hours with frequent changes of water. The dialyzate was then incubated with a known amount of hyoscyamine sulfate for forty-eight hours and assayed. There was found to be no change in the alkaloidal content, which would seem to indicate the absence of any enzyme having a destructive effect on the alkaloid; however, more tests are necessary before definite conclusions can be made. Portions of the alcoholic extracts which contained the same amount of alkaloid as determined by chemical assay were evaporated. Water was added, filtered, and the filtrate diluted until it contained 1 part alkaloid to 5000 parts water. These extracts were then tested on the intact eyes of rabbits which had been previously standardized for maximum dilatation with a known solution of atropine. I t was found that the action produced by extracts of the lyophilized sample and the oven-dried sample were parallel. This indicates that the proportion of 1hyoscyamine to atropine is not altered by the method of drying or is altered t o the same degree. CONCLUSIONS
From the data obtained in this investigation, the following conclusions were drawn: 1. Lyophilization is applicable to the leaves of belladonna as a method of preservation and it is suggested that this method of drying be investigated on other leaves and on other plant organs. 2. Lyophilization reduces the moisture content as low as oven drying at 50’ for 36 hours.
3. With respect to the preservation of the alkaloids of belladonna leaf, lyophilization is as effective as oven drying at 50 ’. 4. Vacuum drying does not facilitate extraction. 5. Oven drying causes more proteolysis than lyophilization. 6. No difference in starch content was found, but extraction of the starch was found to be incomplete and it is suggested that more work be done on extraction problems especially in respect to powder size. 7. Lyophilization preserves the sugars as compared to oven drying a t 50’. It is recommended that a similar investigation be carried out on glycosidal- and tannin-bearing drug plants. 8. Oven drying has a destructive effect on chlorophyll while lyophilization preserves this pigment. 9. No evidence was presented in the preliminary tests conducted that the proportion of I-hyoscyamine to atropine is different in the two methods of drying. 10. It appears that lyophilization preserves certain constituents better than oven drying at 50 ’; therefore, it is recommended that lyophilization be adopted as a standard method of drying material which is to be used in any original investigation of these constituents. REFERENCES (1) Sizer, I. W., and Josephson, E. S., Food Research, 8, 200(1942). (2) Balls,Arnold K.,and Lineweaver, Hans, ibid.. 3, 57 (1938). (3) Koch, G. P.,THISJOURNAL, 8.390(1919). (4) Fliick, H.,Pharm. J . , 159,375(1947). (5) Rowson, J. M.,Quart. J . Pharm. Pharntatol., 17. 234(1944). (6) Todd, J. P., J . Roy. Tech. Coll. (Clasgow), 2, 353 (1930). (7) Flosdorf, Earl W., J . Chcm. Educafion, 22,470(1945). (8) Flosdorf, E. -W., Hull, I,. W., and Mudd, S., J . Immunol., 50, 21(104a). (9) Elser, W. J., Thomas, R. A., and Steffen, G . I., ibid., 28. 433(1935). (10) A. 0. A. C., “Methods of Analysis,” 1945,p. 571. (11) Ibid., p. 132. (12) I b i d . , p. 142.
reported that solanaceous leaves yield 2 per cent to 10 per cent more tropine alkaloids if the leaves are stabilized (4). It was further found that the temperature of drying had a significant influence on the degree of racemization (5). Fermentation of belladonna leaves (6) was carried out at 32" while preventing the loss of water, and upon an alkaloidal determination of the leaf, it was found to contain less alkaloid than the control. It was concluded that enzymes were the causative COMPARATIVE study of the effects produced by lyophilizing and oven drying belladonna agents for the loss in alkaloidal content. Thus leaves was made. The object was to determine if it can be seen that the temperature of drying the newer method of desiccation, lyophilization, affects the alkaloidal content. A method of drying without the aid of heat is is applicable to belladonna leaves, and to determine what changes, if any, other than desiccation lyophilization. With this method the moisture in the material is frozen and removed by means of are produced by drying at 50". The basis for the activities of plant cells is to be reduced pressure directly from the frozen state found in their enzymes. The rate at which an to the vapor state. It is sublimation of the enzymatic reaction proceeds is influenced not ice. This process has been used successfully with only by the temperature but also by the length the biologicals (7). It was possible to keep conof time the reaction mixture has already been valescent human sera for two or three years at that temperature. Within certain limits, an increase in the temperature increases the while retaining its full potency (8). Antisera and velocity of enzymatic reactions. If plant ma- antigens as well as certain related products, when terial is being dried within these limits, these dried in the frozen state and stored in vacuo, enzymatic reactions will continue for a period of suffered no deterioration in numerical values and time, thus altering the quantity of the constit- the dried products when reclaimed after a number of years showed no loss in potency. uents contained therein. Bacteria and yeasts have been preserved unThe effect of low temperatures (1) upon enchanged by lyophilization, and ip no instance zymic systems has been studied and it was found that the rate was retarded and there was no in- have the strains shown any changes even in activation of the enzymes which were studied. minor particulars (9). The apparatus that was used in this laboratory Attempts have been made to demonstrate hydrolis shown in the photograph while in operation and ysis in the solid state ( 2 ) . The rate of hydrolysis was found to be not more than 2 per cent in the accompanying diagram which was drawn of that occurringin supercooled liquid at the same to scale. This apparatus has proved to be very satisfactory. Pressures as low as 0.006 mm. of temperature. mercury were obtained which kept the material Upon application of these results to the desiccation of plant materials similar effects caused frozen while being dried. The drying chamber and moisture trap were by temperature would be expected. The effect 4l/1 and 4 inches in diameter, respectively, with of the temperature used for the drying of bella71/60 ground glass joints. Tubes of this size donna leaves has been noted by Koch (3). It is made it possible to lyophilize several hundred * Received October 29, 1949, from the College of Phar- grams of fresh material at one time. The mamacy Ohio State University Columbus Ohio t 6bstract of a thesis subkitted to th;?Graduate School in terial to be lyophilized was shell-frozen on the partial fulfillment of the requirements for the degree of sides of the drying chamber so as to expose a Doctor of Philosophy. X Fellow of the American Foundation for Pharmaceutical maximum of surface area to facilitate drying. Education: present address: College of Pharmacy, Universitv of Texas. Austin. The ground glass joints were sealed with Apiezon, $, Akociate Piofessor; College af Pharmacy, Ohio State a wax (obtained from James G. Biddle Co., University.
A
323
324
JOURNAL OF THE
AMERICAN PHARMACEUTICAL ASSOCIATION
Philadelphia, Pa.), which prevented freezing of the joints. The moisture trap was surrounded by a Dewar flask containing dry ice and acetone. The mercury trap, also surrounded by a Dewar flask containing dry ice and acetone, was included in the apparatus to prevent mercury from the diffusion pump entering the moisture trap and the drying chamber. In order to break the vacuum and to admit dry air into the system a drying tube containing anhydrous calcium sulfate was attached to a stopcock in the system. The McLeod gauge was calibrated and used to measure the pressure in the system. After making the pressure readings, the column of mercury in the gauge was pulled down by means of an aspirator.
EXPERIMENTAL The material used for the investigation was the leaves of Atropa belladonna L. Opposite leaves were gathered whenever possible to obtain leaves of the same age, size, metabolic processes, and which have been under the same environmental conditions. They were divided into two analogous samples to be oven dried and lyophilized; analyses could be made and the results compared on these samples. Sample IA was dried at 65" for twelve hours. It had a dark olive green color. Sample 1B was dried at room temperature under reduced pressure. It had essentially the same color as 1A. Sample 2A was dried at 45-50' for thirty-six hours and it also had a dark olive green color. Sample 2B was lyophilized. It was very fluffy and occupied a volume more than twice as large as its corresponding oven-dried sample. It had a bright green color. Samples 3A and 3B were dried in the same manner as-2A and 2B, respectively, and appeared the same in every respect.
-0
Fig. 1.-Lyophilizing
apparatus in operation.
The mercury diffusion pump was included in the system to insure obtaining a pressure low enough to effect lyophilization. A Cenco-Megavac pump was used as a forepump. The arms of the mercury pump were uniformly heated to prevent condensation by wrapping them with asbestos and nichrome wire as a resistor. The water intake and outlet are for the condenser in the pump.
Fig. 2.-To scale drawing of apparatus in Fig. 1. After drying, the samples were reduced to a No. The average results in per cent on a dry-weight basis are given in Table I for the various determinations. The moisture was determined by the U. S. P. XI11 method for drugs containing no constituents volatile at 100". It can be seen that lyophilization reduces the moisture content as low as oven drying at 50' for thirty-six hours. Ash determinations were made according to the U. S. P. XI11 method. The drug is well within 40 powder and the various analyses made.
SCIENTIFIC EDITION
325
TABLE I.-SUMMAR~ Moisture Total ash Acid-insoluble ash Total alkaloids Total alcoholic extractive Total nitrogen in extract calculated as protein Total nitrogen in residue calculated as protein Amino nitrogen Ammonia and amide nitrogen Ammonia nitrogen Amide nitrogen Starch Sugars Total chlorophyll Chlorophyll a Chlorophyll b % Chlorophyll a of total chlorophyll % Chlorophyll b of total chlorophyll Animal tests Q
'
1A
1B
2A
2B
5.51 15.84 1.46
6.24 15.59 1.69
22:33 0.62 4.50
22129 0.53 4.28
...
.. . ... ...
3.12 14.59 0.90 0.328 23.00 0.53 4.53 0.41
3.42 15.43 1.34 0.338 22.97 0.59 4.52 0.35 0.027 0.006 0.021 4.40 3.44 4.33 2.71 1.61 62.6 37.4
...
... ... ...
...
2.60 1.55 0.98 59.9 37.5
...
... ... ...
2.50 1.62 0.99 64.8 37.3
...
0.054 0.010 0.044 4.81 2.25 3.30 2.09 1.21 63.3 36.7
...
...
3A
3B
3.88 4.40 15.81 15.45 1.08 1.06 0.285 0.307 23.29 25.79 0.79 0.90 4.18 4.51 0.35 0.26 0.049 0.015 0.018 0.001 0.031 0.014 5.81 4.85 2.31 4.31 2.45 4.30 1.50 2.64 0.95 1.66 61.4 61.3 38.9 38.7 No difference
Average results of multiple determinations.
U. S. P. XI11 requirements for acid-insoluble ash. The leaves were assayed for alkaloids by the U. S. P. XI11 method. The data presented here do not indicate any difference, or at least no significant difference, between the lyophilized and oven-dried products. It does not appear, under the conditions of drying here, that the total alkaloidal content is affected by the temperature. It may be that oven drying was carried out so rapidly that no detectable change had taken place. Total alcoholic extractive was determined to see if heat had any coagulating or other effect. AIso because of the greater bulk, it was thought the lyophilized drug would allow greater penetrability resulting in more extractives and quicker complete extraction. The results showed no difference between the two products. Aliquot portions of the alcoholic extracts were evaporated and total nitrogen, amino nitrogen, and amide nitrogen determined for the purpose of detecting if any proteolysis had taken place. Total nitrogen in the extracts and residues were found to be the same, as well as the amino nitrogen; however, a difference was found in the amide nitrogen. The oven-dried products contained about twice as much as the lyophilized products. This indicates that some proteolysis had occurred during oven drying. For determining sugar and starch the samples were extracted in a Soxhlet extractor for twenty-four hours with 95% alcohol. The extract was then made up t o a 500-cc. volume with 95% alcohol. The residue after extraction was dried a t room temperature. The residue was placed in a beaker with 50 cc. of water and heated on a steam plate for fifteen minutes. After cooling, 20 cc. of a freshly prepared 0.5% solution of pancreatin was added. The mixture was digested in a n oven a t 55' for one hour and again placed on a steam plate for fifteen minutes. After cooling, the mixture was digested with an additional 20-cc. portion of the pancreatin solution at 55" for one hour. It was then filtered and washed. The filtrate was diluted to 200 cc. and hydrolyzed with 20 cc. of 36% hydrochloric acid in a 600-1~. Erlenmeyer flask on a steam plate for two and one-half hours.
The hydrolyzed extract was partially neutralized with 15 cc. of 30% sodium hydroxide, completely neutralized with solid anhydrous sodium carbonate, and then filtered, washed thoroughly, and diluted to 500 cc. Aliquot portions were analyzed according to the A. 0. A. C. methods of analysis (10). Duplicate results on the same samples treated in the same manner could not be obtained. It was believed the extractions were incomplete. New samples were run in exactly the same manner. The residue, after incubating with pancreatin and washing, was dried; the filter paper and residue were put into a Waring blendor with a small amount of water and broken up. It was then washed into a beaker and treated in exactly the same manner as previously. The second time each sample yielded from one-half t o the same amount as it yielded the first time. The combined results are given in Table I. The previous assumption that the extractions were incomplete was correct. It is possible that some of the cells never came into contact with the extracting medium. Since the material was dried, the tissues would become filled with air. In multicellular particles, it is possible that air would become trapped in the internal cells and thus prevent the extracting medium from entering those cells. This would prevent complete extraction. The alcoholic extracts were treated and analyzed according t o A. 0. A. C. methods (11). The carbohydrates were determined according to the Munson and Walker method and calculated as dextrose. There is a noticeable difference between the samples as a result of the method of drying. The lyophilized samples have a higher sugar content than the oven-dried products. A possible explanation for the difference was through the hydrolysis of starch, but a n examination of the results from the starch determinations will immediately eliminate this as a source of the difference. Of the other processes going on in a plant the one most likely t o account for the difference in the amount of sugar is respiration. It has been found that sugars are utilized in enzymatic and respiratory processes. In the sample in which these processes were inhibited, the sugar content was greater than in the sample in
326
JOURNAL OF THE
AMERICAN PHARMACEUTICAL ASSOCIATION
which these processes were not inhibited. Lyophilization then preserves the carbohydrates as compared’to oven drying a t 45-50”. Chlorophyll was determined according to A. 0. A. C. methods (12). I t can be seen in the table that the lyophilized product has a greater amount of total chlorophyll and that chlorophyll a and chlorophyll b occur in the same ratio in the samples which were dried by different methods. It appears that lyophilization prevents some process which has a destructive effect on chlorophyll. An aqueous extract of both samples was made, saturated with ammonium sulfate, and allowed to set for twenty-four hours. The lyophilized sample had considerably more flocculent precipitate than the oven-dried. The precipitate was filtered, washed into a dialyzing bag, and dialyzed for seventy-two hours with frequent changes of water. The dialyzate was then incubated with a known amount of hyoscyamine sulfate for forty-eight hours and assayed. There was found to be no change in the alkaloidal content, which would seem to indicate the absence of any enzyme having a destructive effect on the alkaloid; however, more tests are necessary before definite conclusions can be made. Portions of the alcoholic extracts which contained the same amount of alkaloid as determined by chemical assay were evaporated. Water was added, filtered, and the filtrate diluted until it contained 1 part alkaloid to 5000 parts water. These extracts were then tested on the intact eyes of rabbits which had been previously standardized for maximum dilatation with a known solution of atropine. I t was found that the action produced by extracts of the lyophilized sample and the oven-dried sample were parallel. This indicates that the proportion of 1hyoscyamine to atropine is not altered by the method of drying or is altered t o the same degree. CONCLUSIONS
From the data obtained in this investigation, the following conclusions were drawn: 1. Lyophilization is applicable to the leaves of belladonna as a method of preservation and it is suggested that this method of drying be investigated on other leaves and on other plant organs. 2. Lyophilization reduces the moisture content as low as oven drying at 50’ for 36 hours.
3. With respect to the preservation of the alkaloids of belladonna leaf, lyophilization is as effective as oven drying at 50 ’. 4. Vacuum drying does not facilitate extraction. 5. Oven drying causes more proteolysis than lyophilization. 6. No difference in starch content was found, but extraction of the starch was found to be incomplete and it is suggested that more work be done on extraction problems especially in respect to powder size. 7. Lyophilization preserves the sugars as compared to oven drying a t 50’. It is recommended that a similar investigation be carried out on glycosidal- and tannin-bearing drug plants. 8. Oven drying has a destructive effect on chlorophyll while lyophilization preserves this pigment. 9. No evidence was presented in the preliminary tests conducted that the proportion of I-hyoscyamine to atropine is different in the two methods of drying. 10. It appears that lyophilization preserves certain constituents better than oven drying at 50 ’; therefore, it is recommended that lyophilization be adopted as a standard method of drying material which is to be used in any original investigation of these constituents. REFERENCES (1) Sizer, I. W., and Josephson, E. S., Food Research, 8, 200(1942). (2) Balls,Arnold K.,and Lineweaver, Hans, ibid.. 3, 57 (1938). (3) Koch, G. P.,THISJOURNAL, 8.390(1919). (4) Fliick, H.,Pharm. J . , 159,375(1947). (5) Rowson, J. M.,Quart. J . Pharm. Pharntatol., 17. 234(1944). (6) Todd, J. P., J . Roy. Tech. Coll. (Clasgow), 2, 353 (1930). (7) Flosdorf, Earl W., J . Chcm. Educafion, 22,470(1945). (8) Flosdorf, E. -W., Hull, I,. W., and Mudd, S., J . Immunol., 50, 21(104a). (9) Elser, W. J., Thomas, R. A., and Steffen, G . I., ibid., 28. 433(1935). (10) A. 0. A. C., “Methods of Analysis,” 1945,p. 571. (11) Ibid., p. 132. (12) I b i d . , p. 142.