Carbohydrate Research 352 (2012) 137–142

Contents lists available at SciVerse ScienceDirect

Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres

De novo biosynthesis of starch chains without a primer and the mechanism

for its biosynthesis by potato starch-synthase
Rupendra Mukerjea, John F. Robyt ⇑
Laboratory of Carbohydrate Chemistry and Enzymology, Department of Biochemistry, Biophysics, and Molecular Biology, Iowa State University, Ames, IA 50011, USA

a r t i c l e i n f o a b s t r a c t

Article history: The starch-synthase enzymes used in this study were the second acetone precipitate and Fractions 21 and
Received 4 October 2011 23, Table 1, [Mukerjea, Ru.; Falconer, D. J.; Yoon, S.-H.; Robyt, J. F. Carbohydr. Res. 2010, 345, 1555–1563].
Received in revised form 22 November 2011 Fractions 21 and 23 had high specific activities of 544 and 944 International Units/mg, respectively. When
Accepted 25 November 2011
the enzymes and buffer and substrate were treated with immobilized a-amylase and glucoamylase for
Available online 9 December 2011
30 min, they all had the same activity, before and after treatment, indicating that the enzymes were free
of putative primers and synthesized amylose chains de novo, without the addition of primers. Starch-
Keywords:
synthase was immobilized and reacted with ADP-[14C]Glc; the immobilized enzyme was removed,
Starch-synthase
De novo biosynthesis
washed and treated at pH 2 and 50 °C for 30 min, giving the release of 14C-D-glucopyranose and 14C-amy-
Pulse and chase reactions lose, showing that during catalysis they were covalently attached to the enzyme active-site. Pulse and
Synthesis from the reducing-end chase reactions of starch-synthase with ADP-[14C]Glc and ADPGlc, respectively, followed by reduction
Covalent amylose and D-glucose enzyme and acid hydrolysis of the starch-chain product, gave 14C-D-glucitol from the pulse reaction and a signif-
intermediates icant decrease of 14C-D-glucitol from the chase reaction, showing that the addition of D-glucose from
Two catalytic-site insertion mechanism ADPGlc was to the reducing-ends of the growing amylose chains. Reactions of four different concentra-
tions of starch-synthase, with constant ADPGlc concentration and temperature, gave four amylose chains,
each with different number average molecular weights that were inversely proportional to the concentra-
tion of the enzyme, indicating that the synthesis was processive. From the results, a two catalytic-site,
insertion mechanism is proposed for the biosynthesis of starch chains.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction
We previously studied the mechanism for the biosynthesis of
starch chains, using starch granules from eight different plant
sources that had active starch-synthase [EC 2.4.1.2] (a-1?4-D-glu-
can synthase) and starch-branching enzyme [EC 2.4.1.18] (a-1?4-
D-glucan, a-1?6-D-glucan transferase) entrapped inside the starch
granules.1–3 The first set of experiments involved the pulsing of the
starch granules with ADP-[14C]Glc and chasing with nonlabeled
ADPGlc. After isolation and reduction of the pulsed and chased
starch chains, followed by reduction with NaBH4, acid hydrolysis,
and descending paper chromatography, 14C-labeled D-glucitol,
resulting from the reduction of the reducing-end D-glucose residues
of the synthesized pulsed glucans, was obtained, with the amount
of [14C]-D-glucitol decreased in the chase reactions. This indicated
that the synthesis was occurring by the addition of D-glucose to
the reducing-ends of the growing chains, because if primers were
required for the synthesis of the starch chains,4,5 the D-glucose res-
idues would have been added to the nonreducing-ends of the prim-
ers, and it would have been impossible to obtain [14C]-D-glucitol
⇑ Corresponding author. Tel.: +1 515 294 1964; fax: +1 515 294 0453.
E-mail address: jrobyt@iastate.edu (J.F. Robyt).
from the pulse and a decrease in [14C]-D-glucitol from the chase
reactions.1 We now have studied the mechanism of the biosynthe-
sis of starch chains by using highly purified, solubilized, starch-syn-
thases that were isolated and purified from white potato tubers.6
We have used immobilized a-amylase and glucoamylase to
show that soluble starch-synthase and the buffer and substrate
were free of any primers and that starch-synthase synthesizes
amylose chains de novo, without the presence or addition of prim-
ers. It also is shown that D-glucose and a growing glucan chain are
covalently attached to the active-site of starch-synthase, during
catalysis, and that the polymerization is processive and is termi-
nated by hydrolysis of the amylose-chain from the active-site of
the enzyme. It is proposed that the de novo synthesis from the
reducing-end occurs by a two catalytic-site, insertion mechanism.
2. Experimental
2.1. Materials
Adenosine 50 -diphospho-a-D-glucopyranose (ADPGlc); dithio-
threitol (DTT); polyvinyl alcohol 50 k; Immobead 150, cross-linked
methacrylamide/epoxide beads7; and POP and PoPoP were obtained
from Sigma-Aldrich Chemical Co. (St. Louis, MO). ADP-[14C]Glc

0008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.11.024
138 R. Mukerjea, J. F. Robyt / Carbohydrate Research 352 (2012) 137–142
(333 mCi/mmol) was obtained from American Radiolabeled Chemi-
cals, Inc., 101 Arc Drive, St. Louis, MO 63147. Liquid scintillation
cocktail was prepared, containing 5.0 g PPO and 0.1 g PoPoP, in
1.0 L of toluene. All chemicals were of the highest grade obtainable.
Bacillus amyloliquefaciens a-amylase [EC 3.2.1.1] was crystal-
lized from Miles Laboratories (Elkhart, IN) HT-concentrate, using
the procedures of Stein and Fischer.8 Rhizopus sp. glucoamylase
[EC 3.2.1.3] was obtained from Megazyme International, Ltd. Wick-
low, Ireland. Starch-synthase Fractions 21 and 23 (Table 1, Ref. 6)
are highly purified enzymes, with specific activities of 544 and
944 IU/mg, respectively, as previously presented in the isolation,
fractionation, and purification of white Idaho Russet potato tuber
starch synthesizing enzymes.6
2.2. Methods
2.2.1. General
(1) Standard Buffer (pH 8.5), contained 10 mM glycine, 2 mM
EDTA, 0.04% (w/v) polyvinyl alcohol 50 K, and 1 mM DTT.6 (2) All
of the glasswares were treated with chromic acid before use in
the experiments.
2.2.2. Determination of whether starch-synthase does or does
not require a primer for the synthesis of starch chains
2.2.2.1. Immobilization of a-amylase and glucoamylase. (1)
Immobead 150 (617 mg) was suspended in 3 mL of 40 mM imidaz-
ole-HCl buffer (pH 6.2) and allowed to swell at 4 °C for 15 h. (2) The
beads were filtered on a 0.2 l membrane filter and washed with
6 mL of imidazole-HCl buffer (pH 6.2) 5-times and suspended in
6 mL of buffer and divided into two 3 mL parts. (3) B. amylolique-
faciens a-amylase (100 IU) was added to 3 mL (300 mg) of Beads
and glucoamylase (100 IU) was added to 3 mL (300 mg) of Beads
and allowed to incubate for 15 h at 4 °C. (4) The a-amylase-beads
and the glucoamylase-beads were filtered on 0.2 l membranes and
were washed with 6 mL of buffer (pH 6.2) 5-times and then were
each suspended in 5 mL of 40 mM imidazole-HCl (pH 6.2) buffer.
(5) The enzyme-beads were assayed and found to have 26.8 IU/
300 mg/5 mL immobilized a-amylase and 12.8 IU/300 mg/5 mL
immobilized glucoamylase.
2.2.2.2. Treatment of starch-synthases and buffered substrate
with immobilized a-amylase and glucoamylase. (1) a-
Amylase-Beads (5.4 IU) and glucoamylase-Beads (2.6 IU) were
added: (a) to 1 mL of Fraction 21 (136 IU/mL) starch-synthase;
(b) to 1 mL of Fraction 23 (236 mIU) starch-synthase; (c) to 1 mL
of second acetone precipitate (423 mIU/mL) starch-synthase; and
(d) to 1 mL of buffered (pH 6.2) 20 mM ADP-[14C]Glc. Reactions
were allowed to proceed 60 min at 30 °C. (2) The amylase-beads
were removed from the four samples by filtration through a
0.2 l membrane filter and the starch-synthase activities were as-
sayed (see Section 2.2.3 for the assay procedure) for: (a) Fraction
21, (b) Fraction 23, (c) the second acetone precipitate, and (d) the
buffered substrate solution, containing Fraction 21.
Table 1
Results of the treatments of starch-synthases with immobilized a-amylase and
glucoamylase
Starch-synthase Activitya before Activitya after
amylase treatment amylase treatment
Second acetone precipitateb 428 ± 10 427 ± 12
Fraction 21b 136 ± 15 138 ± 13
Fraction 23b 237 ± 13 235 ± 14
Buffer and ADPGlc/ADP- 138 ± 12 136 ± 15
[14C]Glc with Fraction 21
a
Activity = 1.0 mIU = 1.0 nmol of D-glucose incorporated into amylose/min.
b
Fractions obtained from Table 1 of Ref. 6.
2.2.3. Assay of starch-synthase activity
The assay was performed in 75 lL of Standard Buffer (pH 8.5),
containing 20 mM (0.05 lCi) ADP-[14C]Glc; 25 lL of starch-syn-
thase was added to the assay solution and was incubated at
37 °C for 30 min; 25 lL aliquots were removed and 2.5 lL of 1 M
NaOH was added to the aliquots to stop the reaction. This was then
added to 1.5 cm-square Whatman 3MM paper, which was immedi-
ately added to 100 mL of MeOH, with stirring for 10 min, to precip-
itate the synthesized amylose onto the paper and remove soluble
materials, such as unreacted ADPGlc, ADP, and buffer components.
The paper was washed two more times with 100 mL MeOH, dried,
and counted in toluene cocktail. Background controls were ob-
tained by adding 2.5 lL of 1 M NaOH to 25 lL of the buffer solu-
tion, containing the ADP-[14C]Glc substrate, without the enzyme,
followed by adding it to a Whatman 3MM 1.5 cm paper square.
2.3. Investigation of the possible formation of covalent
intermediates with starch-synthase during synthesis of starch
chains
2.3.1. Preparation of immobilized starch-synthase
(1) Immobeads 150 (100 mg) was added to 1 mL of water to
swell the beads for 15 h, and after swelling, the beads were centri-
fuged and washed with 1 mL of water. (2) Fraction 21 (Table 1, Ref.
6) starch-synthase (2 mL) was added to 100 mg of the swollen
Beads for 120 min for adsorption; the suspension was centrifuged
and washed with 500 lL of Standard Buffer, without DTT. The
starch-synthase-beads were diluted to 20 mL with the Standard
Buffer, containing 1 mM DTT. (3) Starch-synthase-Beads were as-
sayed for starch-synthase activity and had 1.0 IU/mg of Beads.
2.3.2. Reaction of immobilized starch-synthase-beads with
ADP-[14C]Glc to investigate the possible formation of covalent
intermediates with the enzyme during catalysis
(1) Standard Buffer (20 lL without DTT), pH 8.4, containing
0.25 lCi ADP-[14C]Glc + 25.0 mg ADPGlc, all in 1.92 mL, was added
to 200 lL (8.0 IU) Starch-Synthase-Beads. (2) The reaction was al-
lowed to proceed at 37 °C for 23 min or 0.5 conversion period
(one conversion period is the theoretical amount of time necessary
to completely convert the substrate into product). (3) The reaction
was stopped by dilution (1?30) with Standard buffer and centri-
fuged for 5 min and the supernatant was removed. (4) The
starch-synthase-beads were washed 3-times with 200 lL of water
to remove any unreacted ADP-[14C]Glc and the washes were dis-
carded. (5) The reacted starch-synthase-beads were then treated
with 200 lL of 0.01 M trifluoroacetic acid (TFA), pH 2 at 50 °C for
30 min, to release any intermediates attached to the starch-syn-
thase. (6) The starch-synthase-beads were removed by centrifuga-
tion and washed 2-times with 100 lL of water that was added to
the 200 lL of the 0.01 M TFA supernatant. (7) The resulting
400 lL solution was taken to dryness and 100 lL of water was
added to the dried material and transferred to a microcentrifuge-
tube; a second 100 lL of water was added and transferred to the
tube. The 200 lL of the dissolved material was added in a narrow
line to the top of a 23 cm  36 cm Whatman 3MM chromatogra-
phy paper for descending paper chromatography, using 70:30 (v/
v) 1-propanol/water at 20 °C, with D-glucose and maltodextrin
standards.9 (8) The solvent was allowed to descend for 16 h at
22 °C. The paper was removed and dried. The standards on the pa-
per were developed, using the silver nitrate method.9 (9) A
1.5 cm  15 cm strip at the origin of the chromatogram was cut
out for the released amylodextrin (amylose) and a 1.5 cm  15 cm
strip of the chromatogram, containing the released D-glucose, was
cut from the chromatogram, and the 14C in each sample strip
counted in 10 mL of Toluene Scintillation Cocktail. (10) A control
was performed by adding 56,068 cpm 14C-starch to 200 lL
 R. Mukerjea, J. F. Robyt / Carbohydrate Research 352 (2012) 137–142
(8.0 IU) immobilized starch synthase beads and incubated for
30 min, followed by removal of the beads and washing the beads
three-times, with 200 lL of water, drying the beads, and then
counting them in toluene cocktail.
2.3.3. Determination of the direction of the reaction of starch-
synthase polymerization of starch chains by pulse reactions
with ADP-[14C]Glc and chase reactions with ADPGlc
(1) Three pulse reactions were conducted in 10 mL, containing
the Standard Buffer, with three different concentrations of ADP-
[14C]Glc (10, 15, and 20 mM) and 500 mIU of starch-synthase at
37 °C. (2) The three pulse reactions were allowed to proceed 0.75
conversion period or 118, 177, and 236 min, respectively. The reac-
tions were divided into two parts and each placed into an ice bath.
(3) To one part, 2-volumes of cold (0 °C) ethanol was added to stop
the reaction and precipitate the pulsed synthesized starch chains.
(4) To the second part, 5 mL of Standard Buffer, containing
20 mM nonlabeled ADPGlc, was added and the reaction was al-
lowed to proceed at 37 °C for two conversion-periods or 629 min
for the chase reaction. (5) Two-volumes of 0 °C ethanol was added
to the chase reaction and placed at 4 °C for 15 h to completely pre-
cipitate the synthesized chased polysaccharide. (6) The pulse and
chase precipitates were centrifuged and washed 5-times with
75:25 (v/v) ethanol/water to remove buffer, ADP, and any unre-
acted ADP-[14C]Glc. (7) The pulsed and chased precipitates were
then dissolved in 0.5 mL of 0.1 M NaOH and 10 lL, containing
2 mg of NaBH4, was added to each solution and reduction was al-
lowed to proceed at 70 °C for 30 min; (8) 100 lL of 1 M TFA was
added to the pulse and chase precipitates to stop the reduction;
they were then concentrated to dryness in a Speed-vac to remove
the resulting borate as BF3; (9) 0.5 mL of 4 M TFA was then added
to the pulsed and chased precipitates and placed in an autoclave at
121 °C for 30 min to hydrolyze the polysaccharides. (10) After
autoclaving, 0.5 mL of methanol was added to the pulse and chase
samples, and taken to dryness in a Speed-vac to remove the excess
TFA; 200 lL of water was added to the dried pulsed and chased
samples. (11) D-Glucose and D-glucitol standards were added to
each side of the top of a 23  52 cm Whatman 3MM paper. (12)
The hydrolyzed pulsed and the chased solutions were separately
added to the tops of two papers, in a thin stream, with drying by
warm air. The chromatograms were irrigated with 8:1:3:2 (v/v/v/
v) nitromethane/acetic acid/ethanol/water-saturated boric acid
(20 °C) by descending chromatography for 15 h at 21–22 °C to sep-
arate the D-glucitol from D-glucose. (13) The chromatograms were
dried and the standard strips were removed and developed by the
silver nitrate method.9 D-Glucitol sections were cut from the chro-
matogram and added to 10 mL of Toluene Cocktail for liquid scin-
tillation counting to determine the amounts of D-glucitol in the
pulsed and the chased samples.
2.4. Determination of whether starch-synthase synthesizes
starch processively
Reactions were conducted with 20 mM (0.25 lCi) ADP-[14C]Glc
at 37 °C, using four different concentrations of enzyme (1440, 840,
130, and 60 mIU/mL), each for or 0.75 conversion period, corre-
sponding to 11, 19, 120, and 325 min of reaction, respectively.
Two-volumes of ethanol were added to stop the reactions and pre-
cipitate the synthesized amylose chains. The precipitates were cen-
trifuged and washed three times with 75:25 (v/v) ethanol/water.
They were then dissolved in water, reduced by NaBH4 at pH 8.5, acid
hydrolyzed with 4 M TFA for 30 min in the autoclave at 121 °C. The
resulting mixture of D-glucose and D-glucitol was separated by
descending paper chromatography,9 using 25 cm  52 cm
Whatman 3MM paper that was irrigated with 8:1:3:2 (v/v/v/v)
nitromethane/acetic acid/ethanol/water-saturated with boric acid
139
for 15 h. The D-glucose and D-glucitol strips were cut from the paper
and added to 10 mL of Toluene Cocktail for liquid scintillation count-
ing. The number average molecular weights of the synthesized
amylodextrin (amylose) chains were determined by (MWn) = (cpm
of D-glucose + cpm of D-glucitol)  (cpm of D-glucitol) for each
sample.
3. Results and discussion
3.1. Enzymatic characterization of the solubilized and purified
potato starch-synthases
The enzymes used in this study were (1) the second acetone
precipitate, (2) Fraction 21, and (3) Fraction 23 starch-synthases
that were isolated and purified from white Idaho Russet (Solanum
tuberosum) potatoes, as previously reported (Table 1 in Ref. 6). The
buffers plus substrate and enzymes were tested for the presence
and absence of putative primers, using the Whelan and co-work-
ers10 techniques of treatment of the samples with immobilized
a-amylase and glucoamylase. The combination of the two enzymes
was found to completely convert 10 mg potato starch into 100% D-
glucose in 30 min. (1) The Standard Buffer, containing ADPGlc and
ADP-[14C]Glc, (2) the second acetone precipitate, (3) Fraction 21,
and (4) Fraction 23 starch-synthases were treated with the immo-
bilized a-amylase and glucoamylase for 60 min. After the treat-
ment, Fraction 21 was added to the Standard Buffer, containing
ADPGlc and ADP-[14C]Glc. The starch synthase assay results for
all four of the treatments gave the same starch-synthase activities,
within experimental error, before and after the treatments (see,
Table 1 in this study). In addition, this also indicated that all
twenty-five fractions obtained from the polysulfone hollow-fiber
cartridge column6 were free of any potential primers, as they too
were obtained from the second acetone precipitate. Fractions 21
and 23 exclusively contained starch-synthase, had a very high
specific activity, and were free of starch-branching enzyme.6
The experiments also show that potato starch-synthase synthe-
sizes amylose chains de novo without the need of a primer.
These enzymes were used throughout the following studies in
Sections 3.2-3.4.
3.2. Demonstration that D-glucopyranosyl and D-glucanyl starch
chains are covalently attached to the active-site of starch-
synthase during catalysis
An immobilized Fraction 21 starch-synthase was reacted with
20 mM (0.25 lCi) ADP-[14C]Glc for 0.5 conversion period or
3 min at 37 °C; the immobilized enzyme was removed from the
reaction digest, washed with water, treated at pH 2 and 50 °C for
30 min, followed by chromatography on Whatman 3MM paper.
The results of the chromatography are shown in Figure 1. This fig-
ure shows that two 14C-components were released: (1) 1470 cpm
14
C-starch and high molecular weight (dp >6) 14C-maltodextrin
chains, at the origin of the paper chromatogram, and (2) a second
component, 650 cpm 14C-D-glucopyranose (10 cm from the origin).
It should be noted that the condition of hydrolysis used in this
experiment (pH 2, 50 °C, 30 min) will not hydrolyze a-(1?4)-gly-
cosidic linkages but will hydrolyze high-energy carboxyl acetal-
linkages. In addition, the control experiment in which 14C-starch
was incubated with the immobilized starch synthase for 30 min
showed that none of the 14C-starch was adsorbed onto the Imm-
obeads or the starch synthase. The results of these experiments
definitively show that both D-glucopyranose and growing glucan
chains were covalently linked to the enzyme active-sites during
catalysis in that they could only be released from the immobilized
starch synthase by the pH 2, 50 °C, 30 min treatment.
140 R. Mukerjea, J. F. Robyt / Carbohydrate Research 352 (2012) 137–142
Figure 1. Demonstration of the covalent intermediates of starch chain and
glucose occurring during catalysis.
3.3. Pulse reaction of starch-synthase with ADP-[14C]Glc and
chase reaction with nonlabeled ADPGlc demonstrating the
addition of D-glucose to the reducing-end of the growing chain
Three pulse and chase experiments, using three concentrations
of ADPGlc were conducted with Fraction 23, starch-synthase. The
results are shown in Table 2. All three pulse experiments gave
14
C-D-glucitol that could only come from the reducing-ends of
the synthesized starch-chain. On chasing the synthesized starch-
chains with nonlabeled ADPGlc, a decrease in the amount of 14C-
D-glucitol was observed (see, Table 2 for the results), which also
could only occur at the reducing-ends of the synthesized starch
chains. These three pulse and chase experiments of starch-syn-
thase definitively show that the starch chains are synthesized by
the addition of D-glucose from ADPGlc to the reducing-ends of
the growing starch-chains. Figure 2 shows the products of the
pulsed reaction, the chased reaction, and the blue triiodide color
of the product from the chased reaction. This figure shows that
the ethanol precipitated products of the pulse and chase reactions
were polymeric in nature and that the blue triiodide color of the
chased product shows that it is amylose.
3.4. Demonstration of the processivity of the starch-synthase
reaction
Four different amounts of starch-synthase (1440, 840, 130, and
60 mIU) were reacted with a constant amount of substrate (20 mM
0.25 lCi ADP-[14C]Glc) at 37 °C in a constant volume, each for a
constant amount of reaction of 0.75 conversion period for 11, 19,
120, and 325 min, respectively. The synthesized 14C-labeled starch
chains were precipitated with 2-volumes of ethanol and washed 3-
times with 75:25 (v/v) ethanol/water; reduced with NaBH4; acid
hydrolyzed; and chromatographed, using descending paper chro-
matography.9 The number average molecular weights were com-
puted for each of the synthesized glucans from the cpm obtained
for 14C-D-glucitol and 14C-D-glucose. The results are shown in Table
3. The four experiments show that as the concentration of the en-
zyme was decreased, the number average molecular weight was
Table 2
Results of pulse and chase reactions of ADP-[14C]Glc with starch-synthase
Experimenta Concentration of ADPGlc Pulsed D-glucitol Chased D-glucitol
in the pulsed reaction mM (cpm) (cpm)
I 10 775 296
II 15 1003 429
III 20 1332 734
a
Starch-synthase (500 mIUb) pulse reactions were conducted with 10, 15, and
20 mM (0.25 lCi) ADP-[14C]Glc for 0.75 conversion period (118, 177, and 236 min,
respectively, for the three substrate concentrations) and the chase reactions with
20 mM ADPGlc at 37 °C for two conversion periodsc or 629 min.
b
mIU = nmol of D-glucose incorporated into amylose/min.
c
One conversion period is the amount of time to theoretically convert the sub-
strate to product, under the conditions of the reaction.
D-
PULSE CHASE
Synthesized Starches from Triiodide Color
Pulse and Chase Reactions of Synthesized
Starch
Figure 2. Picture of the synthesis of starch chains during, pulse and chase reactions,
and the blue triiodide color of the synthesized chased product.
increased, indicating that the synthesis of the amylose chains by
starch-synthase is inversely proportional to the concentration of
the enzyme. This result shows that the reaction is processive, that
is, many D-glucopyranose units are added sequentially to the
reducing-ends of the growing chains, giving the extrusion of the
chains from the active-site of the starch-synthase, until the chain
is released by hydrolysis.
The reason that the inverse relationship between the concentra-
tion of the enzyme and the number average molecular weight of
the synthesized glucan demonstrates that the processive character
of the starch-synthase polymerization is the following: (1) there
are a fixed number of enzyme molecules in the reaction digest that
give a certain number of synthesized polysaccharide chains with a
certain number average molecular weight; (2) if the concentration
of the enzyme is significantly reduced (e.g., by one half) there
would be a lower number of enzyme molecules in the digest, and
when the concentration of the substrate remains the same, there
would be longer polysaccharide chains synthesized, because there
are fewer enzyme active-sites, with the resulting increase in the
number average molecular weight; (3) continuing to decrease
the concentration of the enzyme would also result in an increase
in the number average molecular weight, giving an inverse rela-
tionship of molecular weight to enzyme concentration; (4) if, how-
ever, the synthesis is not processive, the number of monomer units
incorporated into the polymer would be the same, or nearly the
same, at any concentration of the enzyme, if the same degree of
conversion is allowed to occur. A similar result was previously ob-
tained for dextransucrase, showing that the number average
molecular weight of the dextrans also was inversely proportional
to the concentration of the enzyme, indicating that the polymeriza-
tion of dextran by dextransucrase also is highly processive.11
 R. Mukerjea, J. F. Robyt / Carbohydrate Research 352 (2012) 137–142 141

Table 3 carboxyl-acetal of the growing glucan chains that are covalently

Number average molecular weights of synthesized amyloses as a function of the
concentration of starch-synthase
Experimenta Concentration of starch-synthaseb (mIU/mL) MWnc (Da)
I 1440 3658
II 840 4817
III 130 6547
IV 60 7632
a
The reactions were conducted at different concentrations of starch-synthase,
each with 20 mM (0.25 lCi) ADP-[14C]Glc at 37 °C for 0.75 conversion period or 11,
19, 120, and 325 min reaction times, respectively.
b D-glucopyranose
Concentration of starch-synthase in mIU = nmol of glucose incorporated into
amylose/min.
c
MWn = number average molecular weight, computed from [cpm D-glu-
cose + cpm D-glucitol]  [cpm D-glucitol].
The processive character of the polymerization indicates that a
high-energy, covalent intermediate holds the growing polymer at
the active-site of the enzyme, where the high-energy monomer is
added. A nonprocessive reaction would indicate the addition of the
high-energy donor to the ends of an acceptor or putative primer
and would not give a high molecular weight polymer.
The studies of Damager et al.12, using various maltodextrin
putative primers for starch synthesis showed that the putative
primers did not give the synthesis of starch chains. The putative
primers used were maltotriose, maltohexaose, and a branched,
6III-a-maltotriosyl maltohexaose that were added to ADP-
[14C]Glc/starch-synthase digests and only one or two D-glucose
units were added to the nonreducing-ends of each of the malto-
dextrin putative primers. Further, when the branched pentasac-
charide, a-maltosyl-6II-a-methyl-maltotrioside was tested as a
linked to the active-site of starch-synthase. This occurs by a trans-
glycosylation reaction of the growing glucan chain to the D-gluco-
pyranosyl unit (see, Fig. 3 for the mechanism).
3.5. The proposed mechanism for the de novo, processive
synthesis of starch chains by starch-synthase without the
requirement of a primer
To explain all of the properties of the synthesis of the a-(1?4)-
linked glucan chains, as shown in this study, for
example, (1) de novo synthesis without the requirement of a pri-
mer; (2) covalent D-glucopyranosyl and D-glucanyl covalent inter-
mediates; (3) the addition of D-glucopyranosyl units to the
reducing-ends of the covalently linked D-glucanyl chains; (4) the
processive synthesis by the sequential addition of D-glucopyranose
residues to the growing D-glucanyl chain; and (5) termination of
the synthesis by hydrolysis of the starch chain from the active-site
of starch synthase. Figure 3 presents a two catalytic-site, insertion
mechanism that encompasses these experimental results, and
shows the mechanism for the polymerization of amylose chains
by starch-synthase, involving an initiation step, polymerization
steps, and a termination step. X1 and X2 are two carboxylate cata-
lytic groups at the active-site of starch-synthase. The black circles
represent D-glucopyranosyl units linked a-1?4, and a black circle
with a slanted line through it represents a reducing D-glucopyra-
nose unit.
Initiation of the polymerization begins with the formation of
two D-glucopyranosyl groups covalently attached to the two-car-
boxylate groups. The C4–OH group of one of the covalently-linked
D-glucopyranosyl residues attached to a carboxyl group at the ac-

primer,13 it too was found that only a single D-glucose unit was
added to one of the two nonreducing-ends of the branched penta-
saccharide, giving two hexasaccharides that were not further elon-
gated.12 These experiments of Damager et al.12,13 show that the
starch-synthase transglycosylation reaction of the transfer of D-
glucose to the nonreducing-ends of the putative primer oligosac-
charides is a limited, nonprocessive, acceptor, side-reaction in
which only one or two D-glucose units were added to the nonre-
ducing-ends of the acceptor molecules. These acceptor (putative
primer) reactions do not lead to the polymerization of starch
chains. Instead, the synthesis of starch chains occurs by a proces-
sive de novo reaction without the need of an added primer, with
the apparent addition of D-glucopyranose to the high-energy,
tive-site, makes a nucleophilic attack onto the covalently linked
C1-carbon of the other D-glucopyranosyl residue to form an a-
(1?4) glucosidic linkage; the next step involves the C4–OH group
of the other D-glucopyranosyl intermediate making a nucleophilic
attack onto C1 of the potential reducing-end of the growing amy-
lose chain, with the new D-glucopyranosyl group being inserted be-
tween the reducing-end of the growing amylose chain and the
carboxyl group at the enzyme active-site. This continues, going
back and forth between the two-carboxyl groups having a D-gluco-
pyranosyl group and a growing glucanyl chain, alternately at-
tached to X1 and X2 carboxyl groups, giving the processive
synthesis of the starch chain and its extrusion from the active-site
of the enzyme.

ADP ADP ADP

X1 X1 X1 X1 X1
I II OH
II OH
OH
X2 X2 X2 X2 X2
enzyme Di-D-glucopyranosyl II
ADP
active-site ADP
enzyme complex Polymerization occurs
n-times going back and
I = Initiation step II = polymerization steps forth between X1 and X2

H 2O X1
X1
n +
Synthesized Amylose n
X2
chain X2 III
Mono-D-glucopyanosyl D-glucopyranosyl attached
enzyme complex ready to X1 and amylose chain
to continue polymeriation attached to X2
III = Termination step: hydrolysis of amylose from the enzyme active site

Figure 3. Two catalytic-site, insertion mechanism for the biosynthesis of starch chains by starch-synthase.
142 R. Mukerjea, J. F. Robyt / Carbohydrate Research 352 (2012) 137–142
It is hypothesized that a starch-binding domain binds the D-glu-
cose residues of the synthesized glucan chain, as it is being ex-
truded, glucose by glucose unit, from the active-site, and
eventually the glucose units extend beyond the binding sub-sites.
This puts a strain, and an increase in energy, on the covalent link-
age of the chain to the enzyme active-site carboxyl group and the
linkage is then hydrolyzed, releasing the chain from the active-site,
terminating the synthesis.
4. Summary and conclusions
(1) It has been shown, using immobilized a-amylase and gluco-
amylase that the potato starch synthesizing starch-syn-
thases, in the second acetone precipitate, Fractions 21 and
23, obtained from the polysulfone hollow fiber 100 kDa car-
tridge previously reported,6 was free of any putative
primers.
(2) Fractions 21 and 23 had specific activities of 544 and 944 IU/
mg of protein, respectively, indicating a high degree of pur-
ity. They synthesized a-(1?4)-linked amylose chains de
novo, without the need of an added primer.
(3) Immobilization of Fraction 21 starch-synthase and its reac-
tion with ADP-[14C]Glc, followed by removal and washing
of the immobilized starch-synthase from the reaction digest,
and its treatment at pH 2 and 50 °C for 30 min, released both
14
C-D-glucopyranose and 14C-labeled amylose chains, indi-
cating that both D-glucopyranose and the growing amylose
chain were covalently attached to the active-site of starch-
synthase, during reaction.
(4) Starch-synthase (Fraction 23) was pulsed with ADP-[14C]Glc
and chased with ADPGlc. After reduction and acid hydroly-
sis, the pulsed starch chains gave a significant amount
14
C-D-glucitol, and after chasing with nonlabeled ADPG, the
14
C-D-glucitol was significantly decreased, showing that the
synthesis of the amylose chains was by the addition of D-glu-
cose from ADPGlc to the reducing-ends of the covalently
linked growing starch chains.
(5) Reaction of four different concentrations of starch-synthase,
in decreasing amounts, with a constant amount of 20 mM
ADP-[14C]Glc, constant temperature of 37 °C and equal
amounts of reaction (0.75 conversion period), gave four
amylose chains, whose number average molecular weights
were inversely proportional to the amount of enzyme in
the digests, indicating that the polymerization of the amy-
lose chains is processive.
(6) A two catalytic-site, insertion mechanism is proposed to
explain the experimental data, showing the initiation, poly-
merization, and termination of the synthesis that occurs by
the addition of D-glucose, from ADPGlc, to the reducing-ends
of the growing glucan chains that are covalently linked to
the carboxyl groups at the active-sites of starch-synthase.
References
1. Mukerjea, R.; Yu, L.; Robyt, J. F. Carbohydr. Res. 2002, 337, 1015–1022.
2. Mukerjea, R.; Robyt, J. F. Carbohydr. Res. 2005, 340, 245–250.
3. Mukerjea, R.; Robyt, J. F. Carbohydr. Res. 2005, 340, 2206–2211.
4. Hanes, C. S. Proc. R. Soc. London, Ser. B 1940, 129, 174–208.
5. Leloir, L. F.; de Fekete, M. A. R.; Cardini, C. E. J. Biol. Chem. 1961, 236, 636–641.
6. Mukerjea, R.; Falconer, D. J.; Yoon, S.-H.; Robyt, J. F. Carbohydr. Res. 2010, 345,
1555–1563.
7. Chiang, C.-J.; Lee, W.-C.; Sheu, D.-C.; Duan, K.-J. Biotechnol. Prog. 1997, 13, 577–
582.
8. Stein, E. A.; Fischer, E. H. Biochem. Prep. 1961, 8, 34–38.
9. Robyt, J. F.; White, B. J. Biochemical Techniques: Theory and Practice; Waveland
Press: Prospect Heights, IL, 1990.
10. Schiefer, S.; Lee, E. Y. C.; Whelan, W. J. Carbohydr. Res. 1978, 61, 239–252.
11. Robyt, J. F.; Yoon, S.-H.; Mukerjea, R. Carbohydr. Res. 2008, 343, 3039–3048.
12. Damager, I.; Denyer, K.; Motawia, M. S.; Møller, B. L.; Blennow, A. Eur. J.
Biochem. 2001, 268, 4847–4884.
13. Damager, I.; Olsen, C. E.; Blennow, A.; Denyer, K.; Møller, B. L.; Motawia, M. S.
Carbohydr. Res. 2003, 338, 189–197.