STUDIES ON THE HEMORRHAGIC SWEET CLOVER DISEASE

STUDIES ON THE HEMORRHAGIC SWEET CLOVER DISEASE I. THE PREPARATION OF HEMORRHAGIC CONCENTRATES* BY HAROLD A. CAMPBELL, WILLARD L. ROBERTS, WILLIAM K. SMITH, AND KARL PAUL LINK (From the Deparlments of Biochemistry and Genetics, Wisconsin Agricultural Experiment Station, University of Wisconsin, Madison) (Received for publication, July 5, 940) The disease of cattle, known in veterinary practice as sweet clover disease, arises from the eating of improperly cured hay or silage made from the common sweet clovers Melilotus alba and Melilotus oficinalis. Its occurrence was originally observed practically simultaneously in 92 by Schofield in Canada and Roderick in this country. Schofield (, 2) published his observations in 922 and 924. Subsequently Roderick (3, 4) and Roderick and Schalk (5) recorded the etiology and pathology of the disease in a comprehensive and thorough study made at the North Dakota Agricultural Experiment Station during the interval of 92-32. This disease is in a sense without parallel in animal pathology or human medicine (6-8). When cows, sheep, or rabbits are fed spoiled sweet clover hay made from any of the common Melilotus varieties, the disease is manifested by a progressive diminution in t.he clotting power of the blood and resultant hemorrhages which usually become fatal. It was recognized by Schofield (2) that the disease could be controlled in cattle by the withdrawal of the spoiled hay from the diet and by the injection of blood * Published with the approval of the Director of the Wisconsin Agricultural Experiment Station. Cooperative studies with the Division of Forage Crops and Diseases, United States Department of Agriculture, and the Wisconsin Agricultural Experiment Station, paper No. 264, Genetics Department; personnel and supply assistance since July, 938, in part through the Natural Science Research Project, 65-l-53-2349 of the Federal Works Progress Administration (Madison). 47

Hemorrhagic Sweet Clover Discasc. I strum freshly drawn from normal cattle, provided the hemorrhagic cxtrarasation had not proceeded too far. Roderick (4, 5) observed that hemorrhage may occur in almost any part of the body, but. most frequently in the subcutaneous and intermuscular fasciae. No visible alteration was found in the blood vessels to explain the internal hemorrhagic diathesis, but external hemorrhage may be readily induced by surgical treatment or accidental injuries. The causative agent in the spoiled hay apparently survives the digestive processes, enters the blood stream, and by some mechanism at present unknown, either prevents the formation of prothrombin in the liver, or effects an inactivation of formed prothrombin. Recently Quick (9) confirmed the fundamental observation of Roderick that the delay in t.hc coagulation t.ime of the blood is due t.o a disturbance of the prothrombin, which is generally recognized as one of the essential components of the blood coagulation mechanism (6-8, 0, ). Prior to the biochemical studies begun at this Station in 934, some efforts had been made by others to extract the causative agent from hay known to induce the hemorrhagic condition characteristic of the disease. Schofield (2) attempted to prepare concentrates by extraction with water. The resulting aqueous extracts were fed to six rabbits. Three of t,hcm remained normal. In the other three a delay in the clotting time of the blood was observed and in addition it appeared that some destruction of the red blood cells was induced by the aqueous concentrate. Roderick and Schalk (5) extracted spoiled hay with wabcr and fed the extract daily to a group of rabbits for approximately a month. The clotting power of the blood of these animals was not altered. In another attempt an aqueous extract of spoiled hay was concentrated under reduced pressure (50-55 ). The concentrate was fed to rabbits but no pathological response was noted. When the undissolved residue was fed to rabbits, the usual symptoms of the disease resulted, which indicated clearly that the causative agent, as it occurred in the hay, was not watersoluble. Roderick and Schalk (5) also attempted to extract the causative agent with 95 per cent ethanol, ethyl ether, and per cent acetic acid. None of these attempts was successful. Subsequently Brown, Savage, and Robinson (2) prepared aqueous extracts of

Campbell, Roberts, Smith, and Link 49 the spoiled hay which were fed to rabbits without any deleterious effects. The extensive experimentation done by Roderick and Schalk with aqueous extracts of spoiled hay and the findings of Brown et al. cast some doubt on Schofield s work, which stands alone in its claim that the causative agent, as it occurs in spoiled hay, can be extracted with water. All other attempts to extract the hemorrhagic agent which had come to our attention when the chemical investigations were initiated at this Station were unsuccessful. Two fundamental issues confronted us when this study was begun. There were no chemical criteria available to establish the presence of the hemorrhagic agent, and a bioassay offered the only available means of testing the physiological activity of preparations and extracts. In the absence of a chemical test complete reliance had to be placed on a determination of the prothrombin level or activity in rabbit blood or plasma. We soon recognized that extraction of the hemorrhagic agent in a physiologically active form through a simple and direct operation was not feasible. When the clotting time of whole blood was used for the biological assay, it was found that the hemorrhagic agent was insoluble in petroleum ether, ethyl ether, ethanol, acetone, dioxane, water, and aqueous acid. On the other hand, dilute sodium hydroxide effected the solution of the hemorrhagic principle. In 936-37 one of us developed a scheme for the preparation of active concentrates (3). With the prothrombin met hod of Quick (4) for the bioassay, t,his extraction scheme, which will not bc given in detail, yielded concentrates with much higher activity than the original hay. A bluish green ether solution was obtained which upon removal of the ether gave a green residue weighing 0.52 gm. This concentrat,e showed a physiological activity cquivalcnt to about 75 gm. of the original hay and thus represents a concentration of about 50-fold. Treatment of the blue-green ether solution with the activated carbon known as Nuchar, after addition of 3 volumes of petroleum ether, removed 0.4 gm. of solid material from the solution. This treatment left a colorless inactive solution. Attempts to elute the physiologically active material wcrc unsuccessful. But when the Private communications to -C.P.L. from various investigators who abandoned the work without publishing their findings.

50 Hemorrhagic Sweet Clover Disease. I carbon was fed to a rabbit, a reduction in prothrombin activity was noted equivalent to that obtained after 50 gm. of the original spoiled sweet clover hay were fed. This indicated that at least partial elution of the active principle from the carbon was effected by the digestive system of the rabbit. After repeated trials we have succeeded in separating the hemorrhagic substance quantitatively from the spoiled hay. We are now able to produce consistently an active concentrate with a potency approximately 200 times greater than the original spoiled hay.2 This concentrate is essentially free from fats, waxes, certain pigments, sugars, glycosides, water-soluble polysaccharides, watersoluble acids, amines, alkaloids, water-soluble proteins, and watersoluble decomposition products of chlorophyll. At this stage of the investigation the chemical nature of the hemorrhagic agent is still unknown. Consequently each step in the fractionation scheme has been developed and controlled separately through prothrombin assays on standardized susceptible rabbits, selected specifically for this study (5).2 The act,ivity of each fraction was related to the original spoiled hay rather than to the total solids of the individual active fractions. By evaluating each fraction by the prothrombin assay it became possible to include in the scheme only those operations that effect essentially a quantitative separation of the physiologically active agent. EXPERIMENTAL The spoiled sweet clover hay used in this work was produced experimentally3 from Melilotus alba by the procedure developed here by Smith and Brink (6). In its color and tobaccolike suffocating odor, the artificially spoiled hay corresponded to the spoiled hays that are encountered in agricultural practice and, as Roderick showed, are usually fatal to cattle ((5) p. 39). The hay was milled so that it would pass through a screen having openings about 0.5 mm. in diam&er. * The method employed to assay potent spoiled sweet clover hays and physiologically active concentrates prepared therefrom by evaluating their capacity to reduce the prothrombin level in rabbit plasma will be given in detail in a subsequent paper. 3 The spoiled hay used by Quick (9) was obtained in October, 936, from this experimentally produced stock.

Campbell, Roberts, Smith, and Link 5 Extraction Procedure (See Diagram) Step I-The milled hay (3 kilos) is extracted with Skellysolve,4 (pentane) for 24 hours in a large Soxhlet extractor (Barnstead Still and Sterilizer Company). The extract containing waxes, fats, and other lipoidal material is inactive. The solvent is allowed to evaporate from the extracted hay at room temperatures. Step 2-The extracted hay* is placed in 30 liters of water to which 260 cc. of concentrated HCl have been added. During the steeping for 48 hours at 25-30, the suspension is shaken occasionally and if necessary concentrated HCl is added to maintain the ph in the range of to 2. After 48 hours the steepwaters are removed by filtration. They are inactive and contain materials soluble in dilute aqueous acid (carbohydrates, acids, proteins, alkaloids, etc.). Step S-The wet filter cake from Step 2 is placed in 30 liters of water containing 20 gm. of NaOH and steeped for 48 hours at 25-30. The mash is shaken occasionally and if necessary concentrated NaOH is added to maintain a ph of to 2. This treatment effects an extensive swelling and disintegration of the sweet clover tissue. The pectins, hemicelluloses, gums, and waterinsoluble acids dissolve in the form of sodium salts. If the liquid is separated from the fiber at this point by filtering or centrifuging, the hemorrhagic fraction is found in the alkaline solution. The fiber is inactive. However, the dispersed pectins, gums, and hemicelluloses make the separation of the fiber and insoluble material very difficult and laborious. In practice this laborious and time-consuming separation is avoided through the technique embodied in Step 4. Step b--the alkaline solution (without separation from the solid fibrous material) is acidified by adding concentrated HCI until the solution is at ph 3.0. This precipitates the pectins, hemicelluloses, gums, and water-insoluble acids on the suspended fibers which then act as a carrier and greatly facilitate filtration. The acidified suspension should stand several hours to allow the acid to diffuse into the fibrous material before the filtration is made. The filtrate is inactive, while the precipitated material contains the total physiological activity of the spoiled sweet clover hay. 4 0 gallon glass-lined vessels are used in Steps 2 and 3.

52 Hemorrhagic Sweet Clover Disease. I Step 5-The precipitated material collected on the filter in Step 4 is leached with 20 liters of et,hanol (90 to 95 per cent) for 48 hours at 25-30. The ethanolic extract is removed by filtration. This leaching process is repeated three times and the four Extract (discard) Steps in Concentration of Hemorrhagic Agent Spoiled sweef clover hay. Extract with n-pentane I (Residue) 2. St;yg,ln water 48 hrs., ph -2; I Filtrati (discard) 3. Steep in water 48 hrs., p$ -2 4. Adjust to ph 3.0; filter Filtrate (discard) (Residue) 5. Extract with 95% ethano:; filter Resid& (discard) (Filtrate) 6. Remove EtOH at 30 7. Dissolve residue in 0.50/o NaOH; adjust ph to 3.0; filter I Filtrate (discard) dt.) 8. Dissolve in methanol; ppt. with ethyl ether; filter Ppt. (discard) Water layer and insoluble (discard) ppt. (Filtrhte) 9. Remove solvent; dissolve residue in 0.5% NaOH.; adjust ztfjie~ 3.0; extract with ethyl I I Ether solution Sctivity 200 X original hay (minimum) ethanolic extracts are combined. They contain the acids, proteins, chlorophyll degradation products, and some lipoidal material soluble in ethanol. The ethanolic extract is active, while the insoluble fibrous material is inactive. Step &-The ethanolic extracts are concentrated in a 5 gallon Pfaudler glass-lined vacuum still under reduced pressure at 25-30.

Campbell, Roberts, Smith, and Link 53 Step r-the residue from Step 6 is removed from the still by dissolving it in 0.5 per cent NaOH. The alkaline solution is acidified by adding HCl until ph 3 is reached to precipitate the active material. The flocculent precipitate formed settles on standing for about 2 hours. The clear supernatant liquid is filtered through an asbestos mat on a Buchner funnel. The brownish green precipitate is finally transferred to the filter and sucked dry. The inactive filtrate is discarded. Step g--the active brownish green precipitate obtained in Step 7 is removed from the filter, and dissolved in liter of methanol. To this solution arc added 3 volumes of ether, which gives a large flocculent precipitate. This precipitate is removed by filtration through a fluted filter paper. The precipitate is dissolved in methanol and again precipitated by adding ether. This treatment gives a large gummy precipitate which is inactive, and a brownish green active solution. Step g-the physiologically active solution from Step 8 is concentrated under reduced pressure at 25-30 to remove volatile solvents. The residue is dissolved in about a liter of 0.5 per cent NaOH and is then precipitated in a finely divided condition by acidifying with HCl. The solution should be definitely acid to Congo red (about ph 3). The suspension is shaken with an equal amount of ethyl ether. After the ether layer is separated from the insoluble material and aqueous layer, the solid material is dissolved by adding NaOH, again precipitated, and shaken with ether. The ether layers, which have a high activity, are combined. The insoluble material suspended in the aqueous layer retains some activity. By repeating the extraction procedure of Steps 8 and 9 on this residue, the remaining activity is almost completely obtained in the ether layer. The ether solution has a blue-green color and contains the ethersoluble, water-insoluble, acidic compounds such as chlorophyll degradation products, fatty acids, phenolic compounds, organic acids, etc. After the ether is removed under reduced pressure, a concentrate (5 to 8 gm.) is obtained which has constant physiological activity. When 0.60 gm. of this concentrate was fed to st.andardized susceptible rabbits2 the plasma prothrombin level was reduced to 0 per cent of the normal in 40 to 48 hours. This

54 Hemorrhagic Sweet Clover Disease. I represents a 200-fold concentration (minimum) of the hemorrhagic principle. SUMMARY. An extraction and fractionation scheme has been developed for the concentration of the hemorrhagic substance in spoiled sweet clover hay (Melilotus alba). 2. When 0.60 gm. of this concentrate is fed to standardized susceptible rabbits, the plasma prothrombin is reduced to 0 per cent of the normal in 40 to 48 hours. This is equivalent to a 200-fold concentration (minimum) of the hemorrhagic agent. 3. The chemical nature of the hemorrhagic agent present in the concentrates is not known at this stage. However, it can be stated that the concentrate is essentially free from the following classes of substances: fats, waxes, certain pigments, sugars, glycosides, water-soluble polysaccharides, water-soluble acids, amines, alkaloids, water-soluble proteins, and water-soluble decomposition products of chlorophyll. This series of investigations on the hemorrhagic sweet clover disease was undertaken at the suggestion of Dr. R. A. Brink of the Genetics Department, who in 933 initiated the first trials at this Station to select a non-bitter strain of sweet clover. We are indebted to Dr. Brink and to Dr. E. A. Hollowell of the Office of Forage Crops and Diseases, United States Department of Agriculture, Washington, for their generous counsel and financial assistance throughout the course of this study. BIBLIOGRAPHY. Schofield, F. S., Ccznud. Bet. Rec., 3, 74 (922). 2. Schofield, F. S., J. Am. Vet. Med. Assn., 64, 553 (923-24). 3. Roderick, L. M., J. Am. Vet. Med. Assn., 74,34 (929). 4. Roderick, L. M., Am. J. Physiol., 96, 43 (93). 5. Roderick, L. M., and Schalk, A. F., North Dakota Agric. Exp. Stat., Bull. %50 (93). 6. Pickering, J. W., Blood plasma in health and disease, London (928). 7. WBhlisch, E., Ergebn. Physiol., 28, 443 (929). 8. Eagle, H., A symposium on blood and blood-forming organs, Madison, 242-264 (940). 9. Quick, A. J., Am. J. Physiol., 8, 260 (937).

Campbell, Roberts, Smith, and Link 55 0. Howell, W. H., Physiol. Reu., 6, 435 (935).. Eagle, H., Medicine, 6, 95 (937). 2. Brown, J. M., Savage, A., and Robinson, A. II., SC. Agric., 3, 56 (933). 3. Roberts, W. L., Summaries of doctoral dissert,ations, 936-37, Madison, 2, 204-206 (938). 4. Quick, A. J., Stanley-Brown, III., and Bancroft, F. W., Am. J. Med. SC., 90, 50 (935). 5. Smith, W. K., Science, 87, 49 (938); J. Agric. Research, 69,2 (939). 6. Smith, W. K., and Brink, R. A., J. Agric. Research, 67, 45 (938).

STUDIES ON THE HEMORRHAGIC SWEET CLOVER DISEASE: I. THE PREPARATION OF HEMORRHAGIC CONCENTRATES Harold A. Campbell, Willard L. Roberts, William K. Smith and Karl Paul Link J. Biol. Chem. 940, 36:47-55. Access the most updated version of this article at http://www.jbc.org/content/36//47.citation Alerts: When this article is cited When a correction for this article is posted Click here to choose from all of JBC's e-mail alerts This article cites 0 references, 0 of which can be accessed free at http://www.jbc.org/content/36//47.citation.full.ht ml#ref-list-