First evidence for polyamine conjugation mediated by an enzymic activity in plants.

Abstract

An enzyme activity, found for the first time in plants, mainly located in the 22,000g supernatant of the crude extract of sprout apices of Helianthus tuberosus L. cv OB1 tubers, is able in vitro to covalently bind polyamines to endogenous substrates of different molecular weights. The major assay parameters, such as pH, dithiothreitol, and extract concentrations were optimized. The time course of the reaction, the dependence on putrescine concentration, its competition with histamine, the capacity to bind spermidine and spermine better than putrescine, the stability of the reaction product and analysis of the latter by sodium dodecyl sulfate polyacrylamide gel electrophoresis and thin-layer chromatography suggest that putrescine is linked to endogenous substrates by means of an enzyme reaction that shows some similarities with transglutaminase activities detected in animals. However, the activities of the crude extract and of a fraction eluted from a Sephadex G25 column were not affected by CaCl(2) concentrations lower or equal to 5 millimolar; 4 or 10 millimolar EGTA caused a very small reduction; higher concentrations of CaCl(2) and 15 millimolar or more of EDTA were inhibitory. N,N'-dimethylcasein was not recognized as a substrate. These data indicate that this activity also displays some characteristics which are different from those of animal transglutaminases.

Full text

Page 1

Plant Physiol. (1988) 87, 757-761
0032-0889/88/87/0757/05/$0 1.00/0
First Evidence for Polyamine Conjugation Mediated by an
Enzymic Activity in Plants1
Received for publication August 4, 1987 and in revised form March 10, 1988
DONATELLA SERAFINI-FRACASSINI*, STEFANO DEL DUCA, AND DARIo D'ORAZI
Dipartimento di Biologia evoluzionistica sperimentale, Universitia di Bologna,
Via Irnerio, 42, 40126 Bologna, Italy
ABSTRACT
An enzyme activity, found for the first time in plants, mainly located
in the 22,000g supernatant of the crude extract of sprout apices of
Helianthus tuberosus L. cv OB1 tubers, is able in vitro to covalently bind
polyamines to endogenous substrates of different molecular weights. The
major assay parameters, such as pH, dithiothreitol, and extract concen-
trations were optimized. The time course of the reaction, the dependence
on putrescine concentration, its competition with histamine, the capacity
to bind spermidine and spermine better than putrescine, the stability of
the reaction product and analysis of the latter by sodium dodecyl sulfate
polyacrylamide gel electrophoresis and thin-layer chromatography sug-
gest that putrescine is linked to endogenous substrates by means of an
enzyme reaction that shows some similarities with transglutaminase
activities detected in animals. However, the activities ofthe crude extract
and of a fraction eluted from a Sephadex G25 column were not affected
by CaCI2 concentrations lower or equal to 5 millimolar,4 or 10 millimolar
EGTA caused a very small reduction; higher concentrations ofCaCI2 and
15 millimolar or more of EDTA were inhibitory. N,N'-dimethylcasein
was not recognized as a substrate. These data indicate that this activity
also displays some characteristics which are different from those of
animal transglutaminases.
One of the reasons for the increasing importance assumed by
spermine, spermidine, and putrescine in biology is the role that
they play in modulating some enzyme activities (proteolysis,
protein phosphorylation, ornithine decarboxylation) or in inter-
acting with structural proteins (actin, myosin, fibrin and fibro-
nectin, tubulin) (2, 24, 26, 27). Polyamines, owing to their
chemical nature, can form hydrogen, ionic, or covalent linkages
with other molecules. In some cases, polyamines linked to par-
ticular proteins have been isolated (1, 2, 7, 11, 13, 23, 26, 27).
In animals, posttranslational covalent linkages of polyamines to
numerous proteins have been demonstrated. They are solely
catalyzed by transglutaminases (R-glutaminyl-peptide: amine- 'y-
glutamyl-transferase, EC 2.3.2.13) and form cross-linked and
noncross-linked complexes with two or one peptide-bound glu-
tamine residues respectively (12, 13, 28, 33; reviewed in Refs. 11
and 17). In plants, although various other types of conjugates
(e.g. with hydroxycinnamic acids) have been detected (21), these
covalently bound polyamine-protein complexes are practically
unknown. However, in activated parenchyma slices of Helian-
thus tuberosus tubers, bound polyamines seem to be mainly
linked to proteins (31), and we had preliminary data suggesting
the existence of a transglutaminase-like activity (22). Moreover,
'Work supported by a grant of Ministero Pubblica Istruzione.
when labeled putrescine was added to the culture medium,
radioactive complexes, extracted in particular phases of the cell
cycle from these synchronously growing cells, have been detected
in various SDS-gel electrophoretic bands (22).
In the present study, we examined the possibility that also in
plants an enzyme activity, perhaps a transglutaminase activity,
might catalyze the formation of such complexes. We used the
shoot apices of Helianthus sprouts, where many meristematic
cells divide actively, but asynchronously; therefore, ifthisenzyme
activity is present and expressed in certain phases of the cell
cycle, it should be easily detected.
MATERIALS AND METHODS
Plant Material and Enzyme Assay. The apices (3-4 mm ofthe
apical region) of the sprouts growing from tubers of Helianthus
tuberosus L. cv OBl were sterilized with sodium hypochlorite
(6% active C1) for 15 min and used fresh or after storage at
-80C. Apices were rapidly ground in an ice-cold mortar with
two volumes per g fresh weight of 50 mM Tris-HCl buffer (pH
8) containing 45 mm freshly prepared DTT. After filtration
through four layers of cheesecloth and avoiding as much as
possible contact with air, one part ofthe extract was used for the
enzyme assay, performed with some modification according to
Haddox and Russell (13). The other part was precipitated with
10% TCA (w/v) for protein determination. One hundred 11 of
the crude extract (containing more than 0.4 mg protein) or its
22,000g supernatant (or pellet) were added to 100 ,ul of 150 mM
Tris-HCl (pH 8.2), 50 Ml of either N,N'-dimethylcasein (1.5 or
30 mg ml-' in 50 mM Tris-HCl, pH 8) or of 50 mM Tris-HCl
(pH 8), and 50IAlof 1.2 mM putrescine labeled with 1,480 kBq
ml-' of [1,4(n)-3H]putrescine (833 TBq mol-', unless otherwise
indicated) (Amersham, UK). This 300 Ml assay mixture, having
a pH of 7.8, was incubated at 30°C for 30 min, was placed on
Whatman No. 3 MM filter paper discs after stopping the reaction
with 20% TCA (w/v), and then was filtered using a multifilter
apparatus (Millipore) and repeatedly washed with 5% TCA also
containing 0.2 M KCl. The dried filters were cut into small pieces,
put in vials containing 4.5 ml of scintillation cocktail (MP,
Beckman) and counted for 20 min in a Beckman LS 1800
counter.
Other extraction media were also tested, but were less effective:
0.25 M sucrose and Tris buffer containing different molarities of
DTT.
Other experiments were performed with 92.5 kBq per assay of
[1,4-'4C]putrescine or [1,4-'4C]spermidine or [1,4-'4C]spermine
(all having a specific activity of 4.37 TBq molP'), (Amersham,
UK). The effects of 5 mM histamine, or various concentrations
ofCaCl2, EDTA, EGTA, DTT, or ion deprivation by elution on
a Sephadex G25 column were also tested as well as different
amounts of plant extract, N,N'-dimethylcasein, and [3H]putres-
cine.
757

Page 2

SERAFINI-FRACASSINI ET AL.
To evaluate the formation of other types of linkages (ionic or
hydrogen bonds), at the end of the assay, 3 ml 20% TCA were
added to the mixture, which was then filtered, treated with 9 ml
1 M NaCl, and washed as described; alternatively, NaCl was
added directly to the assay mixture, which was then filtered and
the filters were treated with 20% TCA and washed. In another
experiment, the reaction product treated with 20% TCA was
filtered and washed with 30 ml of 5% TCA plus 0.2 M KC1 also
containing 5 mM putrescine.
All the reagents, unless otherwise indicated, were purchased
from Sigma. Each experiment was repeated at least twice with
different extracts, with 3 to 6 replicates for each assay. The data
are expressed as pmol [3H]putrescine (mg protein)-' 30 min-'
where proteins are only the endogenous ones (casein through
present was not considered). The isotopic dilution is reported in
the legends of tables and figures. The physiological stage of the
sprouts influenced the binding capacity ofthe extracts. Therefore,
the data from separate experiments may be somewhat different.
Polyamine Determination. Free polyamines in crude extracts
of the sprout apices were detected after dansylation (32) in the
5% TCA supernatant after centrifugation (3000 rpm, 10 min).
Polyamines were separated on TLC plates with a concentrating
zone (Merck). Mono- or bi-dimensional chromatographies were
eluted in cyclohexane:ethylacetate (3:2, v/v) and/or in chloro-
form:triethylamine (5:1, v/v). Fluorescence was measured in
scraped spots extracted with
conjugated polyamines of the assay product collected on filters
or precipitated twice with 10% TCA were either separated on
SDS-PAGE or hydrolyzed overnight with 6 N HCI at 120C. The
hydrolysate was taken to dryness, resuspended in 5% TCA,
brought to pH 8.5, dansylated, and chromatographed as de-
scribed. Radioactivity in the scraped spots was measured after
fluorescence decay.
Protein Determination. Proteins in the crude extract (or in the
22,000g supernatant and pellet) were measured on the TCA
precipitate, solubilized in
method (Bio-Rad Protein Assay) because DTT did not interfere
with this assay, as it did with the Lowry method (20).
SDS-PAGE. Assay products with labeled polyamines were
separated by gel electrophoresis with sodium dodecyl sulphate
(16). The entire lanes were cut into either Coomassie positive or
negative bands, and their radioactivity was counted after gel
dissolution.
Calcium Determination. Total calcium content in the crude
homogenate and 22,000g supernatant was measured by atomic
absorbance spectrophotometry (Varian S 11 spectrophotometer).
1 ml of anhydrous acetone. The
1 N NaOH, using the Bradford (6)
RESULTS AND DISCUSSION
Definition of the assay conditions. In order to optimize the
assay conditions, various parameters were checked and some
modifications ofthe method normally used with animal extracts
were adopted. It is known that oxidases are very active in
Helianthus tuberosus (3). In fact, the concentration of DTT in
the extraction buffer and in the assay mixture was critical to
prevent inactivation of the enzyme activity (Table I). If, in
addition to the concentrations reported in Table I, the same 10
or 30 mm concentration ofDTT was present in the crude extract
as well as in the assay mixture, the reaction was blocked (data
not shown). The polyamine-binding activity, recently found in
the storage parenchyma ofH. tuberosus tubers, could proceed in
the presence ofmuch lower concentrations ofDTT (2 mM) (30).
In animal systems, DTT is used, if at all, at similarly low
concentrations in the assay mixture (13, 15, 23).
The optimum pH was 7.8 at 30C, and a similar pH has been
used for tranglutaminase assays in animals (4, 10, 13). N,N'-
dimethylcasein, normally used as the exogenous acyl donor
substrate when assaying animal tranglutaminases, did not influ-
Table I. Putrescine-Binding Capacity as a Function ofDTT
Concentration
Various DTT concentrations were tested both in the crude extract and
in the assay mixture; 0.3 Mm [3H]putrescine and 200 MM cold putrescine
were added in the test tube. The averages were significantly different (1%
Student's t-test).
DTT
Concentration
In the
crude
extract
In the
assay
mixture
mm dpm30 min.-
mM
Bound [3H]Putrescine
pmolmgprotein-'
30min-'
0.165
1.026
0.093
dpm 30 mln
10
30
30
13
10
40
5,605 ± 598
45,997 ± 5,809
3,457 ± 396
Table II. Putrescine-Binding Capacity as a Function ofN,N'-
Dimethylcasein Concentration
The assay was performed for 30 min; 0.2 Mm [3H]putrescine and 200
Mm coldputrescinewere added in the test tube. Different letters indicate
averages significantlydifferent from each other(5%, Student'st-test).
N,N'-dimethylcasein
Bound [3H]Putrescine
tg test tube-'
Mgtest
dpm30 min'
pmol mgprotein-'
~~~~~~~30
1.321
1.284
1.520
1.231
0.536
0.363
min-'
0
52,712 + 9,874a
50,602 ± 5,828a
59,742 + 12,222a
48,601 ± 4,687a
21,743± 11,755ab
15,089 ± 7,597b
75
150
450
750
1,500
ence the activity of this plant enzyme at low concentrations. On
the contrary, at high concentrations, it significantly inhibited
putrescine incorporation (Table II) as was confirmed by an
experiment performed for 3 h. Casein at 750 or 1,500 ,ug test
tube-' caused an inhibition of 33 and 37%, respectively, com-
pared to untreated controls. It appeared not to be recognized as
a substrate by the plant enzyme, as confirmed by the absence of
radioactivity in the casein bands separated on SDS-PAGE after
the assay performed with the 22,000g supernatant (Fig. 1).
The calcium requirement was checked, since animal tranglu-
taminases are Ca2"-dependent. Some animal tranglutaminases
do not require exogenously supplied Ca2" (23, 25) and/or are
inhibited by high Ca2" concentrations (9, 14). The total endoge-
nous Ca2" content in the sprout apices was approximately 1 mM
in the crude extract filtered through cheesecloth and was diluted
to 300 Mm in the assay mixture. Two hundred Mm Ca2" is usually
sufficient to activate transglutaminases in animal systems (8, 17).
In our system, the addition of up to 5 mM CaCl2 either to the
crude extract or after its elution from a Sephadex G25 column
(Fig. 2 and inset) did not significantly affect activity.
Experiments were also performed with chelating agents. EDTA
markedly inhibited the activity of the crude extract at 15 mm or
more (Fig. 3) and EGTA, up to concentrations much higher (4
or 10 mM) than those of endogenous Ca2" which should reduce
the activity to zero, caused instead a very small but significant
inhibition (respectively, 16 and 17%). These data suggest that
Ca2` is not an absolute requirement for this activity.
Controls and Time Dependence. The controls, performed in
the absence of the plant extract, always gave a very low average
value: 710 dpm (30 min)-'. This value was not affected by the
presence of different molarities of dimethylcasein, DTT, and
putrescine, of CaC12 or its chelating agents, by different pH
758
PlantPhysiol.Vol.87, 1988

Page 3

ENZYME MEDIATED POLYAMINE CONJUGATION IN PLANTS
SUPERNATANT
-CASEIN
dpm
66
45
36
53,231
2,545
13,351
10,349
9,832
844
4,710
36
4,198
4,792
12,514
2,481
29
20
1,460
789
14,2
1,587
i
..
408
0
0
+ CASEIN
dpm
19,015
4,170
25,01 2
7,804
6,057
6,206
7,222
2,2 87
90
1,382
2,494
1,426
97
1,440
3 031
CAS
3,633
2,756
0
FIG.
1. SDS-PAGE pattern and amounts of bound radioactivity, ex-
pressed as dpm per mg total protein present in the products obtained
after the enzyme assay. The experiment was performed with the super-
natant in the presence of 0.6 ,M
putrescine, with or without 750 ,ug N,N'-dimethylcasein (test tube)-'.
[3H]putrescine and 200 gM
cold
'
0
O\
5
10
nj
t ~~~aCi, ImM)
0
10
20
30
40
50
CaCIC(mM)
FIG. 2. Percentage inhibition of the putrescine-binding capacity of
the crude extract as a function ofCaCl2 concentration added to the assay
mixture. 0.4 Mm [3H]putrescine and 200AM cold putrescine were added
to each test tube. (0), Crude extract; (A), crude extract eluted from a
Sephadex G25 column.
values, nor by the addition of a previously boiled extract. It also
remained constant at various intervals during the time course of
the assay, as shown in Figure 4, where the time dependence of
the activity in the crude extract is also reported.
The enzyme activity remained substantially unaltered when
the crude extract was stored, while avoiding contact with air, at
various temperatures (-80C, -20C, +4C, +20°C) for 24 h.
Distribution of the Polyamine-Binding Activity. The polya-
mine-binding activity was mainly located in the 22,000g super-
natant of the crude extract (65%). Animal transglutaminase
activity is also mainly localized in the supernatant obtained from
centrifugation at various speeds (from 12,000-105,000g) (5, 8,
>.k
0(Us
050-
0
30
60
90
EDTA (mM)
FIG. 3. Percentage inhibition of the putrescine-binding capacity of
the crude extract as a function of the EDTA concentration in the assay
mixture. 0.4 MM [3H]putrescine and 200jAM cold putrescine were added
to each test tube.
8 100 _
Time of Incubation (mi)
FIG. 4. Putrescine-binding capacity ofthe crude extract as a function
of time. 0.4 ,uM (3H]putrescine and 200 MM cold putrescine were added
to each test tube. (-), Crude extract; (A), control (performed without
crude extract). Pu, putrescine.
15). In addition, Juprelle-Soret et al. (14) found a transglutami-
nase activity in rat liver crude homogenates and another in
lysosomes. Boiling the crude extract, its 22,000g supernatant, or
its precipitate for 10 to 20 mmn destroyed all capacity to induce
the binding of putrescine, thus demonstrating the reaction was
enzyme catalyzed. The addition of an equal volume of boiled
supernatant to the crude extract reduced the binding activity by
30% (with respect to a control consisting of equal volumes of
unboiled supernatant and extraction buffer). This can be due to
the presence of competitive amines in the supernatant or to
inhibitory oxidation products formed as a result of boiling.
Dependence of the Activity on the Amine Substrate Concen-
tration. An analysis of the data shows that the activity was
dependent on the putrescine concentration in the assay mixture
(Fig. 5, inset). Plotting activity versus low putrescine concentra-
tions yielded a sigmoidal curve (Fig. 5, unconnected square
symbols). This suggests the enzyme is cooperative, nevertheless,
the interpolation of the data based on a Hill plot gave a poor fit.
In fact, at putrescine concentrations lower than 0.5 mM, the
curve resembled more a hyperbola than a sigmoid. This could
be explained by the presence of two enzymes, or two forms of
one enzyme, contributing to the total activity, one with high
affinity and low velocity hyperbolic kinetics (component A,
evident at low concentrations) and the other with low affinity
and high velocity sigmoidal kinetics (component B, evident at
high concentrations).
759

Page 4

SERAFINI-FRACASSINI ET AL.
[Pu](mM)
FIG. 5. Putrescine-binding capacity ofthe crude extract in relation to
putrescine concentration (endogenous putrescine concentration in the
crude extract was 7jAM:[3H]putrescine concentration was 0.4 MuM). (5),
Experimental data. A and B, curves for components A and B, calculated
as described in the text; inset, overall curve; Pu, putrescine.
As described in Segel (29), it was assumed that component B
can be neglected at low substrate concentrations and a Line-
weaver-Burk plot was used to obtain the apparent kinetic param-
eters of component A (Km = 0.21 mm and Vm. = 2.4 nmol mg
protein-' h-').
With these theoretical parameters, the curve for component A
was constructed (Fig. 5, curve A). The values on this curve were
then subtracted from the overall activity curve to give the curve
for component B (Fig. 5, curve B) whose values were used to
construct a Hill plot. Based on the latter, the following were
calculated: [S]0.s = 22 mm and n = 2, confirming a positive
cooperativity. The correlation coefficients for curves A and B
were 0.99 and 0.95, respectively.
For animal transglutaminases, Km values range from 0.09 mM
(19) to 0.7 mm (14) frequently with N,N'-dimethylcasein as the
protein substrate. Furthermore, the existence of multiple forms
ofthe enzyme has also been reported in guinea pig prostate and
sea urchin egg (17). In evaluating these data one must remember
that in Helianthus, putrescine may be bound to more than one
endogenous substrate and not to casein, as shown by SDS-PAGE
(Figs.
Specificity of the Activity for the Amine Substrate.The trans-
glutaminases ofanimal origin are known to catalyze the binding
of many amines other than polyamines (18). In extracts of H.
tuberosus sprout apices, 5 mm histamine, a competitive inhibitor
of transglutaminase, caused a 64% inhibition of the activity.
Similarly, Paonessa et al. (23) observed in rat sperm a 79%
inhibition of spermine conjugation at the same histamine mo-
larity.
To verify the specificity toward various polyamines, the bind-
ing efficiency for equimolar concentrations of [14C]-putrescine,-
spermidine, or -spermine were compared. Spermine and sper-
midine were much more efficiently linked than putrescine. The
relative binding activities were; putrescine 34, spermidine 100,
and spermine 87, while in guinea pig liver transglutaminase, the
corresponding values were 38, 100, and 92, respectively (23).
Analysis of the Complexes. Putrescine, or its derivatives, can
also bind to proteins by ionic linkages as demonstrated by
Balestreri et al. (2) in Medicago sativa leaves where protease
activity was inhibited by binding of spermine to the enzyme.
This inhibition, due to a spermine-induced conformational
change of the protease molecule, was prevented by the ionic
strength of a 50 mm Tris buffer.
To verify the occurrence ofsuch ionic linkages in H. tuberosus
1 and 6).
(RUJDE EXTRAC t
dpnM
MW.#t
:'.kd rzC41-1
').39C.
)}0
10,49'"
,,
.50O57
11,533
18.980
26,952
45
FIG. 6. SDS-PAGE and amounts of bound radioactivity, expressed
as dpm per mg total protein present in the products obtained after the
enzyme assay. The experiment was performed with the crude extract in
the presence of0.6 MM [3H]putrescine and 200 Mm cold putrescine.
Table III. Putrescine-Binding Capacity as a Function ofCold
Putrescine Washing
The assays were performed at three different pH values and the
resulting products, collected on the filters, were washed, as indicated,
with 100 volumes 5 mm cold putrescine; 0.3 Mm [3H]putrescine and 200
Mm cold putrescine were added in the test tube. At each pH value, the
average of incorporation in the samples washed with or without putres-
cine was never significantly different one from the other (5%, Student's
t-test).
pH of the
Bound [3H]Putrescine
AssayMixture
dp 0min-'
dpm 30min~'
6.6
With Pu
1,387 ± 405
Without Pu
1,411 ± 151
7.8
With Pu
61,713 ± 9,937
Without Pu
59,180 ± 16,559
8.0
With Pu
34,516 ± 2,264
Without Pu
27,768 ± 5,660
pmol mgprotein-'
30 min'
0.007
0.007
0.845
0.810
0.356
0.263
extracts, the reaction product, previously precipitated with TCA,
was washed with NaCl. There was no significant decrease in the
amount ofradioactive complexes recovered on the filters. How-
ever, if NaCl was added prior to filtering and TCA treatment,
considerably fewer labeled complexes (22%) were recovered, but
this was due to the fact that, in this way, some of the labeled
proteins pass through the filters (50% of those present in the
assay mixture compared to 1 to 5% in the controls not washed
with NaCi).
In addition, when the reaction products (obtained from assays
performed with mixtures having three different final pH values)
were washed with cold putrescine at high concentrations (25-fold
that ofthe assay mixture), no further release oflabeled putrescine
was observed (Table III).
Acid hydrolysis, which destroys the linkage of polyamines to
760
PlantPhysiol.Vol.87,1988

Page 5

ENZYME MEDIATED POLYAMINE CONJUGATION IN PLANTS
partnermolecules in the labeledcomplexes,and the subsequent
comigrationof labeledcompoundswith cold standard polyam-
ines on mono- and bi-dimensional TLC plates allowed us to
determine that most of the label in the benzene phase was still
putrescine, whereas spermidine and spermine together repre-
sentedonly 1
putrescine derivatives formed during the assay procedure.
Theelectrophoretic pattern
usingeither the crude extract or its supernatant exhibited two
labeledregions (Figs. 1
substrates that were either too heavyor too insoluble to enter
thestacking geland others which entered the running gel and
had a mol wthigherthan 92.5 kD.
that 40% of thelabel,aftertransglutaminase-mediatedbinding,
associated withproteinsthat could not enter the gel. Birckbichler
et al.(5) suggestedthat rat liver transglutaminaseitself, which
has ahighmolwt, mayalso act as the acceptor substrate. In the
secondregion,labeled bandscorrespondingto molecules having
a mol wt ofapproximately 30 kD or less were observed. As
expected,freeputrescinewas notpresentin the gel. Finally, SDS-
PAGE was carried out on a mixture oflabeled putrescine previ-
ouslyincubated with standard reference proteinsused as mol wt
markers. Neither labeledbands nor freeputrescinewere detected,
demonstratingthatputrescine bindingis dependent on the en-
zyme activity present in the plant extract.
All
covalentlylinkpolyaminestoendogenoussubstrates exists and
that it has anefficiencysimilar to that of a mouse liver crude
extract (that we preliminarily measured). This plant enzyme
activityshows some similarities with mammalian transglutami-
nases, eventhoughthe fact that it does not recognize N,N'-
dimethylcaseinas a substrate and that it does not require exog-
enousCa2+ mayindicate that this activity is different from that
of animal transglutaminases which, incidentally, are the only
knownenzymesable tocatalyzethe covalent binding of poly-
amines. TherequirementforexogenousCa2", however, may not
be adiscriminatingfactor since some unpurified animal trans-
glutaminasesareindependentofadded Ca2+ (23, 25). In the final
analysis, onlythe identification ofglutamyl-polyaminederiva-
tives willprove beyond anydoubt the existence in plants of a
transglutaminase activity.
to 2%; theremaininglabel was probably due to
of the reaction products obtained
and6).In thefirst,there were endogenous
Similarly, Folk(1 1) found
these data indicate that,in H. tuberosus, the capacity to
Acknowledgments-Theauthors acknowledge Prof. G. Williams-Ashman for
the kindsuggestions,the ResearchGroupof Prof. F. Autuori ofthe Dipartimento
diBiologia,2ndUniversityof Rome for technical advice, Prof. P. Pupillo of the
Dipartimento di Biologia, University ofBologna, and Prof. R. Federico of the
Dipartimento di Biologia Vegetale, 1st University "La Sapienza" of Rome for
fruitful discussions and Prof. E.Carpene of the Istituto di Biochimica (Veterinary
Faculty), UniversityofBologna,for the Ca2" determination. We acknowledge Mr.
N. Mele for technical assistance.
LITERATURE CITED
1. AMBRONRT,LT KREMZNER 1982 Post-translational modificationofneuronal
proteins:evidence fortransglutaminase activityin R2, the giant cholinergic
neuron inAplysia.Proc Natl Acad Sci USA 79: 3442-3446
2. BALESTRERI E,PCIONI,AROMAGNOLI,S BERNINI, A Fissi, R FELICIOLI 1987
Mechanism ofpolyamine inhibition of a leaf protease. Arch Biochem
Biophys 255: 460-463
3.BASTIN M,0UNLUER1971 Studyon inactivation of enzymes in Jerusalem
artichoke tuber slices. Can J Bot 50: 727-731
4. BENINATIS,MPIACENTINI,MPARGENTO-CERU', S Russo CAIA, F AUTUORI
1985 Presence ofdi- andpolyamines covalentlybound to proteinin rat liver.
Biochim Biophys Acta 841: 120-126
5. BIRCKBICHLER PJ, GR ORR, MK Patterson Jr 1976 Differential transglutami-
nase distribution in normal rat liver and rat hepatoma. Cancer Res 36:
2911-2914
6. BRADFORD MM 1976 Arapid and sensitive method for the quantitation of
microgram quantitiesofprotein utilizingtheprinciplesof protein-dye bind-
ing.Anal Biochem 72: 248-254
7. CANELLAKIS ZN, PK BONDY, AA INFANTE 1985 Spermidine is bound to a
unique protein inearlysea urchinembryos. Proc Natl Acad Sci USA 82:
7613-7615
8. CHUANG DM 1984
,8-Adrenergic receptorinternalization and processing: role
oftransglutaminaseandlysosomes.Mol Cell Biochem 58: 79-89
9. CLARKE DD,MJ MYCEK, A NEIDLE, H WAELSCH 1959 The incorporation of
amines intoprotein.Arch Biochem Biophys 79: 338-354
10. DELCROSJG,AM
transglutaminase by a-difluoromethylornithine.FEBS Lett 171: 221-226
11. FOLK JE 1980Transglutaminases.Annu Rev Biochem 49: 517-531
12. FOLKJE, SI CHUNG 1973 Molecular andcatalytic propertiesof transglutami-
nases. AdvEnzymolRelat Areas Mol Biol 38: 109-191
13. HADDOX MK, HDRUSSELL1981 Increased nuclear conjugated polyamines
andtransglutaminase duringliverregeneration.Proc Natl Acad Sci USA 78:
1712-1716
14. JUPRELLE-SORET M, SWATTIAux-DE CONINCK, R WATTIAUX
ofatransglutaminase activityin rat liver lysosomes.Eur J Cell Biol 34: 271-
274
15. KORNERG,U BACHRACH 1987 Intracellulardistribution ofactive and inactive
transglutaminasein stimulated cultured C6 gliomacells. J Cell Physiol 130:
44-50
16. LAEMMLI UK 1970Cleavageof structuralproteinsduring the assembly of the
head ofbacteriophageT4. Nature 227: 680-683
17. LORAND L, SM CONRAD 1984Transglutaminases.Mol Cell Biochem 58: 9-
35
18.LORANDL, KN PARAMESWARAN, P STENBERG, YS TONG, PT VELASCO, NA
JONSSON,LMIKIVER,PMOSES
taminase for amine substrates. Biochemistry 18: 1756-1765
19. LORANDL,LKCAMPBELL-WILKES,L COOPERSTEIN 1972 A filter paper assay
for transamidating enzymes using radioactive amine substrates. Anal
Biochem 50: 623-631
20.LOWRYOH,NJ ROSEBROUGH,ALFARR,RJ RANDALL 1951 Protein measure-
ment with the Folinphenol reagent.J Biol Chem 193: 265-275
21. MARTIN-TANGUY J 1985 The occurrence andpossiblefunction of hydroxycin-
namoylacid amides inplants.Plant Growth Regul 3: 381-399
22.MOSSESI= U,DSERAFINi-FRACAssINI,S DELDUCA,D'ORAZI, 1987 Conjugated
polyamines during dormancyand activation oftuber ofJerusalemartichoke.
In KSchreiber,HRSchutte,GSembdner, eds, ConjugatedPlant Hormones,
Structure Metabolism and Function. Deutcher Verlag der Wissenschaften,
Berlin,GDR pp 369-375
23.PAONESSA G,S METAFORA,GTAJANA,P ABRESCIA, A DE SANTIS, V GENTILE,
R PORTA 1984 Translutaminase-mediated modifications of the rat sperm
surface in vitro. Science 226: 852-855
24. POHJANPELTO P, I VIRTANEN, E HOLTTA
disappearanceofactin filaments and microtubules in polyamine-auxotrophic
CHO cells. Nature 293: 475-477
25.RICERH,H GREEN 1978 Relation ofprotein synthesisand transglutaminase
activityto formation of the cross-linked envelope during terminal differen-
tiation of the cultured humanepidermal keratinocyte. J Cell Biol 76: 705-
711
26. ROCH AM, GQUASH,HRIPOLL,M MARDON 1983 Differences in polyamine
availabilityand insertion into fibronectins released from normal and trans-
formed cells. Biochem J 210: 137-144
27. RUSSELLHD 1981 Posttranslational modification of ornithine decarboxylase
byitsproduct putrescine.BiochemBiophysRes Commun 99: 1167-1172
28. SCHRODEJ, JE FOLK 1978Transglutaminase-catalyzed cross-linkingthrough
diamines andpolyamines.J Biol Chem 253: 4837-4840
29. SEGEL IH 1975EnzymeKinetics. JWileyand Sons,Chichester, pp 64-71
30. SERAFINi-FRACAssINi D, S DELDUCA, D D'ORAZI, U MOSSETTI 1988 Conju-
gationofpolyamines byanenzyme activityin dividing cells of Helianthus
tuberosus. In APerin, DScalabrino, A-Sessa, EM Ferioli, eds, Perspective
inPolyamineResearch.WichtigEditore, Milano. In press
31. SERAFINi-FRACAssINi D, U MossETrI 1985 Free and bound polyamines in
differentphysiological stagesof Helianthus tuberosus tuber. In L Selmeci,
MEBrosnan,NSeiler, eds, Recent Progress in Polyamine Research. Aka-
demiai Kiado', Budapest, pp 551-560
32. SMITHTA,GRBEST 1977Polyaminesinbarley seedlings. Phytochemistry16:
841-843
33. WILLIAMS-ASHMAN HG,YWILSON,REBEIL, L LORAND 1977 Transglutami-
nase reactions associated with the rat semen clotting system: modulationby
macromolecular polyanions. Biochem Biophys Res Commun 79: 1192-
1198
ROCH, GQUASH 1984 The competitive inhibition of tissue
1984 Presence
1979Specificityofguinea pig liver transglu-
1981 Polyamine starvation causes
761