Comparison of starch branching enzyme I and II from potato
Ulrika Rydberg
1
, Lena Andersson
2
, Roger Andersson
2
, Per A
˚
man
2
and Ha
˚
kan Larsson
1
1
Department of Plant Biology, and
2
Department of Food Science, Swedish University of Agricultural Sciences, Uppsala, Sweden
The in vitro activities of purified potato starch branching
enzyme (SBE) I and II expressed in Escherichia coli were
compared using several assay methods. With the starch–
iodine method, it was found that SBE I was more active than
SBE II on an amylose substrate, whereas SBE II was more
active than SBE I on an amylopectin substrate. Both
enzymes were stimulated by the presence of phosphate. On a
substrate consisting of linear dextrins (chain length 8–200
glucose residues), no significant net increase in molecular
mass was seen on gel-permeation chromatography after
incubation with the enzymes. This indicates intrachain
branching of the substrate. After debranching of the

extender maize mutants that lack SBE II [5]. There is no
known mutant with reduced SBE I; however, the chain
length distribution in amylopectin was not significantly
affected in transgenic potato plants with a reduced level of
SBE I [6,7]. Interestingly, the physical properties of the
starch from transgenic potato with reduced SBE I levels are
clearly changed [6–8].
SBE I from potato was first characterized as having a
relative mass of 80/85 kDa [9,10]. In 1991 it was shown that
intact potato SBE I had a relative molecular mass of
103 kDa [11]. The active 80/85-kDa form present in potato
tubers was isolated and shown to have an almost intact
N-terminus and thus thought to result from proteolytic
cleavage in the C-terminal part [12]. Both intact SBE I and
the 85-kDa form have been shown to transfer chains from a
donor chain to an acceptor chain (interchain branching)
[13,14]. The occurrence of intrachain branching, i.e. transfer
within one and the same chain, could not be excluded in
those experiments.
Thorough studies of the activity of maize SBE I and II
isolated from endosperm [3] or expressed in Escherichia
coli [15,16] have been performed on various substrates.
Potato SBE II was first observed to be present as a granule-
bound protein in tuber starch [17]. SBE II seems to be less
abundant in potato tubers than SBE I and has not been
isolated from potato in amounts required for activity
analysis. Recently, however, both isoforms of potato SBE
have been expressed in E. coli [18–20], and the present
paper reports the activity of potato SBE I and SBE II with
amylose, amylopectin and linear dextrins as substrates.

Amylose (type III, Sigma) and amylopectin (Sigma) from
potato were typically dissolved at 10 mg
:
mL
21
in 0.5 M
NaOH. The solutions were buffered with 1 M KH
2
PO
4
and
pH adjusted to 7.5 with NaOH. The reaction mixtures
contained 0.6 mg
:
mL
21
substrate, 90 mM KH
2
PO
4
, and
0.01 m
M branching enzyme. Incubations were performed at
room temperature (22 8C), and aliquots were withdrawn at
several intervals between 5 and 180 min after the addition of
the SBE and terminated by heating at 95 8C for 5 min. A
100-mL sample of each aliquot was mixed with 900 mL
iodine solution (0.0125% I
2
and 0.04% KI, freshly made

dextrins and 50 mM Tris/HCl, pH 7.6. To
900 mL of this solution was added 100 mL1m
M branching
enzyme or water (control sample). The samples were
incubated at room temperature for 16 h, and the reactions
terminated by heating at 100 8C for 5 min. After addition of
150 mL1
M acetate buffer, pH 3.6, the samples (1 mL)
were debranched with 295 U isoamylase (Hayashibara
Biochemical Laboratories Inc., Okayama, Japan) for 5 h at
38 8C. Before injection on to a column, the reaction was
terminated by heating to 100 8C for 5 min, and the pH
adjusted to . 10 with NaOH as described by Andersson
et al. [21].
Chromatographic methods
GPC was conducted as previously described [22] using a
Sepharose CL-6B column eluted with 0.25
M KOH. The
relative amounts of carbohydrate in the collected fractions
were measured by the phenol/sulfuric acid method [23].
HPAEC-PAD and a CarboPac PA-100 column was used as
described by Koch et al. [24]. In this method, correction for
detector response is performed. All experiments were run in
duplicate with only small differences between the samples.
RESULTS AND DISCUSSION
Comparison of SBE I and SBE II on the amylose and
amylopectin substrates
To compare the activity properties of potato SBE I and
SBE II, commercially available amylose and amylopectin
were used as substrates in a kinetic study using the starch –

(Fig. 2A). Incubation overnight did not notably further
change the l
max
(data not shown). The difference in final
l
max
values and a comparison of the shapes of the two
spectra indicate that SBE I reduced more efficiently than
SBE II the long linear chains that mainly give rise to
absorbance above 600 nm. Similar differences between the
final l
max
values with amylose as a substrate were
previously observed with maize SBE I and II [3,15,16].
The l
max
values after incubation with the amylopectin
substrate shifted from 551 nm to 522 nm with SBE I, and to
538 nm with SBE II, after 180 min of incubation (Fig. 2B).
Similar values were obtained after incubation overnight (not
shown). These results differed from those with the maize
isoforms, which both reduced the l
max
from 530 nm to
about 490 nm [15,16]. Although this suggests that there may
be a difference between the enzymes from maize and potato,
the divergent results could also be due to a difference
between the substrates, with relatively long and linear chains
in potato amylopectin as indicated by the relatively high
l

(dashed lines), SBE II (dotted lines), or control sample without enzyme
(solid lines) in 90 m
M phosphate buffer. The vertical lines denote the
l
max
of the spectra.
Fig. 3. Effect of phosphate on the activity of SBE I and SBE II. The
delta absorbance of the amylose substrate (A), measured at 655 nm, and
the amylopectin substrate (B), mesured at 520 nm, after 120 min of
incubation with SBE I (K) or SBE II (Â) in increasing concentrations of
phosphate. Delta absorbance is defined as the difference between the
absorbance of the starch–iodine complex of the control and the
samples.
6142 U. Rydberg et al.(Eur. J. Biochem. 268) q FEBS 2001
activation was obtained for both isoforms at 15–20 mM
phosphate with both substrates, which is similar to that
reported for wheat SBE I and II [25]. A fivefold activation
by 10 m
M phosphate of potato and wheat SBE I has been
reported previously [25,26]. The effect of phosphate is
dependent on the buffer conditions [26], which could
explain the divergent results for potato SBE I. From the
studies performed by us and others, it cannot be excluded
that the observed stimulatory effect is a consequence of the
phosphate ions interacting with the substrate, and thereby
changing its structure, leading to enhanced enzyme
reactions. Further investigations are required to clarify this
and whether the effect of phosphate is of relevance in vivo.
Branching of linear dextrins
To obtain a more detailed comparison of the mode of action

intrachain branches. The discrepancies between the studies
may be explained by differences in molecular masses and
phosphorylation of the substrates [13] or by differences
between the enzymes used. The experiments of Viksø-
Nielsen et al. [13] and Borovsky et al. [14] were performed
in 50 m
M phosphate and 100 mM citrate, respectively. The
results shown in Fig. 4 were obtained in the absence of
phosphate, but the same elution pattern was obtained in the
presence of 90 m
M phosphate (data not shown). Thus it
seems that SBE can produce branches by both intrachain
and interchain branching, depending on external factors.
After debranching with isoamylase, the GPC elution
profiles were shifted to lower molecular masses compared
with the original substrate, showing that extensive branching
had taken place (Fig. 4B). A more pronounced effect was
seen for SBE I than for SBE II. It is notable that, for both
enzymes, essentially all high-molecular-mass material had
disappeared. For SBE I, the majority of the dextrins with a
dp greater than 60 were missing and for SBE II those greater
than 70. At the same time, the proportion of short chains was
slightly increased for both enzymes and some new chains
shorter than those in the original substrate were detected.
These results are in agreement with the results from the
starch–iodine assay. Similarly, the product of maize SBE II
contained a higher amount of the longest chains than the
SBE I product [4].
To obtain a more detailed picture of the individual chains
produced by the enzymes, quantitative analyses of the

branched by SBE II.
The mechanism of chain transfer for maize branching
enzymes has previously been investigated using reduced
amylose (chain length 405) as substrate. The study of maize
SBE I showed populations of transferred chains with a dp
of 11 –14 and 31 after debranching of the enzyme products
[4]. A more detailed investigation of the shorter chains
(, dp 34) produced by maize SBE I revealed an increase in
chains of dp 11–12 as well as of dp 6 [27]. Maize SBE II
has been shown to transfer shorter chains than maize SBE I,
and the most abundant chains were reported to be around
dp 9 by Takeda et al. [4], whereas Guan et al. [27] reported
an increase in chains of dp 6 –7 with a smaller peak at dp
10–12. In accordance with this, incubation with potato
SBE I and II generated chains of dp 6–9, in decreasing
concentrations, which has been shown to be a general
feature for amylopectin in potato [28]. Thus, it is possible
that during biosynthesis of amylopectin the branching
enzymes produce a fraction of very short chains which are
normally elongated by starch synthase III, as indicated by
the interesting results of Edwards et al. [29] and work by
Abel, as reviewed in Kossmann & Lloyd [8], showing that
the relative amount of dp 6 chains in amylopectin was
significantly higher in transgenic potato lines with reduced
levels of starch synthase III.
The presence of phosphate interfered with the chroma-
tography of the carbohydrates on the HPAEC column.
Therefore the samples shown here were incubated in a Tris-
buffer. However, samples incubated in a phosphate buffer
gave the same elution patterns (not shown). The absence of

We are grateful to Dr E. Johansson who expressed and purified the
starch branching enzymes used in our experiments. This work was
funded by the Swedish Foundation for Strategic Research and the
Swedish Farmer’s Foundation for Agricultural Research.
REFERENCES
1. Burton, R.A., Bewley, J.D., Smith, A.M., Bhattacharyya, M.K.,
Tatge, H., Ring, S., Bull., V., Hamilton, W.D.O. & Martin, C.
Fig. 5. Activity of SBE I and SBE II on linear dextrins analysed
with HPAEC. Products of linear dextrins incubated with SBE I (A),
SBE II (B) or a control sample without SBE added (C) in 50 m
M Tris
buffer. All samples have been debranched by incubation with
isoamylase. The bars represent relative amounts of individual chain
lengths for the different samples.
6144 U. Rydberg et al.(Eur. J. Biochem. 268) q FEBS 2001
(1995) Starch branching enzymes belonging to distinct enzyme
families are differentially expressed during pea embryo develop-
ment. Plant J. 7, 3–15.
2. Larsson, C T., Khoshnoodi, J., Ek, B., Rask, L. & Larsson, H.
(1998) Molecular cloning and characterization of starch branching
enzyme II from potato. Plant Mol. Biol. 37, 505– 511.
3. Guan, H.P. & Preiss, J. (1993) Differentiation of the properties of
the branching isozymes from maize (Zea mays ). Plant Physiol.
102, 1269–1273.
4. Takeda, Y., Guan, H.P. & Preiss, J. (1993) Branching of amylose by
the branching isoenzymes of maize endosperm. Carbohydr. Res.
240, 253–263.
5. Hizukuri, S. (1985) Relationship between the distribution of the
chain length of amylopectin and the crystalline structure of starch
granules. Carbohydr. Res. 141, 295–306.

mechanism of amylose branching by potato Q-enzyme. Eur.
J. Biochem. 62, 307–312.
15. Guan, H.P., Baba, T. & Preiss, J. (1994) Expression of branching
enzyme I of maize endosperm in Escherichia coli. Plant Physiol.
104, 1449–1453.
16. Guan, H.P., Baba, T. & Preiss, J. (1994) Expression of branching
enzyme II of maize endosperm in Escherichia coli. Cell. Mol. Biol.
40, 981–988.
17. Larsson, C T., Hofvander, P., Khoshnoodi, J., Ek, B., Rask, L. &
Larsson, H. (1996) Three isoforms of starch synthase and two
isoforms of branching enzyme are present in potato tuber starch.
Plant Sci. 117, 9–16.
18. Khoshnoodi, J. (1997) Starch branching enzymes in Solanum
tuberosum and Arabidopsis thaliana. PhD thesis, Swedish
University of Agricultural Sciences, Uppsala, Sweden.
19. Larsson, C T. (1999) Studies of starch branching enzyme A from
potato (Solanum tuberosum). PhD thesis, Swedish University of
Agricultural Sciences, Uppsala, Sweden.
20. Jobling, S.A., Schwall, G.P., Westcott, R.J., Sidebottom, C.M.,
Debet, M., Gidley, M.J., Jeffcoat, R. & Safford, R. (1999) A
minor form of starch branching enzyme in potato (Solanum
tuberosum L.) tubers has a major effect on starch structure:
cloning and characterisation of multiple forms of SBE A. Plant J.
18, 163–171.
21. Andersson, L., Rydberg, U., Larsson, H., Andersson, R. & A
˚
man, P.
(2002) Preparation and characterisation of linear dextrins and their
use as substrates in in vitro studies of starch branching enzymes.
Carbohydr. Polym. 47, 53–58.

amylopectin retrogradation of starches with different amylopectin
unit chain length distribution. Carbohydr. Polym. 42, 175–184.
29. Edwards, A., Fulton, D.C., Hylton, C.M., Jobling, S.A., Gidley, M.,
Ro
¨
ssner, U., Martin, C. & Smith, A.M. (1999) A combined
reduction in activity of starch synthases II and III of potato has
novel effects on the starch of tubers. Plant J. 17, 251–261.
q FEBS 2001 Comparison of SBE I and II from potato (Eur. J. Biochem. 268) 6145