Protoplasts from Plant Materials - ACS Publications

the cell contents (protoplasts) liberated (4, 7). The protoplasts remained as discrete entities and could readily be separated from the bagasse (cutic...
1 downloads 0 Views 769KB Size
I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY

February 1948 (13)

Hirschler, A. E., papers presented before Div. of Petroleum Chemistry, A.C.S., pp. 153-69, Memphis, Tenn., April 1942; J . Znst. Petroleum, 32, 33-61 (1946).

Lipkin, M. R., and Kurtz, S. S., Jr., papers presented before Div. of Pet.roleum Chemistry, A.C.S., pp. 47-53, Atlantic City, N. J., September 1940. (15) Lipkin, M. R., and Kurtz, S. S., Jr., IND. ENO.CHEM.,ANAL. ED..13,291-5 (1941). (16) Mills, I. W., Hirschler, A. E., and Kurtz, S. S., Jr., papers presented before Div. of Petroleum Chemistry, A.C.S. Meeting in Print, pp. 91-114, September 1945. (17) Pearce, G. W., and Avens, A. W., “A Colorimetric Method of Determining Oil Deposit on Plant Surfaces Treated vith Oil Sprays,” in preparation. (18)Pearce, G. W., and Chapman, P. J., N. Y. Agr. Expt. Sta., Bull.

(22) (23)

(14)

293

Standard Oil Development Co., ‘‘Basic Values for Calculatinp Viscosity Index,” Circ. 30.50 (1938). Standard Oil Development Go., “Conversion Tables for Kine. matic and Saybolt Viscosities,” Supplement to Circ. 30.90A (1938).

(24)

Taschenberg. E. F., “Studies on the Control of the Grape-Berry Moth, Polgchrosb viteana (Clem.),” Ph.D. thesis, Cornell Univ..

(25)

Vlugter, J. C., Waterman, H. I., and van Westen, H. A., J

(26)

Wilhelm, R. M.,“Tag Manual for Inspectors of Petroleum’ 25th ed., Brooklyn, N. Y., C. J. Tagliabue Mfg. Co., 1929

1945.

Znst. Petroleum Tech., 21, 661-76, 701-08 (1935).

698, 15-16 (1942).

(19)Pearce, G. W., Chapman, P. J., and Avens, A. W., Enfomol.,35, 211-20 (1942). (20) Penny, D. D., Ibid., 14,428-33 (1921). 121)

J. Econ.

Sohiessler, R. W., Clarke, D. J., Rowland, C. S., Sloatman, W. S., and Herr, C. H., Petroleum Refiner, 22, 390 (1943); Proc. Am. Petroleum Znst.. 24, Seo. 3, 49-74 (1943).

RECEIYEDFebruary 24, 1947. Abstract of a theeis submitted to the Gradu. ate School of the Pennsylvania State College by G. ITr. Pearce, June 1946, BC partial fulfillment of the requirements for the Ph.D. degree. Journal Paper 698, New York State Agricultural Experiment Station, Geneva, N. Y . Au. thorired for publication as Paper 1360 in the Journal Series of the Penn. sylvania Agricultural Experiment Station.

Protoplasts from Plant Materials Properties of Protoplasts Released by Anaerobic Fermentation with Clostridium roseurn JONATHAN W. WHITE, JR., LEOPOLD WEIL, JOSEPH NAGHSKI, EDWARD S. DELLA MONICA, AND J. J. WILLAMAN Eastern Regional Research Laboratory, Philadelphia 18, Pa. ’I’he 2,000,000 tons of leaf waste incident to the commercial production of vegetable crops contain 50,000 tons of protein and 20 tons of carotene, besides other lipoid constituents. As a preliminary to the recovery of these materials, some method of concentration is desirable. It can be accomplished by release of the protoplasts as a result of a 2-day fermentation of the cell walls. The dried protoplasts contain from 31 to 56% protein, 18 to 27% (ipoids, and 0.04 to 0.2% carotene, representing a two- to aeven fold increase in concentration ovcr the original leaves. The protein of leaf protoplasts was digestible by trypsin. The process is also applicable to fleshy tissue such as carrot roots, sweet potato tubers, and winter squash. Concentrates containing up to 2.2% carotene were preQaredfrom carrots without solvent extraction.

W

HEN cryptostegia leaves (C. grandijoru) were subjected to anaerobic fermentation by Clostridium roseurn as a pre-

treatment in recovery of rubber, the cell walls were digested and the cell contents (protoplasts) liberated (4, 7). The protoplaste remained as discrete entities and could readily be separated from .the bagasse (cuticle, ribs, vascular tissue). Since the fractionation .of the leaves was so clean cut and the recovery of the protoplasts and of their water-insoluble constituents so high, it was deemed advisable to apply the fermentation process to other plant materials as a first step in the preparation of proteins and lipoids. This paper reports the successful preparation of protein, fat, and carotene concentrates from leaves of turnip, lettuce, beet, pass, and broccoli; from whole pea and snap bean vines and alfalfa; from the roots of carrots and sweet potatoes; and from the fruit of winter squash. Because of the special interest of this laboratory in the utilization of the leaf wastes incident to the .m-~ductionand processing of vegetables, the leaves chosen for

study were mostly from this source. The 2,000,000 tons of such fresh waste produced annually in the United States (8) contain 50,000 tons of protein and 20 tons of carotene. Any method for recovering that protein and carotene is intriguing. The preparation of protein concentrates from grass has beeu attempted with some success. A mechanical process patented in England (8) recovered only part of the proteinaccous material Sullivan (9, 10) dissolved the proteins with dilute alkali and reprecipitated them with acid. The crude product contained 58% protein, which was 56% of the’total protein in the grass. Surplus and cull carrots are the predominant source of commer. cia1 carotene, Many present methods for the preparation of carotene necessitate drying the fresh roots in order to make the carotene accessible t o nonaqueous solvent extraction. This drying ip costly and involves considerable loss of carotene by oxidation. Holmes and Leicester (3) have proposed dehydrating the roote with acetone prior to solvent extraction. This method eliminates the oxidative destruction of carotene, but the large quantities of solvent involved make the process expensive. Barnctt (9) developed a method of recovering a carotene concentrate from fresh carrots by mechanical means, This method requires extensive disintegration of the plant tissues. In fact, unless the walls of the individual cells are broken to allow the escape of the carotenecontaining particles, the yield of carotene is low; furthermore, a considerable amount of pulp is incorporated into the concentrate. resulting in much lower carotene”purity. FERMENTATION OF LEAFY VEGETABLE WASTES

Leaf material from eight species of plants was used in thia study to determine the applicability of the fermentation process for the recoverv of protoplasts. The source, degree of maturity, and portions of plants used in the experiments were as follows:

INDUSTRIAL AND ENGINEERING CHEMISTRY

294

Vol. 40, No 2

ture of C. rosew7~(AlcCoj aiid SlcClung). Tncubation L v a G conducted at 35O to 39" G Chemical Analysis= Recovery in Products, __ yo of Original Anaerbiosis was maintained in Etherthe fermenter by leading in soluble True CaroEtherDrv wt.. material protein, ish, tenr5, Plant soluble True Carocarbon dioxide generated from Fraction 4% % 7% y/g nr> wt material protein Ash tene g. dry ice. Complete sterilizaPea Vines tion r a s not necessary. Thcre 10?5 13.2 15.1 3.5 Original 118 (100) (1;;) (100) (100) (100) 4.8 12.7 4.8 24 20.0 206 Bagasse 19.3 6.4 4.1 was no contamination by othc? 24.3 147 35.4 16.0 692 14.3 100 Protoplasts 38.6 15.2 84 organisms. This was also true Snap Bean Vines In the experiments with erypto12.7 3.7 11.7 Original ti90 155 (100) (100) (100) h g i a (7). 3.5 181 7.4 5.7 39 26.2 Bagasse 24.9 15.3 20.5 39.6 17.6 908 15.9 Protoplasts 108 86.7 48.8 h 22-liter flask held leavw equivalent t o about 750 granie Alfalfa of dry neight, Tvhich supplied 4.1 12.4 6.8 (lty)3 Original 727 4.6 12.4 2.6 373 Bagasse sufficient protoplasts for chem17.9 52.4 2 . 7 1460 1 3 . 3 Protoplasts 97.0 ('81 evaluation. Rhen several Bluegrass (Lawn Clippings) pounds of protoplasts neie 9.5 24.8 8.9 722 Original 234 (100) (100) (100 desiied, a 40-gallon fermenta21.4 5.3 207 116 28.7 8,9 Bagasse 24.7 5.1 14 *?I5 8 I,fr 23.0 140 075 19.3 Protoplasts 46.8 43.7 3.1 31 tion nas carried out In a Turnip Topi barrel, as nas done u ith crypio13.0 21.0 3.7 381 iaoo) (100) 229 (100) Original stegia leaves ( 7 ) . (100) (lOU) 13.9 15.4 11.5 54 3.8 14.4 11.8 4.0 Bagasse 2.8 0.8 The fermentations iwre coin31.4 2 9 . 2 2 2 . 3 846 2 1 . 8 131 52.7 8 3 . 0 Protoplasts 26.4 81 plete in 2 day? The tissues Beet Tops n ere then disintegrated by miid 13.2 17.6 (100) (100) 716 6.5 Original (100) (100) 21.2 9.5 (l:%i 8.8 18.6 21.8 97.6 Bagasfie 4.7 13 agitation, and the relederd 33.8 !J.6 755 25.5 22.1 183 86.7 65.3 Protoplasts 14 4 8X protoplasts were separated froin Lettuce the fibrous materials by neA 12 0 15.8 10.3 (100) (1.00) 126 (10:) Original 486 (100) (100) screening through a combina 7.0 5.7 10.2 49 , 0 34.3 7.0 4.1 Bagasse 2.6 0 2 38 3 12 7 490 13.1 26.6 58.1 30.8 38.1 rion of 20- and 80-mesh sieves Protoplash 9.6 46 Although on a large scale a n Broccoli 80-mesh gyrating screen alone 17.6 13.3 5.8 469 (100) Original 746 (100) (100) (100) (100) 2.3 !0.4 102 3.0 32.5 22.5 17.0 1.7 Bagasse 0.6 O.B Ras adequatr, on a laboratori 04 a 2 0 21 8 2000 39.5 149 118 295 Protoplasts 5.9 168 scale the separation was facili'\Inistiire-frre basis. b i n adsorption of pigment solution from ferinentation products in Wall and Relley procedure, several additional tated by the use of a coarse bands dereloped: these were included as "carotene." whirh may mean t h a t the values for carotene after fermenscreen followed by a fins tation are too high. This phenomenon was confinpd to carotene determination on fermentation produots of leafy materials and was not noted in products reported in Table 111. one. The residue, bagasqe, was iesuspended in a small volume of water and screened again t o Pea vings. From garden-grovn, fully mature plants with many recover mechanically trapped protoplasts and then dried in circudying leaves. The whole vine, minus roots and pods. lating air at 34' C. Snap beans. Garden-grown, fully mature, whole plants, The protoplasts, which settled to one-fifth volume or less over minus roots and pods. night, were washed twice by decantation, roncentrated to n Alfalfa. Ficld-grown plants just starting to bloom. Dead and woody stalks discarded. thick paste by centrifuging, and dried in circulating air at 34" C Bluegrass. 2-Inch clippings from a lawn consisting predomiThe following analyses were made on the original plants anti nantly of bluegrass; very lush growth col1:hcted in May, air-dried on fermentation products: a t 34" C., and stored a t 30-34" C. for 3 months before use. Turnip tops. Leaves, including petioles, from garden-grown, Moirture was determined by heating at 65' C;. I n varuo to CVII mature plants. stant weight. Total nitrogen was determined by the Kjeldahl Beet tops, Leaves, including petioles, from farm-grown, maprocedure on ether-extracted material (to exclude lipide and t m e plan ts. other ether-soluble nitrogen) and calculated t o the original dr> Lettuce. Leaves from garden-grown plants which had bolted basis. To determine the nonprotein nitrogen, 2 giams of dried and were starting to blossom. product were trcxated with 80 ml. of a 57, solution of trichloroBroccoli leaves. Webs stripped from midribs of farm-grown acetic acid, heatcd to boiling, and filtered into a 100-ml. volu plants in the edible st,age. metric flask. The filtrate was made up to volume by continued washing of the residue with 575 trichloroacetic acid, and the niIt was found that the fermentatioii procedure developed for trogen was determined by the Kjeldahl method. The differenre recovery of cryptostegia protoplasts ( 7 ) could be applied directly bctwcen total nitrogen of the ether-extracted material and the or with only slight modification to these materials, It was unnonprotein nitrogcn was considered t o be thc true protrin nitro necessary to preboil them, since, unlike cryptostegia leaves, none gen, which, multiplied by 6.25, gave the true protein. Ash, ethei extract, and crude fiber were determined according to official contained substances inhibitory to C. Toseum. It was deemed methods (f). Carotene was determined by the mcthod of Wnl' advisable, however, to blanch the tissues by steaming to destroy and Kelley (12). The results are shown in Table I. enzyme activity and to wilt the leaves so that they would pack more readily into the fermenter. DIGESTION OF PROTOPLASTS BY TRYPSIN The leaves were first rinsed several times n-it,h water to reWith the possibility in mind of using the protoplasts in fwd, move the adhcring soil and sand particles. They m r e cut into it ~ ~ a ,deemed s advisable to demonstrate their digrstion by pro2- t,o 3-inch (5- to 7.5-cm.) pieces, packed in a 22-liter roundteolytic enzymes. To protoplasis (40 mg of protein nitrogen or bottomed flask, steamed for 10 minutes a t atmospheric pressure, 1.82 nig. of nitrogen per ml. in final volume) suspended in 10 ml and cooled. Thcy were then diluted with tap water to a concenof 0.1 M phosphate buffer, pH 8.8, were added 2 ml. of trypsin tration of 5% solids, calculated on the dry ncight of the original solution (I gram of commercial trypsin in 50 ml. of 0.1 M phoa leaves, and inoculated with 10% volume of an 18-hour brot,h cul'r4BLE

I.

COMPOSITION O F PLAXT SIATERIALS AND RECOVERY O F CONsTITUENTS IN FERJIBNTATION PRODGCTS

THEIR

._-

I _

!i!

(ly)

E

February 1948 T 4 R r a

INDUSTRIAL A N D ENGINEERING CHEMISTRY

OF VARIOUS PLANT 11. DIQESTIBILITY TRYPSIN

PROTOPLASTS BY

Nonprotein Nitrogen in Sample after Digestion,

~i~~~~~~ Protein Nitrogen Digested0 Plant in Blank, Protoplasts Mg./Ml. Mg./MI. Mg./MI. % 37 0.68 0.27 0.95 Turnip top 35 0.64 0.27 0.91 Pea vine 35 0.64 0.27 0.91 Bean 27 0.49 0.25 0.74 Lettuce 62 1.12 0.27 1.39 Alfalfa 40 0.24 0.73 0.98 Beet tops ' 1.47 81 0.39 1.86 Grass 1.22 67 0.39 1.62 Broccoli Each sample oontained 1.82 mg. of protein N per ml. before digestion.

ph2tte buffer, pH 8.8). To each were added 4 drops of toluene as a preservative. The enzyme solution, incubated for a similar period, was added to the controls a t the end of the incubation period. After incubation a t 34" C. for 24 hours, 10 ml. of 10% solution of trichloroacetic acid were added to each, the mixtures were filtered, and nitrogen in the filtrates was determined. Table I1 shows that under these conditions the protoplasts are susceptible to digestion by trypsin. FERMENTATION OF ROOTS AND FRUITS

In contrast to the cells in leaf tissue, those of storage tissue, such as the ones in Table 111, are highly vacuolated, with masses of carbohydrates and oils, so that the proteinaceous phase is not so continuous as it is in the leaf chlorenchyma. Destruction of the cell walls and removal of the carbohydrates by fermentation and solution resulted in fragmented undefined protoplasts rich in oil. Microscopic examination of the material from carrots showed free oil globules and carotene crystals. Furthermore, the high carbohydrate content of the storage tissues led to formation of considerable gum. This appeared ae a viscous, slimy layer full of gas bubbles, which floated on the surface of the fermentatiop liquor. The protoplast fraction, together with the skins and unfermented cellulosic materials, waa trapped within this gum, from which it was separated with some difficulty.

CARROTS.Fresh carrots were ground fine and extracted once with boiling water to remove excess cprbohydrates and hence to reduce the amount of slime formation. The extracted plant material was then made up to 5% solids, inoculated with an 18hour broth culture of C. roseurn and covered with nitrogen. Incubation was carried out a t 34' C. for 2 days. Bv this time the fermentation was complete, and the material, reduced considerably in volume, had formed a frothy layer on the surface. The liquor was drained off, and the gummy layer was diluted with an equal quantity of water made neutral with sodium hydrouide, treated with 0.03% (weight by volume) of aluminum sulfate to facilitate subsequent flocculation, boiled and passed through an 80-mesh screen. The bagasse (outer cork tissues and the central uvlem) stayed on the screen; the carotene-containing oil globules rLhd a small quantity of adhering protoplasts passed through. The proteinaceous material flocculated on cooling, and when it settled carried down the carotene-containing oil globules. The coagulum was recovered and dried. Seven experiments on the preparation of a carotene concentrate from carrots were carried out, with analytical control. The results are. shown in Table 111. Sweet potatoes obtained on the market SWEETPOTATOES. were ground in a meat grinder, and 1 kg. of the pulp (252 grama dry weight) was boiled for 15 minutes with 1.5 liters of water and atraincd. A preliminary experiment indicated the necessity of preboiling the pulp. After cooling, the pulp was inoculated with 200 cc. of an 18-hour broth culture of C. roseurn, the volume made up to 2 liters, covercd with nitrogen, and incubated a t 37' C. for 4 days, although the fermentation was almost complete in 2 days. At the end of fermentation, about 400 cc. of a slimy, yellowbrown, gummy layer were on the surface, containing the skins and other debris. The liquid underlying this was clear but orange colored. The upper layer was recovered and diluted with 300 cc. of water, neutralized, treated with 3 grams of aluminum sulfate, and boiled for 20 minutes. Screening removed 2.2 grams dry weight, of undigested corky tissue. The solids flocculated and

295

after settling overnight had compacted to a small volume. This was further concentrated by centrifuging and dried at 50" C. in vacuo. Table IT1 shows that a sixteen fold inoreasc in carotene concentration was obtained. SQUASH. Winter squash (Hubbard) was round in a meat grinder and 1 kg. (167.8'graas, dry weight? of the pulp waa boiled !or 20 minutes in 1.5 liters of water and drained on cheese cloth. The pulp was inoculated with 250 cc. of an 18-hour culture of C. meum, made up to 2 liters, covered with nitrogen, and fermented a t 37" C. To determine the effect of escess soluble carbohydrate 1 kg. of ulp was fermented without hot-water extraction. h e n the krmentations were complete, the pulpe had been reduced in amount and had risen to the surface as a gummy layer. The boiled pulp yielded a larger volume of a more slimy gum than the unboiled. The gummy layers were diluted with water to 1.5 liters, neutralized, treated with 3 grams of aluminum sulfate, and boiled for 20 minutes. They wcre then screened to remove coame particles of corky tissue and set aside to allow the proteinaceous material to flocoulate. The material from the extracted pulp did not flocculate until 1.5 grains more of aluminum sulfato were added and it WRS boiled again. The supernatnnts were decanted and the sediments centrifu The resulting concentrates were then dried in vacuo a t 50 Analytical results are shown in Table 111. DISCUSSION

All plants fermented rapidly, loss in dry weight ranging from 35% in alfalfa to 80% in lettuce. Plants containing a large proportion of stems and fibers (pea vines, bean vines, alfalfa, and grass) showed less loss in weight than those consisting mostly of leaves and yielded more bagasse (20 to 51%). Thc authors have observed that C. roseum does not attack the cellulose of the vascular system, which comprises the greater part of stems and veins (7). Photomicrographs in that publication show this type of disintegration. The protoplast fraction, which represented from 13 to 25% of the weight of the original material, was rich in protein (31 t o 56%), ether-soluble materials (18 to 27%), and carotene (0.04 to 0.20%). This concentration was accomplished by the extensive digestion of carbohydrate and cellulosic materials during fermentation and also by the removal by the screening process of unfermented gross structures (cuticle, vascular tissue, ctc.). Lettuce leaves and lawn clippings gave low yields of protoplasts, protein, and carotene, probably because the protoplasts became highly fragmented during the fermentation, did not sediment properly, and consequently were not completely recovered. The data of Table I show that ether-soluble substances werc recovered in fairly good yields from most of the fermented plants The ether extracts of the protoplasts were usually high. Recoveries above 100% may have been due to the absorption of compounds produced in the fermentation, or possibly to the greater availability of certain constituents to cstraction, since in most cases in which the recovery of ether-soluble substances was more than loo'%, that of carotene was also high. In the case of cryptostegia leaves (4), rubber could not be complete11 removed from untreated leaves even after prolonged extraction After fermentation, however, the acetone-soluble constituent8 and the rubber were extracted rapidly and completely. Except for alfalfa and broccoli, protein yields were somewhat low. Loss of protein in the process could be caused by three factors. incomplete recovery of the protoplasts during washing. mechanical entrapment of protoplasts in the bagasse fraction, and destruction of protein by the organism. Tho first loss is dependent on the nature of the plant material; bluegrass and lettuce yielded fragmented protoplasts with poor settling properties. Bluegrass and alfalfa bagasse, which was fibrous, entrapped the protoplasts. The contribution of the third factor wa8 not determined; in two fermentations, alfalfa and broccoli, all the original protein was accounted for in the protoplasts and bagasse. Both these plants yielded relatively large protoplasts, which settled rapidly, with consequent negligible loss of fragmented protoplasts in the wash water. However, no mechanical loss was apparent in turnip tops, beet tops, pea vines, and bean vines, and

INDUSTRIAL AND ENGINEERING CHEMISTRY

296

TAn1.F: 111. CAROTENE COSCESTRATES OBTAINED FERMENTATION OF ROOTS A N D FRCITS .Dry weight . halysis, G. % r/e. Carrots 839 100 1,140 38 8 4 6 1,120 42 (1 5 0 22.400 ~

No

Plant Frnction

14tf

Original Bagasqe concentrate Processing liquur 17A Original Bagasse Concentrate Processing liquor Original Original inA Bagasse Concentrate Proresuing liquor 228 0 Original riginal Bagasse Concentrate Processing liquor Original 24A Bagasse Bagasre Concentrate c Proces-ing liquor z ~ A Original Bagasse Concentrate Processing liquor Z6A Original Bagasse Concencrata Processing liquor 2 2 ~ Original

Bagasse Concentrate

120:4

...

1.7

100 1.4

6.6

5.4

1140 134 1.5 11.3 2060 2960 26.7 190.7

...

439 8.30 14.91 2450 256 3.48 12.30 2 ~ 6 4.48 18.80

..

a70 810 15,500 18.41. 728 960 13, I60 1.4 626

BY

Carotene .-__ Total, % of r Oria. 966,000 43,400 941,000

100 4 5 98 5

~oi,boo ib4

1 .3

6.4

9.312

1,400 101,000 28.700 97,600 1,440 148.700 2,880 1,860,000 25,100 1,778,000

166 1. Y 3.4 100 1.5 5.2

778 1.220 11,200 1.8 778 415 6.800

334,000 9,960 167,000 4,420 183,600 1.4.50 83.700

100 2.9 49.3 1.3 100 0,s 45.0

100 1.9 8.0

778 490 6,240

ii3,koo 2190 117,200

1.2 63.8

806

100

908

112

.,.

100 1.1 8.4

...

100 0.9

...

...

~.

dweet Potatoe, 252.0 100 2.2 0.9 17.4 6.9

940

...

. . , d

..,d

3.2 ,.,

52.0

97,l

27 3 100 2.4 152 2.9 100 1.3 90.6

...

iOo

...

Winter Squash Original 167.8 100 140 23,690 100 Concentrate from unex trac ted pulp 38.3 21.0 875 30,900 131 Concentrate from extracted pulp 12,9 7.7 1,830 23,600 100 * Al-treated processing liquor was not coloied and hence presumably conrained no cnrotene. b Color of liquor was orangs, c Additional 1.34 g. of concelitrata analyzing 7,050 y carotene per g roaovered from fermentation liquor in this experiment d These liquors were similar in color t o t h a t from experiment 24-4 23A

thls leads one to suppusr that parr of the protein was digested by C. roseurn and lost as soluble nitrogen. The proximate analy-

sis cariied out on the protoplasts accounts for only 70 to 80% of the dry wight. What constitutes the remainder is unknown. Since only the protoplasts constitute the enriched fraction, thr percentage of total protein and other constituents recovered in this fraction is of interest. These data are included in Table I Bt>causeof the fibrous nature of alfalfa, it Tvas difficult to make t~ clean separation of the protoplasts from the bagasse, with the result that only 56% of the original protein was recovered in the protoplast fraction. Beet tops gave a better separation-65% while broccoli leaves proved to be the best, rvith complete revovery. It is possible that, by introducing some modification, quantitative recoveries can be obtained from most plants. The true protein in the protoplast fractions of the plants used repre-ents a two- to fourfold concentration over the original plant, gith protein contents ranging from 31 to 56%. Extraction of the ether-soluble substances would increase the protein content by at least 2074 and at the same time recover the carotene and fats. The nonprotein nitrogen, which ranged from 0.6% in snap bean vines to 2,9y0in brocroli, represents a Considerable fraction of the nitrogenous material and one which cannot be recovered by the method described here, a s it is noncoagulable and leaches out of the protoplasts. The crude fiber in the protoplasts, nrhich ranged from 0.8% in broccoli to 5,170 in lettuce, represents a small amount of fine cellulosic debris which passed through the screen along with the protoplasts in the separation process. A considerable quantity of the ash was leached out during the fermentation and washing of the protoplasts. The high values

Vol. 40, No. 2

obtained in the protoplasts from ct:rtain lox-growing vegetahieb (turnips, beets, peas, beans, and lettuce) may he due partly ti, sand not, removed during the washing. Carotene was recovered in good yield except in lettuce and lawl~ clippings. Recoveries of over 100% obtained with alfalfa and broccoli may reflect the greater ease of extraction from fermented plants ( 4 ) , and may also be due to the presence of interfering pigments (see Table I). The protoplast fraction corit,ains most of the carotene and represent8 a four- t,o twelvefold coneencration, in the case of broccoli yielding a product with a.1Inosi 0.27, carotene. Attempts to ferment rhubarb leiLve8 were unsuccessful, 'rklr! fermentation of unentracted leaves was slow arid incomplete, artti even after the leaves m r e extracted three to four timej ivit,li boiling water the results were unsatisfactory. This was al~parently due t o the presence of antibiotic substances (6). Roots (carrot and meet pot,atoes) and fruits (squash) wt:rtfermented successfully. Since these tissues contain onlS sma.11 quantities of protein and cellulosic materials, they ere a.lrnost completely digested. The product from carrots contained 0.9 to 2.2% uurotent; (Table 111) and represented 91 to 98T0 of the initial carotene content of the carrots except for the last three runs, which art discussed below. The Wall-I