Phytochemistry. 1966, Vol. 5, pp. 423 to 438. Pergmum Resl Ltd.

Printed ia En&al

PLANT PHENOLIC COMPOUNDS AND THE

ISOLATION OF PLAN’? ENZYMES*

W. D. LOOMISand J. BAITAILE
Science Research Institute, Oregon State University, Corvallis, Oregon; and Department of
Chemistry, Southern Oregon College, Ashland, Oregon, U.S.A.
(Received I August 1965)

Abstract-The presence of phenolic compounds makes it impossible to isolate active enzymes by conventional
techniques from many plant tissues. Phenols combine with proteins reversibly by hydrogen bonding, and
irreversibly by oxidation followed by covalent condensations. The chemistry of these reactions is reviewed,
and their importance in the isolation of enzymes from plants is discussed. Phenols are e!Tectively removed
from H-bonded complexes with protein by adding large amounts of substanceswhich contain groups similar
to the peptide bond. A technique. is described in which insoluble polyvinylpyrrolidone is used to adsorb
phenols and thus obtain active soluble enzymes. Using this technique, mevalonic kinase, phosphomevalonic
kinase, glutamyl transferase and alkaline phosphatase have been demonstrated in extracts from peppermint
leaves.

INTRODUCTION

THE isolation of enzymes and other cell components from any organism generally involves
disruption of the tissues, with inevitable mixing of substances which in the living organism
were rigidly compartmentalized. This may result in inactivation of enzymes, or in the isola-
tion of an enzyme which, though active, is more or less modified from its native form. Plant
tissues present all of the usual difficulties due to interactions of proteins with other protoplas-
mic constituents, such as nucleic acids and lipids, as well as to the inherent instability of
proteins. In addition they present special problems.
Stahmann’ and Pitie in recent reviews cite a number of factors which make plant proteins
particularly unstable and difficult to work with. Included are vacuole acids, carbohydrates,
hydrolytic and oxidative enzymes, phytic acid, and tannins. Adsorption of soluble proteins
by cell wall fragments has been discussed in a recent review by Newcomb. Cell walls and
vacuoles make up the bulk of most plant tissues, and both contain a variety of materials which
can react with proteins, and which cause special problems not encountered with other
organisms. Our purpose here is to consider problems caused by the plant phenolic com-
pounds, or tannins.? In living cells these compounds are frequently associated with vacuoles,4
but information on their localization is limited. There is a tendency to assume that tannins
cause difficulty only in a few species of plants, and only because they precipitate proteins.
l A preliminary report was presented at the 1964 meetings of the American Society of Plant Physiologists.
An abstract appeared in Plant Physiof. 39, Suppl. xxi (1964).
t The term “tannin” has no precise chemical meaning. It is used loosely to describe. plant phenolic
compounds which precipitate proteins, or in a more restricted sense to designate only those plant phenol&
which are capable of converting raw hide into leather. In this paper the word will be used primarily in dis-
cussing the findings of other investigators and will be used in each case in the same sense that they have used it.
1 M. A. STAHMANN, Ann. Rev. Plant Physiol. 14,137 (1963).
2 N. W. Prtux, Ann. Rev. Pfant Physiol. 10, 33 (1959).
3 E. H. N~WCOMB,Ann. Rev. Plant Physiol. 14,43 (1%3).
* K. &AU, PIant Anatomy (2nd Ed.). Wiley, New York (1965).
423
424 W. D. LOOMISand J. BA~~AILE

In fact, problems due to reactions between plant proteins and plant phenolic compounds are
much more prevalent and much more complex than is generally recognized. Animal tissues
contain few phenolic compounds other than the amino acid tyrosine, and seldom in high
concentration,5 but in plants, phenolic compounds (in addition to tyrosine) are widespread,
and frequently occur in very high concentrations.5-8
All phenols, unless sterically hindered, take part in hydrogen bonding,9 and the bond
formed between phenols and N-substituted amides is one of the strongest types of hydrogen
bond.rO Phenol itself has been shown to form complexes with proteins,” nylon12 and poly-
vinylpyrrolidone (PVP). *13 In addition, many phenols are readily oxidized to quinones,
which are highly reactive compounds. 9S14*I5 Oxidation of phenols may occur nonenzymatic-
ally, or it may be catalyzed by phenol oxidases or peroxidases, enzymes which are widely
distributed in plants. Quinones are oxidizing agents and may oxidize essential groups of
proteins. More important, they polymerize rapidly, and in the presence of protein they also
react rapidly to form covalent bonds to the protein.
The complex reactions of phenols and quinones with proteins and related compounds
have been studied in connection with several seemingly unrelated problems. An understand-
ing of these reactions is important to any one working with plant proteins, but the literature is
widely scattered, and much of it is in specialized publications which may not be readily
accessible to plant biochemists. We would like to review this literature briefly and also
describe an improved technique which we have developed for isolating soluble enzymes from
plants which contain phenolic compounds. We hope that the review will not only explain
our own results but also provide a basic understanding of the processes involved, thus aiding
in the further development of methods for working with plant proteins.
Chemically the plant phenolic compounds are extremely heterogeneous, ranging from
simple monomers to very large polymers. In many cases the structures are still unknown.
However, most of them belong to one of two biochemical groups :16the flavonoid compounds
(including the condensed tannins), or the group of C$-C3 and C&i compounds and their
derivatives (such as caffeic acid and gallic acid, and including the hydrolyzable tannins as
well as tyrosine and lignin). Compounds of the first type generally contain only phenolic
hydroxyl groups as reactive centers. Compounds of the second type commonly contain also
carboxyl groups, either free or esterified. In both types of compounds it is usual for some of
the hydroxyl groups to be substituted.

l Abbreviations used: PVP, polyvinylpyrrolidone; ADP, adenosine diphosphate; ATP, adenosine tri-
phosphate.
J J. B. HARB~RNE and N. W. SIMMONDS, In Biochemistry ofi’henolic Compouti (Edited by J. B. HARBORNE),
p. 77. Academic Press, London (1964).
6 J. B. HARBO~NE,In Biochemistry of Phenolic Compounds (Edited by J. B. HARBORNE), p. 129. Academic
Press, London (1964).
7 R. E. A~sro~ and B. L. TURNER,Biochemical Systematics. Prentice-Hall, Englewood Cliffs, N.J. (1963).
8 E. C. BATE&UTH, In Chemical Pfant Taxonomy (Edited by T. SWAIN), p. 127. Academic Press, London
(1963).
9 R. H. THOMSON, In Biochemisfry of Phenofic Compounds (Edited by J. B. HAIUXDRNE), p. 1. Academic
Press, London (1964).
10 M. ST. C. FLY, J. Sot. Dyers Colourists 68,59 (1952).
11 K. H. Gusr~vso~, l%e Chemistry and Reactivity of Collagen. Academic Press, New York (1956).
12 H. ENDRES and H. H~RMANN,Angew. Chem. 75.288 (1963).
13D. GLJTWANN and T. HIGUCHI, J. Am. Pharm. Assoc., Sci. Edif., 45,659 (1956).
14 H. S. MASON, Advan. Enzymol. 16, 105 (1955).
1s H. S. MASON,In Pigment Cell Biofogy (Edited by M. GORDON),p. 563. Academic Press, New York (1959).
16T. ROBINKIN, Z%e Organic Constituents of Higher Plants. Burgess, Minneapolis (1963).
 Plant phenolic compounds and the isolation of plant enzymes 425

Experiments on the tanning of modified collagen17 and of synthetic polymers*7* 18-22

have established the importance of the peptide or amide linkage in the formation of H-bonded
complexes between tannins and protein. Grassmann and coworkers demonstrated in 1937
(see Gustavson17) that water-soluble urea-formaldehyde condensation products, containing
-CO-NH- as the only reactive group, precipitate tannins from solution. More recently
Batzer and Weissenbergerl** l9 and Gustavson and H0lm~~021 showed that hydrated nylon
precipitates tannins. Further investigations of Gustavson established that both soluble22
and insoluble23 polyvinylpyrrolidone form stable insoluble complexes with tannins. Hide

Polyvinylpyrrolidone

powder, hydrated nylon and insoluble PVP all have nearly the same capacity for binding
tannins. Under the conditions of Gustavson’s experiments23 the amounts of several types of
tannins bound ranged from 31 to 44 per cent of the dry weight of the substrate polymers.
Tannins are partially removed from all of these complexes by 6-8 M urea. Dilute alkali, or
aqueous organic solvents which are capable of H-bonding, such as alcohols or acetone, also
de-tan,17* 23but in certain cases they dissolve the substrate as well, so that general comparisons
cannot be made as they can with urea. From the tanning experiments it is clear that only the
-CO-N< group is required for the formation of complexes with vegetable tannins. It was
concluded that tannins form H-bonds with the peptide linkages, probably through the peptide
oxygen, and with the tannins furnishing the hydrogen. The two types of tannins show very
different pH responses in these experiments. 17*23 Condensed tannins are bound almost inde.
pendently of pH below 7-8, except for an indirect effect due to the swelling of collagen at pH 3
or below. The binding decreases rapidly above pH 8. Hydrolyzable tannins on the other
hand are bound very strongly at pH 34 (80 per cent of the weight of substrate), but the
binding decreases above pH 5. At pH 6 the amount bound is approximately 25 per cent of
that bound at pH 3, and at pH 7.5, only 8-10 per cent.
In the case of condensed tannins the pH effects indicate that the binding involves un-
ionized phenolic hydroxyl groups. It seems clear that these groups react to form H-bonds
with the substituted amide groupings of the protein or synthetic polymer. In the case of
hydrolyzable tannins the pH effects suggest that strong H-bonds are formed by unionized
carboxyl groups of the tannins, and weaker H-bonds by unionized phenolic hydroxyl
groups. Hydrolyzable tannins are predominantly derivatives of pyrogallol(l,2,3-trihydroxy-
benzene) and catechol (o-dihydroxybenzene). Chromatography experiments have shown
that such phenols, as a result of internal H-bonding, have relatively low affinity for polyamide.
Investigations of the chromatography of phenols and related compounds on polyamide
powder (principally nylon 6) have yielded valuable information about the types of interactions

17 K. H. GUSrAVmN, The Chemistry of Tanning Processes. Academic Press, New York (1956).
18 H. BATZER and G. WEISSENBERGER,Makromol. Chem. 7,320 (1952).
19 H. BATZER,Makromol. Chem. 8,183 (1952).
20 K. H. GULXALTON and B. HOLM, J. Am. Leather Chemists’ Assoc. 47,700 (1952).
21 K. H. GU~TA~SON, J. Polymer Sci. 12, 317 (1954).
** K. H. GU~~A~SON, Svensk Kern. Tidrkr. 66,359 (1954).
23 K. H. GUSTAVSON, Leak 14,27 (1963).
426 W. D. Locmas and J. BA’ITAILE

involved. This work has been reviewed recently by Endres and Hormann12 and by Egger.24
Phenolic compounds in general are adsorbed on polyamides by hydrogen bonding. The
affinity increases (RI decreases) with an increase in the number of phenolic hydroxyl groups
on the molecule, except when the hydroxyl groups are ortho to each other or otherwise
favorably arranged to allow internal H-bonding. In this case, additional hydroxyl groups
decrease the a5nity for polyamide. For example, phloroglucinol(1,3,5-trihydroxybenne)
has an Rf of 0.07 with water as the developing solvent, while pyrogallol (1,2,3&ihydroxy-
benzene) has an Rfof 0.36 and phenol has an RIof O-21.l2 The eluting strength of solvents
increases in the order: water, ethanol, methanol, acetone, dilute NaOH, formamide, dimethyl-
formamide. Surprisingly, tyrosine migrates with the solvent front, having very little affinity
for the polyamide. This is not the only instance of anomalous behaviour on the part of
tyrosine. Tyrosine is much less soluble in water than are other amino acids, including phenyl-
alanine.25* 26 The phenolic hydroxy group seemingly reduces the polarity of tyrosine, rather
than increasing it as one would expect. The water solubility of tyrosine is greatly increased
in the presence of urea. 26*27 In addition, the sublimation temperature of tyrosine is con-
siderably higher than that of phenylalanine and most other amino acids.25 These observa-
tions apparently have not been explained, but they suggest unique interactions between the
phenol portion and the alanine portion of tyrosine.
Brown and Wright2* have investigated interactions between milk proteins and tea poly-
phenols by means of electrophoresis. They found that tea infusion drastically altered the
electrophoretic mobilities of milk proteins. In the presence of 7 M urea, tea infusion had no
influence on protein mobility. The investigators concluded that protein-polyphenol com-
plexes are formed and that the interactions are due, at least initially, to hydrogen bonding.
Further evidence of complexing was obtained during attempts to separate ,the colored tea
polyphenols by gel filtration. The colored material of tea extracts would not move with
water on a Sephadex column and was only eluted on addition of O-1N NaOH. When casein
was added to the tea infusion before putting it on the column the brown color passed straight
through the column with the protein.
Although large amounts of phenolic substances can be bound to protein by hydrogen
bonding there have been many indications that other, more stable, bonds are also formed,
and that they are formed very rapidly. i7 It now appears that these stable bonds result from
oxidation of phenols to quinones, and copolymer&ion of the quinones with protein. The
enzymatic oxidation of phenols, and the reactions of quinones with proteins, have been
reviewed by Mason, 14*I5129 by Yasunobu,30 and by Bouchilloux.31 Gustavson” has also
reviewed quinone-protein reactions. Mason’s 1955 review14 is especially comprehensive.
Evidence is cited for the covalent bonding of quinones to proteins by l+addition of sulfhydryl
groups and the imino group of proline, in addition to free amino groups. Sulfhydryl groups
and terminal a-amino groups react much more readily than do the c-amino groups of lysine.
N-terminal proline also reacts rapidly. The initial product is a catechol- or quinol-protein
24 K. EGGER, PIanla Med. 12,265 (1964).
2s J. P. GREENSRIN and M. Wr~rrz, Chemistry ofthe Amino Acids, Vol. 1, p. 565. Wiley, New York (1961).
26 Y. NOZAKI and C. TANFORD,J. Biol. Chem. 238,4074 (1963).
27 P. L. WH~NEY and C. TANFORD.J. Biol. Chem. 237. PC 1735 (1962).
28 P. J. BROWN and W. B. WRIGHT; J. Chromatog. 11,%4 (1963): .
29H. S. MASON,Nature 175,771 (1955).
30 K. T. YASUNOBU.In Pigment Cell Biology (Edited by M. GORDON), p. 583. Academic Press, New York
(1959).
31 S. BOUCHILL~UX, In Plant Phenolics and their Industrial Significance. (Edited by V. C. RIJNJXKLES), p. 1
1962 Symposium Plant Phenol& Group of North Americia (1963).
 Plant phenolic compounds and the isolation of plant enzymes 427

compound, which is readily oxidized by free quinone to the corresponding quinone-protein.

Quinones with a second reactive ring position available can react with a second reactive
group of protein, resulting in cross-linking.
The hardening of cuticle and related structures of invertebrates has been shown in several
cases to be due to quinone tanning of proteins, the quinones arising by enzymatic oxidation
of phenols. Much of the available fundamental knowledge of quinone-protein reactions is
due to investigations of these tanning processes.“* 17,32p33
Recently the importance of quinones in vegetable tanning has been emphasized.23* 34
Endres3’ concludes that the essential reaction of vegetable tanning is l&addition of free
amino groups of the protein to bifunctional low molecular weight quinones formed by oxida-
tion of phenols, resulting in covalent cross-linkage. Quinones are bound irreversibly on
polyamide (nylon 6).12*35 Analyses showed slightly less than one mole of quinone bound
per two moles of free amino groups of the polyamide, and blocking the amino groups com-
pletely removed the afhnity for quinones. This was taken as evidence for successive 1,4-
additions of two terminal amino groups of the polyamide to each quinone molecule.
Wood and IngrahamJ6 have shown that when phenol oxidase oxidizes 14C-phenol, in the
presence of oxygen and ascorbate, radioactive material is bound irreversibly to the enzyme.
Quinone addition reactions are thought to be responsible.
When beer ages, or is exposed to oxidizing conditions, a precipitate frequently appears
which is referred to as “haze”. Chemical analyses have shown that beer hares contain
proteins, “tannins”, and in some cases heavy metals.j7* 38
The beer hare problem has been investigated intensively in the past few years, with
particular emphasis on a search for agents which would specifically remove phenolic com-
pounds from beer. Soluble PVP,38 nylon 66 and nylon 6,39 and keratir#O were all used with
considerable success by different investigators. McFarlane and Bayne’l tested several types
of cross-linked insoluble PVP, and found that one of these, originally designated “Agent
AT-496” was particularly promising. Compared with nylon, the new polymer removed
more of the haze precursors and fewer other substances. It had the obvious advantage over
soluble PVP that it could be used in excess without leaving any soluble residue in the beer.
Agent AT-496 was later referred to simply as “Agent AT”, and it is now available commer-
cially as “Polyclar AT”. In subsequent investigations McFarlane and coworkers have
studied the binding of various flavonoid phenolic compounds by Polyclar AT and have used
it for analysis of polyphenols,42-45 as well as for stabilizing beer. For calorimetric analysis,
beer polyphenols are adsorbed on Polyclar AT and then eluted with N-methyl-2-pyrrolidone.42
32 D. GILMOUR,The BiochemistryofInsects. Academic Press, New York (1961).
33 R. H. HACKMAN, In The Physiology of Insecta (Edited by M. ROCKSTEIN),Vol. 3, p. 471. Academic Press,
New York (1964).
34 H. ENDRJS,Lx&r 12,294 (1961). From Chem. Abstr. 56,13062e (1962).
35 H. ENDRES,~. AMI. Chem. 181,331 (1961).
36 B. J. B. WOOD and L. L. INGRAELW, Nature 205,291 (1965).
37 W. I. BENCXMJGHand G. ti, J. Inst. Brewing 61,134 (1955).
38 W. D. MCFAIUANE,E. WYB and H. L. GRAM, European Brewery COW., Roe. Congr., 5th Boa&&&n,
p. 298 (1955).
39 G. HARRISand R. W. RICKEI-IS,European Brewery Conv., Proc. Congr., 7th Rome, p. 290 (1959).
40 R. VAN CRAENEN~ROECK and R. LONIIE.European Brewery Conv., Proc. Congr.. 9th Brussels, p. 513 (1%3).
41 W. D. MCFARLANEand P. D. BAYNE, European Brewery Conv., Proc. Congr., 8th Viennu, p. 278 (l%l).
‘2 W. D. MCFARLANJZ, /. ht. Brewing 67, 502 (1961).
43 W. D. MCFARLANBand M. J. VADER,J. ht. Brewing 68,254 (1%2).
*4 W. D. MCFARLANLI and P. T. SWORD, J. Inst. Brewing 68.344 (1962).
45 W. D. MCFARLANE,P. T. SWORDand G. BLINOPT,European Brewery Conv., Proc. Congr., 9th Brussels,
p. 174 (1963).
428 W. D. LooMls and J. BAITAIIZ

Chapon et al.4648 have studied the formation of beer hazes, and changes in their solubility,
and suggest that their findings apply generally to work with plant proteins.46 Freshly formed
beer haze dissolved rapidly and almost completely on the addition of PVP of molecular
weight 40,000, while if the haze was allowed to age, its solubility in PVP decreased greatly.
PVP 40,000 (nondialyzable) had the additional effect of converting dialyzable phenolic
substances of beer into soluble non-dialyzable complexes. It was shown48 that a number of
putied proteins and polyphenols can act as precursors of precipitates which are essentially
the same as natural beer hazes. Among the factors which influence haze formation, ionic
strength was found to be important. Haze formation was greatly accelerated at ionic strengths
of less than 0~08.~~
Interference by phenolic compounds in the isolation of plant enzymes has been noted by
several investigators, and in some cases methods have been devised for overcoming the
inhibition.
Quaste14g in 1932 found that certain polyhydric phenols inhibited urease very strongly,
and that thiols could prevent the inhibition. Evidence was obtained which indicated that
the actual inhibitors were quinones and that even traces of quinones were extremely in-
hibitory. Difficulty in assaying enzymes of Kalanchoe blossfeldiana led EhrenbergSo to
investigate the inhibition of phosphatases by tannins extracted from KalanchoC. Phosphatase
from Phaseolus mdtiporus was strongly inhibited by the KalanchoB tannins. This inhibition
was reversed by the addition of hide powder, about 10 mg of hide powder per mg of tannin
being required for complete reversal. The phosphatase activity of suspensions of leaf powder
of KalanchoB was increased as much as 150 per cent by the addition of lightly chromed hide
powder. (Chrome tanning of hide protein increases its capacity to take up vegetable tannins. 17)
Addition of soluble proteins was said to reverse the inhibition of the Phaseolus phosphatase
but to have no effect on the activity of the KalanchoL; enzyme.
Friedrich5’ showed that @glucosidase from bitter almonds was strongly inhibited by
addition of tannic acid, and that the inhibition could be reversed by addition of lightly
chromed hide powder.
Forsyth and coworkerss2 measured the solubility and activity of several enzymes during
the curing of cacao beans. They prepared acetone powders of the beans and extracted the
powders several times with aqueous acetone, a procedure which should remove phenolic
material that is not covalently bonded. As the curing progressed the acetone powders ob-
tained became browner and contained increasing amounts of phenolic materials. At the
same time the enzymes (and all other proteins) became insoluble in water. The enzymes
retained a small amount of activity even after they had become completely insoluble.
Interactions of phenolic compounds with proteins are important in the curing of several
plant products. Forsyths3 has reviewed the physiological aspects of the curing of tea, cacao
and tobacco, with special emphasis on the role of phenols and quinones, and their reactions
with proteins.
Hathway and Seakins54 showed that pectinase was inhibited by the addition of tannins
46 L. CHAPON, B. CHOLLOT and E. URION, Bull. Sot. Chim. Biol. 43,429 (1961).
47 L. CHAPON, B. CHOL~ and E. URION, European Brewery Conv., Proc. Congr., 8th Vienna, p. 319 (1961).
48 B. CHOLLOT, L. CHAPON, and E. URION,European Brewery Cow., Proc. Congr., 8th Viennu, p. 334 (1961).
49J. H. QUASIT.L,Biochem.J. 27, 1116 (1933).
50 M. EHRENBERG, Biochem. 2.325, 102 (1954).
51 H. FRIEDRICH, Arch. Pkwm. 288,583 (1955).
52 W. G. C. FORS-I-H,V. C. QIJESN~L and J. B. ROBERTS, J. Sci. Food Agri. 9,181 (1958).
53W. G. C. FORSYITI,Ann. Rev. Plant Physiol. IS, 443 (1964).
54D. E. HATHWAY and J. W. T. SEAKINS, Biochem. J. 70,158 (1958).
 Plant phenolic compounds and the isolation of plant enzymes 429

and that the enzyme could be partially recovered from the enzyme-tannin complex by pre-
cipitation with 80 % acetone. Porter and Schwar@ isolated a water soluble substance from
grape leaves which had been shown to inhibit pectinase and cellulase and found it to be a
tannin. The inhibitor could be removed from extracts by hide powder, gelatine, caffeine, or
nicotine sulfate. The same group”j screened extracts from leaves of sixty-one plant species
for ability to inhibit enzymes. Extracts from twenty-nine species inhibited pectinase, and
extracts from fourteen inhibited cellulase.
Schwimmer5’ found that ethanolicextracts of potato tubers contain substances, apparently
phenolic, which inhibit potato phosphorylase. Bosers8* 59 studied the inhibition of malic
dehydrogenase and glucose-&phosphate dehydrogenase by known phenolic compounds,
especially flavonoids. Several of these compounds were potent inhibitors at a concentration
of 3 x 1O-4 M, and in some cases the inhibition was partly or completely reversed by addition
of serum proteins. Of the compounds tested, anthocyanins were among the most effective
enzyme inhibitors. Catechin and epicatechin did not in themselves inhibit, but they yielded
“acid condensates” which were very inhibitory.
Goldstein and Swai#’ have studied the inhibition of several soluble enzymes by a purified
hydrolyzable tannin and by a condensed tannin extract. The tannins precipitated all of the
enzymes, but most of the precipitated enzymes retained some activity. Soluble enzyme
activity could be obtained from the precipitates with varying degrees of success by addition
of borate, caffeine, polyethylene glycol, polyvinyl alcohol, methyl cellulose, non-ionic or
cationic detergents, or soluble PVP. In almost all cases PVP was the most effective.
Phenol oxidase of tea leaves has been regarded as a particulate enzyme and has proven
difficult to isolate in soluble form. However, Sanderson61 showed recently that it is completely
soluble if a large quantity of powdered nylon is added during the extraction of the tissue.
The isolation of active mitochondria from plant tissues has been hampered in several
cases by the presence of phenolic compounds. Tager6’ was unable to obtain mitochondria or
soluble protein from banana pulp by conventional techniques. He suggested that tannins
were responsible. When tripotassium phosphate was added to the isolation medium, mito-
chondria and soluble proteins could be extracted, but the mitochondria were extensively
damaged by the high pH. Addition of egg albumen made it possible to isolate active mito-
chondria. Recently Badran and Jones63 have obtained soluble phenol oxidase from bananas
by precipitating tannins with polyethylene glycol.
Lieberman isolated active mitochondria from apple fruit by removing the peel (which
is especially rich in phenolic compounds) and homogenizing the pulp at a pH of 8.4 to 9.9
with ascorbic acid added. Particles isolated at lower pH values were not active.
Jones and Hulme65 obtained highly active mitochondria from the peel of apples by
adding soluble PVP to the extraction medium to bind phenolic compounds. Subsequently
Hulme, Jones and Wooltorton have used this technique very successfully in further investiga-

55W. L. PORTER and J. H. SCHWARTZ, J. Food Sci. 27,416 (1962).

56 T. A. BELL,J. L. ETCHELLS, C. F. WILLLAMS and W. L. PORTER,B&m. Gaz. 123,220 (1962).
57 S. S~HH~~MMER, J. Biol. Gem. 232,715 (1958).
58 H. BOSER,Hoppe-Seyler’s Z. Physiol. Chem. 316,284 (1959).
59 H. BOSER, Planta Med. 9,456 (1961).
60 J. L. GOLDSTEIN and T. SWAIN,Phytochem. 4,185 (1965).
61 G. W. SANDERSON, Biochem. Biophys. Acta 92,622 (1964).
62J. M. TAGER,South African J. Sci. 54,324 (1958).
6) A. N. BADRANand D. E. JONES,Nuture 206,622 (1965).
64 M. LIEBERMAN, Plunt Physiol. 3S,7% (1960).
65 J. D. Jams and A. C. HULMJZ, Nature 191,370 (1961).
28
430 W. D. Loor&xsand J. BATTAILE

tions of mitochondria from apple@-70 and rose petal~.~l They also found6’ that use of
soluble PVP greatly increased the activity of soluble malic enzyme and pyruvic carboxylase in
extracts obtained from apple fruit. References 68 and 69 include detailed discussions of their
findings. They found that ordinary PVP contained impurities which made it ineffective;
pharmaceutical grade PVP was required. Best results were obtained when the pH of the
homogenate was 7 to 7.5. They found also that merely preventing the oxidation of phenolic
compounds did little good; it was necessary to remove them.
Barber and Hassid71a found that addition of soluble PVP made it possible to isolate
active cellulose-synthesizing particles from the cotton boll.
Phenolic compounds also make it difiicult to isolate viruses from many species of plants.
Thresh72 was able to restore infectivity in virus-phenolic mixtures by dilution, or by addition
of pH 8 buffer, nicotine sulfate, or alumina. Cadman has reported similar findings. Brunt
and Kenten found that infective virus could be extracted from cocoa leaves only if excess
protein was added to the medium. Either soluble protein or hide powder was effective, but
the authors preferred hide powder because of the ease of separating it from the extract.
Their data suggest that the amount of protein relative to the leaf weight is more important
than the concentration of protein in the extracting fluid.
Precipitation with tannins has been reported as a technique for purifying or concentrating
enzymes and other proteins. Acetone7s or caffeine76 has been used to recover the protein
from the tannin complex.

RESULTS
Preliminary Experiments
In connection with investigations of monoterpene biosynthesis we attempted to obtain
active enzymes from leaves of peppermint and other monoterpene-producing plants. Extracts
prepared from peppermint leaves by conventional techniques browned rapidly, and no active
enzyme, other than phenol oxidase, could be found in them. It was clear that the plant tissues
contained enzyme inhibitors, but the nature of these inhibitors was not known. It seemed
likely that they were phenols, quinones formed by phenol oxidase activity, or organic acids.
Many extraction and purification procedures were tested, most of them designed to remove
or neutralize these compounds and to prevent oxidation. Techniques tested included the use
of large amounts of buffer; addition of reducing agents, cyanide, and metal chelating agents;
decolorizing with charcoal; dialysis and gel filtration to remove compounds of low molecular
weight; addition of polyethylene glycol, soluble PVP, or albumen. These techniques were

66 A. C. HULMEand L. S. C. WOOLTORTON, Nature I%, 388 (1962).

67 A. C. HULME,J. D. JONESand L. S. C. WOOLTOR~N, Pruc. Roy. Sm. (London) Ser. B 1S8,S I4 (1963).
68 A. C. HULMEand J. D. JONES,In Enzyme Chemistry of Phenolic Compowtds(Edited by 5. B. PRIDHAM),
p. 97. Pergamon Press, Oxford (1963).
69 A. C. HULME,J. D. JONESand L. S. C. WOOLTOR~N, Phyrochem. 3,173 (1964).
70 J. D. JONES, A. C. HULKE and L. S. C. WOOLTOR~N, Phytochem. 3,201(1964).
71 A. C. HULME, J. D. JOW and L. S. C. WOOLTORTON, Nature 201,795 (1964).
718 G. A. BARBER and W. Z. HASSID,Nature 207,295 (1965).
‘2 J. M. THRESH.Ann. Appl. Biol. 44,608 (1956).
73 C. H. CADMAN, J. Gen. Microbioi. 20,113 (1959).
74 A. A. BRUNT and R. H. KENTEN,Yirdogy 19,388 (1963).
75 R. BENTLEY, In Me&& in Enzymology (Edited by S. P. COLOWXCK and N. 0. KAPLAN),WI. 1, p. 340.
Academic Press, New York (1955).
76 W. MEJBAUM~ATZENELLENB~~EN, W. DOBRYSZYCKA,J. BOGVSLAWSKA-JAWORSKA and B. MORAWIECKA,
Nature 184, 1799 (1959).
 Plant phenolic compounds and the isolation of plant enzymes 431

tested both on fresh tissues and on acetone-dried tissues, using glutamyl transferase” and
mevalonic kinase’s as test systems in most cases. Some of the procedures were helpful with
other species, but none were effective with peppermint. It was possible to prevent peppermint
extracts from browning, but none of the extracts had any convincing amount of enzyme
activity. Extracts prepared with albumen became considerably browner than other extracts,
suggesting that the albumen formed soluble complexes with the browning products.
It was found that addition of insoluble proteins (collagen or hide powder), plus buffer
and sodium ascorbate, during the homogenization of peppermint leaves, yielded extracts
which did not brown rapidly, and in which several active enzymes could be detected. An
extract prepared with 0.2 g of collagen per g of fresh tissue was shown to have mevalonic
kinase activity, and the same extract after gel filtration on Sephadex G-50 showed weak
glutamyl transferase activity. An extract prepared with O-2 g of “purified” hide powder per
g of tissue, and fractionated by gel filtration on Sephadex G-50, was shown to contain
glucose&phosphate dehydrogenase as well as weak glutamyl transferase activity. The solid
residue from each of these extraction procedures turned deep brown on standing in air. The
extracts themselves browned slowly.
It was evident that the insoluble proteins had removed substances which were strongly
inhibitory to the peppermint enzymes; it seemed likely that the inhibitors were the phenolic
compounds which are the substrates of the browning reaction. It was also clear that the
substances did not need to be oxidized in order to be inhibitory. Since the use of foreign
protein, even such presumably inert proteins as collagen and hide powder, introduces a possi-
bility of contaminating the extract, a synthetic polymer seemed preferable. Insoluble PVP
(Polyclar AT) was tested and found to be very effective.

Experiments with Insoluble PVP

Preliminary experiments showed that Polyclar AT adsorbed both browning substrates
and browning products from peppermint extracts. To determine the amount of Polyclar
required, peppermint leaves were extracted with water in the presence of 0.5, 1.0, I.25 and
1.5 g of the polymer per g of fresh tissue. With 0.5 g the extract was somewhat brown; with
1.0 g the extract was initially clear but there seemed to be very slight browning after 16 hr
at room temperature. With 1.25 or 1.5 g, the extract was clear and remained clear.
The extraction procedure which was then devised consisted of washing the leaves with
distilled water and grinding them in a mortar with liquid nitrogen, adding 1.5 g of Polyclar
for each g of fresh tissue, and mixing thoroughly by continued grinding. A solution of buffer
and sodium ascorbate was added slowly so that it froze, and was mixed by further grinding.
The mixture was then thawed, squeezed through bolting silk, and centrifuged. The extract
was further purified by gel filtration on Sephadex G-50. This extract contained mevalonic
kinase and also phosphomevalonic kinase. From 62 wmoles of ( + ) mevalonic acid,
15 mpmoles of phosphomevalonate and 1 mpmole pyrophosphomevalonate were formed
in 165 min. at 37”. Thirty per cent of the mevalonic acid, or 60 per cent of the physiologi-
cally active (+) isomer, was phosphorylated.
On further investigation we found that variable amounts of soluble PVP are formed from
Polyclar AT either by grinding in a mortar or by homogenizing in a blendor type of homo-
genizer. The amounts formed represent a very small fraction of the Polyclar (on the order of ’
0.02 per cent) but nevertheless are a substantial contaminant to the small amounts of protein
” W. D. LOOMIS,Plant Physid. 34,541 (1959).
‘a W. D. Loows and J. BA-ITAILE,
Eiochim. Biophys. Actu 67,54 (1963).
432 W. D. LOOMISand J. BATLULE

extracted from the plant tissues. The soluble PVP formed in this way was precipitated, at least
in part, by trichloroacetic acid, but the precipitate frequently redissolved. To avoid contamina-
tion of the extracts by PVP, the procedure was modified. Polyclar, buffer and sodium ascor-
bate are now mixed to form a thin paste, and chilled in an ice bath. Preferably this is allowed
to stand for several hours so that the Polyclar is fully hydrated. The washed plant tissue is
then ground with liquid nitrogen in a mortar, and the frozen powder is added to the Polyclar-
buffer-ascorbate slurry and stirred in gently. The liquid is expressed by squeezing through
bolting silk, and the residue is extracted again with ice-cold, glass-distilled water. Most of the
solid material is removed by the bolting silk; what remains is precipitated by centrifuging.
In order to test the extraction procedure, five extracts were made, with and without Poly-
clar and with and without reducing agents. A large amount of peppermint leaf tissue was
ground in liquid nitrogen, and the frozen powder was divided into equal samples by measuring
with a 5 ml beaker. The samples were extracted with various combinations of buffer, Polyclar
AT, ascorbate and 2-mercaptoethanol, as indicated in Table 1. A portion of each extract
was fractionated on Bio-Gel P-10 polyacrylamide gel to remove substances of low molecular
weight. Ten fractions were thus obtained: five original extracts and the high-molecular-
weight fraction from gel-filtration of each extract. These fractions were analyzed for protein
and for the enzymes mevalonic kinase, alkaline phosphatase, and glutamyl transferase.
The results, except for glutamyl transferase activities (which ‘were consistently very low are
shown in Table 1. Due to differences in the volumes of the extracts the amount of fresh

TABLE 1. PROTEM CONTENTAND ENZYME ACWITIIB OF E~TRACIXFROMPEPPERMINT

LEAVES

Protein in
extract Mevalonic Alkaline
Extract Extraction kinase phosphatase
(mt+f;y
additions* activity? activityt

crude
1 B, A 4.8 0.9 0
2 B, P 5.6 o-2 40
3 B, A, P 7.0 3.2 40
4 B, M, P 75 0 0
5 B, M 2.6 0.5 0
Gel filtered
1G B, A 1.8 0.2 35
2G B, P 6.2 0 55
3G B, A, P 4,9 2.7 75
4G B, M, P 3,8 0 47
5G B, M 1.9 0.4 11

* B = buffer ; P = Polyclar AT ; A = sodium ascorbate; M = mercaptoethanol.

t mpmoles of mevalonate phosphorylated per 10 mg fresh tissue in 180 min. The
reaction mixture contained: buffer, Tris plus maleate, 16 pmoles of each, KOH to give
pH 6.2; MnSO,, 0.5 pmole; ATP, 2 pmoles; mevalonic acid, 30 mpoles (100 m&;
enzymeextract, 0.1 ml forcrudeextracts, 0.15 mlforgel-fiheredextracts. Total volume,
0.16 ml. for crude extracts, 0.21 ml for gel-filtered extracts. Temperature 37”.
$ pmoles of pnitrophenyl phosphate hydrolyzed per 10 mg fresh tissue in 50 min.
The reaction mixture. contained: buffer, Tris, 3-O m-moles, acetic acid to give pH 8.0;
pnitrophenyl phosphate, 3 pmoles; enzyme extract 0.1 ml in all cases. Total volume,
3.1 ml. Temperature, 25”. Readings at 10 min intervals showed the reaction rates
were constant.
 Plant phenolic compounds and the isolation of plant enzymes 433

tissue represented in each assay ranged from 7.9 to 12.9 mg. For ease of comparison, all
results are expressed in terms of 10 mg of fresh tissue.
It is not clear whether the consistently low glutamyl transferase activity of peppermint
extracts is due to an actual low level of the enzyme, or to inadequacy of the extraction and
assay techniques. However, the extraction procedures definitely removed substances which
would interfere with the calorimetric determination of hydroxamic acid in the transferase
assay. Extracts 1 and 5 turned pink on addition of trichloroacetic acid, before the addition
of ferric chloride. These same extracts were intensely yellow at pH 8. None of the other
extracts behaved in this way, indicating that the substances responsible for the colours could
be removed either by gel filtration or by adsorption on Polyclar AT.
The alkaline phosphatase assay was repeated later, after the extracts had been stored for
3 months in the freezer, with occasional thawing. The results agreed generally with the
initial assay, except for a striking decrease of activity in extract number 3. It would appear
that the O-25 M ascorbate damaged proteins when left over such a long period of time. The
same concentration of mercaptoethanol was apparently harmful even in the hour or so
required for preparing the extracts. This may be due to exchange reactions which would
disrupt disulfide bridges of the protein.
Preliminary evidence indicates that peppermint extracts prepared with Polyclar AT contain
an enzyme or enzymes which act on geraniol. When 14C-geraniol is incubated with the
extract plus ATP a labeled water, soluble product is formed. From the ATP requirement,
and from RIvalues in paper chromatography,79 it appears that the new compound may be
geranyl pyrophosphate. Extracts which have been heated do not form the compound.

TABLE 2. EFFECT OF POLYCLA~ AT AND N-MEIHYLPYRROLIDONE

ON GLUTAMYL TRANSFERASEAClIVlTy OF A PUMPKIN SEEDLING
ACETIME POWDER

Additives
, 3 Hydroxamate
Polyclar Methyl- formed
pyrrolidone &moles)
(mg) (me)

0 0 4.5
0 103 (0.1 ml) 4.3
1 0 4.0
2 0 3.7
5 0 4-o
10 0 3.6
20 0 3.6
50 0 3.4
100 0 3.8

The reaction mixture. contained : buffer, potassium maleate,

pH 6.4, 100 pmoles; ghttamine, 17 pmoles; hydroxylamine
hydrochloride (neutralized with NAHCOI), 10 pmoles; MnSOI.
1 pmole; Na&AsO,, 20 pmoles; ADP, 0.2 pmoles; Polyclar AT
and methylpyrrohdone. as indicated. Total volume, before
adding Polyclar or methylpyrrolidone, 2.0 ml. Enzyme, 10 mg
of a pumpkin seedling acetone powder (equivalent to 130 mg
fresh tissue). Reaction time, 180 min; temperature, 35”.
79 F. CRAMER, W. RITTERSD~RF and W. B&M, Ann. Gem. 654,180 (1962).
434 W. D. LOOMISand J. Bmrm

As a general test for possible enzyme inhibition, pumpkin seedling glutamyl transferase
was assayed in the presence of varying amounts of Polyclar AT, and also of N-methylpyr-
rolidone. The results are shown in Table 2. It is clear that neither reagent has any great
inhibitory action on this enzyme, even when added in large amounts. The apparent reduction
of about 10 per cent caused by Polyclar AT is probably due to removal of phenolic compounds
which form interferring colored complexes with ferric ion. Polyclar treatment of pumpkin
extracts causes a decrease in light absorption in the 270 rnp region, as well as a slight reduction
in the visible color formed when FeCl, is added.
In preliminary experiments we have found that addition of Polyclar AT makes it possible
to extract soluble proteins from apple fruit and from Canada thistle leaves (Cirsiwn arvense).
Without Polyclar no protein could be extracted from apples, and very little from thistle
leaves.

DISCUSSION

Phenolic compounds combine with proteins reversibly by hydrogen bonding, and irre-
versibly by oxidation followed by covalent condensations. Techniques for isolating enzymes
from plants which contain phenolic compounds should then specifically separate the phenols
from the proteins, and at the same time prevent oxidation of the phenols.
Those phenols which form H-bonded complexes with protein are effectively removed by
adding large amounts of substances which contain groups simiiar to the peptide linkage. To
date, the most satisfactory agents have been various grades of PVP, soluble PVP for mito-
chondria, and insoluble PVP for soluble enzymes. Proteins and synthetic polyamides have
also been used successfully.
It is important to recognize, as Gustavson 20**l has emphasized, that the adsorption of
phenols from dilute solution by polyamides involves free, or accessible, amide groups. In
ordinary nylon most of the amide groups are tied up by internal hydrogen bonding. Poly-
amides synthesized from mixtures of monomers have irregular structure and, as a result,
reduced internal hydrogen bonding. Methylation of the amide nitrogen atom has a similar
eKect.80~81 Batzer and Weissenberger’8* I9 and Gustavson and Holm20S*l used a mixed
polymer nylon for their tanning experiments. They also hydrated the polymer by dissolving
it in hot methanol and pouring it into cold water, thus making the amide groups accessible.
Both groups of investigators found that the unmodified polymer as obtained from the
manufacturer adsorbed very little tannin.
In PVP there is no possibility of internal H-bonding; the polymer is a strong H-acceptor
but cannot act as an H-donor. This undoubtedly has much to do with its capacity for adsorb-
ing phenolic compounds; and with the high water solubility of most grades of PVP. Even in
PVP, however, there is apparently considerable possibility for blocking of amide groups.
An aqueous “paste” of Polyclar AT is reported to be at least 50 per cent more effective in
removing tannins than is the dry powder. 42 Gustavsonz3 soaked the dry powder in water for
24 hr before use, in order to hydrate it, and we also have found that thorough wetting of the
polymer requires some time.
The same H-bonding characteristics which make PVP effective in adsorbing phenols
should make it relatively inert with respect to possible interactions with protein. Singers2

80C. E. H. BAWN,The Chemistry of High Polymers. Intersciena, New York (1948).

81R. HILL,J. Sot. Dyers Colourists 68,158 (1952).
82 S. J. SINGER,
Advan. Prorein Chem. 17, 1 (1962).
 Plant phenolic compounds and the isolation of plant enzymes 435

has pointed out that proton-accepting solvents have little tendency to disrupt intrapeptide
H-bonding, whereas proton-donating solvents tend to cause considerable disruption. The
same principles should apply to PVP. The fact that soluble PVP has been used as a plasma
substitute in blood transfusions speaks for its inertness. Soluble PVP added during the
isolation of amphibian yolk platelets protects the structure of the platelets.83
It is common practice to add reducing agents such as ascorbate or thiols during the
extraction of enzymes from plant tissues, and these are clearly of some value. However, one
should realize that they do not actually prevent the oxidation of phenols; rather, they rapidly
remove any quinone that is formed, preventing its accumulation and thus reducing the
probability that it will react with protein. Ascorbate reduces the quinones,848 8s while thiols
also react with them to form thioethers.15* 29s31*86 I n one respect reducing agents can be
harmful. It has been shown that, in the presence of oxygen, reducing agents activate the
ortho-hydroxylatbon of monophenols by phenol oxidase.14* 84 The results of Wood and
Ingraham cited above are probably an example of this. Among the phenols that can be
hydroxylated in this way are tyrosyl residues of protein. 30-31*85**’ This would obviously
produce a modified protein; it would also introduce the possibility of “self tanning” of the
protein via quinonoid intermediates. Clearly the use of reducing agents is only an imperfect
substitute for actual inhibition of phenol oxidation.
The question of the optimum pH for extraction is important and needs further attention.
In the experiments described here we used buffers of pH 7.4 or 7.5 because they were satis-
factory in the past for extracting plant enzymes; but a somewhat lower pH would probably
be better when an agent such as PVP is used, in order to suppress ionization of phenols. The
ionized phenols cannot hydrogen bond with either protein or PVP, and at the same time they
are more readily oiidized than are the un-ionized forms. We have found active mevalonic
kinase in peppermint extracts made with either water or pH 6.4 maleate buffer in the presence
of Polyclar AT and ascorbate. Incidental observations from experiments with extracts from
pumpkin seedling acetone powders indicate the importance of pH. These extracts, which
were pale yellow, were decolorized by Polyclar AT after acidification with trichloroacetic
acid, or at the undetermined pH produced by extracting the powder with distilled water.
At pH 7.75 Polyclar removed no color. Data of Jones and Hulme65 showed slightly greater
succinic oxidase activity in apple peel mitochondria prepared at pH 7.15 with PVP than in
mitochondria prepared in the same way at pH 7.5.
It is probable that some phenolic compounds which do not form strong H-bonded
complexes with proteins or PVP, are readily oxidized to quinones. One would expect such
behaviour especially of catechol derivatives, and of free tyrosine. These compounds would
not be removed effectively by adsorption on PVP and would remain in the extract as latent
inhibitors. Gel filtration or dialysis should remove such materials, provided adequate pre-
cautions are taken to prevent oxidation. In this connection, it would be especially interesting
to test poly-N-methylacrylamide. This polymer, if appropriately cross-linked, would very
likely combine a capacity for gel filtration with a capacity to adsorb phenols. Poly-N,N-di-
methylacrylamide, with a structure analogous to PVP, might also be useful.
Organic solvents have been used to some extent in the isolation of enzymes, and they
undoubtedly have much unrealized potential for work with phenol-containing plant tissues.
(3 R. A. WALLACE and S. KARASAKI, J. Cell Biol. 18, 153(1963).
84M. ROLLAND and S. LMTZKY, Biochim. Biophys. Acta 56,83 (1962).
85 S. LIWTZKY and hf. ROLLAND, Bloc&n. Biophys. Actu 56,95 (1962).
86 S. ROSTON, J. Biol. Chem. 235, 1002 (1960).
8’ I. W. SIZER,Advan. Enzymol. 14, 129 (1953).
436 W. D. Lcmus and J. BATMILE

Poly-N-methylacrylamide

Several hydrogen-bonding organic solvents, in mixture with water, have been shown to be
more or less effective in stripping tannins from leather. “* s*, 89 Stripping ability of the
aqueous solvents increased in the order: methyl, isopropyl, ethyl, tert. butyl alcohols, methyl
cellosolve, methyl cellosolve acetate, acetone, dioxane. ** Only a few of the solvents, notably
methyl alcohol, were at all effective in the anhydrous state; water alone was ineffective.
Bendall and Gregorygo were able to obtain soluble phenol oxidase from acetone powders of
tea leaves, provided the acetone used in preparing the powders contained about 20 per cent
water. They suggested removal of tannins as a possible explanation.
Several other areas in the use of organic solvents should be explored. For example, it
might be advantageous to add reducing agents, or oxidase inhibitors, when making acetone
powders. Use of other solvents, particularly those which act as strong H-acceptors in the
formation of hydrogen bondsB2 (e.g. dimethylformamide, dimethylsulfoxide, N-methylpyr-
rolidone) might be of value. Butanol has been used to dissociate lipoprotein complexes,91 and
the techniques developed might be adaptable to protein-phenol complexes. We have found
that extraction of plant acetone powders with methanol or methylpyrrolidone removes
considerable amounts of pigment. We have not investigated this further, except to note that
peppermint acetone powders which had been extracted with methylpyrrolidone still browned
rapidly when moistened with water. They clearly contained active phenol oxidase as well as
phenol oxidase substrates. A recent review by SingerB2 summarizes interactions of proteins
with non-aqueous solvents.

EXPERIMENTAL
Materials
Peppermint plants were the Black Mitcham variety of Mentha pipe&a L., grown in the
greenhouse from the same clone we have used previously. 92 Enzyme extracts were made from
the tips, consisting principally of young leaves which were still expanding.
Polyclar AT (insoluble PVP) and N-methyl-2-pyrrolidone were obtained from General
Aniline and Film Corp. Dyestuff and Chemicals Division, 435 Hudson Street, New York,
N.Y. Polyclar AT was pursed essentially as described by McFarlane and Vader.43 It was
boiled for 10 min in 10% HCl and washed with glass-distilled water until free of Cl-. It
was then washed with acetone and dried. With one batch of Polyclar this washing removed
considerable soluble yellow material. With a later batch it did not appear that anything was
washed out. Methylpyrrolidone was purified by adding 5 drops concentrated HCl per liter

H. L. ELLWN and C. P. HALL,J. Am. Learher Chemists’ Assoc. 42,536

88H. B. MERRILL,D. H. CAMERON,
(1947). From Chem. Absrr. 42,246Od (1948).
139H. B. MERRILL, D. H. CAMERON, H. L. ELLLS~N and C. H. WRIXTON, J. Am. Leather Chemists’ Assoc. 44,
54 (1949). From Chem. Abstr. 43,404Od (1949).
90 D. S. BENDALLand R. P. F. GREGORY,In Enzyme Chemistry of Phenolic Compounds (Edited by J. B.
PRIDHAM),p. 7. Pergamon Press, Oxford (1963).
91 R. K. MORAN, In Methook in Enzymology (Edited by S. P. COLOWICK and N. 0. KAPLAN),Vol. 1, p. 25.
Academic Press, New York, (1955).
5~ J. BAI-~AILEand W. D. Loows, Biochim. Biophys. Actn 51,545 (1961).
 Plant phenolic compounds and the isolation of plant enzymes 437

and distilling in wzcuo, discarding the first 10 per cent of the distillate.42 The colorless re-
distilled methylpyrrolidone was stored in a refrigerator.
American standard hide powder was obtained from the Marshall Laboratory, Ridgway,
Pennsylvania. Before use it was boiled for 40 min in HzO. The insoluble residue was washed
thoroughly with water, acetone, chloroform, dilute base and dilute acid, and finally water,
and dried. This drastic “putication” procedure was used in order to minimize the likelihood
of contaminating the plant extracts with foreign protein or enzymes.
Collagen, from tendon, was obtained from Sigma Chemical Co., St. Louis, Missouri.
Before use it was washed several times with water and dried. Bio-gel P-10, cross-linked
polyacrylamide, was obtained from Bio-Rad Laboratories, Richmond, California. Sephadex
G-50, cross-linked dextran, was obtained from Pharpacia, Uppsala, Sweden. ( k ) [2 - 14C]
Mevalonic acid was obtained in the form of the N,N’dibenzylethylenediamine salt from
Tracerlab, Inc., Waltham, Massachusetts. It was converted to the sodium salt before use.‘*
14C-Geraniol was isolated and purified from rose geranium cuttings (Pelurgonium graveolens)
which had been exposed to 14C02.g3 ATP and ADP were obtained as sodium salts from
Pabst Laboratories, Milwaukee, Wisconsin. Glutamine was obtained from the Pierce
Chemical Co., Rockford, Illinois. Tris was Sigma primary standard grade tris(hydroxy-
methyl)aminomethane. Sodium ascorbate was also obtained from Sigma. p-Nitrophenyl
phosphate was obtained from Calbiochem, Los Angeles, California.
Analytical Methods
Mevalonic kinase was assayed as described previously.78 Briefly, 14C-mevalonate was
used as the substrate, and the reaction products were separated by paper chromatography.
The radioactive compounds were located on the paper by autoradiography with Kodak
single coated medical X-ray film, and radioactivity was counted by means of a Van-
guard Autoscanner 880 automatic chromatogram scanner. Counts were determined
from the recorder trace by cutting out the peaks and weighing them. Due to heavy loading
with buffer and other solutes the chromatograms streaked considerably. However, the
n-butyl alcohol-formic acid-water (77: 10: 13 by vol.) solvent system gave a clear separa-
tion of mevalonic acid from the phosphorylated derivatives in spite of the streaking,
and it was therefore used in routine assays. When separation of pyrophosphomevalonate
from phosphomevalonate was required, two-dimensional chromatography was used, with
tert. butyl alcohol-formic acid-water (40: 10: 16 by vol.) as the second solvent system.
Glutamyl transferase was assayed as described previously.77 The absorption of the ferric
ion-glutamyl hydroxamate complex was measured with a Bausch & Lomb Spectronic 20
spectrophotometer at 540 rnp.
Alkaline phosphatase was measured spectrophotometrically, using p-nitrophenyl-
phosphate as the substrate and measuring the formation of p-nitrophenol by following
absorbancy changes at 410 rnp, with a Bausch & Lomb Spectronic 20 spectrophoto-
meter.g4* gs
For protein determinations, tissues were extracted with 5 % trichloroacetic acid, or tri-
chloroacetic acid was added to extracts to a final concentration of 5 per cent. The precipi-
tate, or the insoluble residue, was analyzed by the micro-Kjeldahl method as described by
Steyermark.g6 The weight of nitrogen was multiplied by 6.25 to obtain the weight of protein.
93D. J. BAISTEDand W. D. LOOMIS,Unpublished data.
9* A. GARENand C. LEVINTHAL, Biochim. Biophys. Acta 38,470 (1960).
95 Data sheet on E. coli alkalinephosphatase, Worthington Biochemical Corp., Freehold, N.J. (1963).
9~5A. STEYERhMRK, Quantitative Organic Microanalysis (2nd Ed.). Academic Press, New York (1961).
438 W. D. Loohns and J. BATTALE

Preparation of Peppermint Extracts

AI1 procedures were carried out in the cold, either in an ice bath, or in a cold room or
refrigerated centrifuge at about 2”.
The extracts used for the experiment shown in Table 1 were prepared as follows: A large
amount of peppermint leaf tissue was ground in a mortar with liquid nitrogen, and a 5 ml
beaker was used as a measure to obtain equal samples of the frozen powder. One sample was
transferred to a screw-cap vial and weighed (2.3 g). The frozen tissue samples were added to
20 ml of ice-cold solutions containing different combinations of O-3 M potassium phosphate,
pH 7.5; 0.25 M sodium ascorbate; and O-25 M 2-mercaptoethanol; with or without 2.5 g
Polyclar AT. The various combinations are indicated in the Table. The extracts were squeezed
through 97 mesh bolting silk, made to 20 ml each, and centrifuged for 30 min at 35,000 g.
From each extract, 4.0 ml was fractionated by gel filtration on a 20 ml column of Bio-Gel
P-10 polyacrylamide gel, eluting with glass distilled water. Calibration of the columns showed
that large molecules started to appear after a forerun of 10 ml. It was arbitrarily assumed that
the protein fraction would be diluted 50 per cent, and hence a 6.0 ml fraction was to be taken.
However, the columns contracted slightly when the extracts were put through them, and
highly refractive solutes began to appear before 6.0 ml had been collected. This fraction was
therefore removed when the first drop of refractive material appeared, and varied between
5.3 and 5.8 ml.

Acknowledgements-This investigation was supported by research grants from the U.S. Public Health Service
(GM 08818 from the National Institute of General Medical Sciences) and the National Science Foundation
(GEl&O), and by a U.S. Public Health Service research career program award (K3-GM-17,064) from the
National Institute of General Medical Sciences. Part of the work was done during a program of research
participation for college teachers supported by the National Science Foundation. We are grateful to Alice J.
Burbott for assistance with the experimental work.