Qualitative Profiling of Polyglucose Degradation Products in Peritoneal Dialysis Fluids

Article abstract of DOI:10.1021/acs.analchem.5b00665

Heat sterilization of peritoneal dialysis (PD) fluids leads to partial degradation of the osmotic agent to form reactive carbonyl structures, which significantly reduce the biocompatibility of PD fluids and impair long-term PD therapy. Hence, it is important to know the exact composition of the degradation products to improve biocompatibility of PD fluids. Our study conducted targeted screening for degradation products in polyglucose (icodextrin)-containing PD fluids (pGDPs) by applying o-phenylenediamine (OPD) to form stable derivatives, which were analyzed by ultrahigh-performance liquid chromatography with hyphenated diode array tandem mass spectrometry (UHPLC-DAD-MS/MS). For the first time, specific degradation products of polyglucose, namely, 4-deoxyglucosone (4-DG) and 3,4-dideoxypentosone (3,4-DDPS), could be identified in PD fluids. Further, a reaction product of 5-hydroxymethylfurfural (5-HMF) and OPD could be characterized to be (5-(1H-benzo[d]imidazol-2-yl)furan-2-yl)methanol. Additionally, 3-deoxyglucosone (3-DG) and 3-deoxygalactosone (3-DGal), both known to be present in glucose-based PD fluids, were also detected in polyglucose-containing fluids. Trapping a hitherto unknown degradation product with OPD yielded 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one, which was present in heat- as well as filter-sterilized PD fluids.

Full text of DOI:10.1021/acs.analchem.5b00665

Subscriber access provided by UNIVERSITY OF LEEDS
Article
Qualitative Profiling of Polyglucose Degradation
Products in Peritoneal Dialysis Fluids
Sabrina Gensberger, Carina Knabner, Reiner Waibel, Jochen Huppert, and Monika Pischetsrieder
Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b00665 • Publication Date (Web): 13 May 2015
Downloaded from http://pubs.acs.org on May 17, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth
Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 31
Analytical Chemistry
1
2
3
1
4
Qualitative Profiling of Polyglucose Degradation Products in Peritoneal Dialysis Fluids
5
6
2
7
8
9
3
4
Sabrina Gensberger,† Carina Knabner,† Reiner Waibel,‡ Jochen Huppert,§ Monika
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Pischetsrieder†*
5
6
† Food Chemistry Unit, Department of Chemistry and Pharmacy, Emil Fischer Center, Friedrich-
Alexander Universität Erlangen-Nürnberg (FAU), Schuhstr. 19, 91052 Erlangen, Germany
‡ Medicinal Chemistry Unit, Department of Chemistry and Pharmacy, Emil Fischer Center,
Friedrich-Alexander Universität Erlangen-Nürnberg (FAU), Schuhstr. 19, 91052 Erlangen,
Germany
7
8
9
10
§ Fresenius Medical Care Deutschland GmbH, Frankfurter Str. 6-8, 66606 St. Wendel, Germany
11
12
* Correspondence to:
13
14
15
16
17
Monika Pischetsrieder, Food Chemistry Unit, Department of Chemistry and Pharmacy, Emil
Fischer Center, Friedrich-Alexander Universität Erlangen-Nürnberg, Schuhstr. 19, 91052
Erlangen, Germany
Email: monika.pischetsrieder@fau.de, Phone: +49-9131-8524102, fax: +49-9131-8522587
1
ACS Paragon Plus Environment

Analytical Chemistry
Page 2 of 31
1
2
3
4
5
6
7
8
9
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Abstract
Heat sterilization of peritoneal dialysis (PD) fluids leads to partial degradation of the osmotic
agent to form reactive carbonyl structures, which significantly reduce the biocompatibility of PD
fluids and impair long-term PD therapy. Hence, it is important to know the exact composition of
the degradation products to improve biocompatibility of PD fluids. Our study conducted targeted
screening for degradation products in polyglucose (icodextrin)-containing PD fluids (pGDPs) by
applying o-phenylenediamine (OPD) to form stable derivatives, which were analyzed by
ultrahigh-performance liquid chromatography with hyphenated diode array tandem mass
spectrometry (UHPLC–DAD–MS/MS). For the first time, specific degradation products of
polyglucose, namely 4-deoxyglucosone (4-DG) and 3,4-dideoxypentosone (3,4-DDPS), could be
identified in PD fluids. Further, a reaction product of 5-hydroxymethylfurfural (5-HMF) and
OPD could be characterized to be (5-(1H-benzo[d]imidazol-2-yl)furan-2-yl)methanol.
Additionally, 3-deoxyglucosone (3-DG) and 3-deoxygalactosone (3-DGal), both known to be
present in glucose-based PD fluids, were also detected in polyglucose-containing fluids.
Trapping a hitherto unknown degradation product with OPD yielded 1,4-bis(1H-
benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one, which was present in heat- as well as in
filter-sterilized PD fluids.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 31
Analytical Chemistry
1
2
3
36
4
Introduction
5
6
7
8
9
37
Peritoneal dialysis (PD) is a common alternative to hemodialysis for blood purification in case of
renal insufficiency. A highly-osmotic PD fluid is administered into the abdominal cavity so that
the peritoneum as semipermeable membrane allows excessive water and uremic waste to diffuse
from the patient’s blood into the PD fluid.1 To enable ultrafiltration and ensure biocompatibility,
PD fluids contain water, electrolytes like calcium or sodium, a buffer substance and, in most
cases, glucose as osmotic agent. In some PD fluids glucose is replaced by other osmotic agents,
such as icodextrin, a high-molecular, water-soluble glucose polymer derived from starch.2
During production, PD fluids are heat-sterilized to assure microbiological safety. It is well-
known that heat treatment leads to the formation of reactive glucose degradation products
(GDPs).3, 4 GDPs, however, reduce the biocompatibility of PD fluids and, thus, limit long-term
therapy, for example by enhanced formation of advanced glycation endproducts (AGEs).5-8
These modifications may impair protein biofunctionality and eventually lead to a decline of
membrane permeability.9-11 Furthermore, GDPs are cytotoxic against human peritoneal
mesothelial cells as well as fibroblasts and may interfere with cell signaling.9, 12-14 Therefore,
minimizing the GDP content by technological measures, changes in the product composition or
the use of additives is crucial to provide maximum biocompatibility,15-17 an objective that
requires detailed knowledge about GDP profiles in PD fluids.
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
54
55
56
57
58
GDP profiles of glucose-based PD fluids have already been extensively investigated. Thus,
several GDPs with monocarbonyl structure, like 5-hydroxymethylfurfural (5-HMF),
acetaldehyde, formaldehyde, and 2-furaldehyde have been identified and quantified in PD
fluids.6, 7, 18, 19 Additionally, the α-dicarbonyls glucosone, 3-deoxyglucosone (3-DG), 3-
deoxygalactosone (3-DGal), 3,4-dideoxyglucosone-3-ene (3,4-DGE), glyoxal, and
3
ACS Paragon Plus Environment

Analytical Chemistry
Page 4 of 31
1
2
3
4
5
6
7
8
9
59
60
61
62
63
64
65
66
67
68
methylglyoxal have been identified as the major GDPs and quantified in different commercial
products .12, 19-22 Thus, it was shown that glucose-based PD fluids contain GDPs in
concentrations up to 450 µM.23 In contrast, the GDP content of polyglucose-based solutions has
hardly been investigated. Exemplarily, selected typical GDPs were analyzed in few samples of
icodextrin-containing fluids.19, 24 However, icodextrin is an oligomer, where glucose monomers
are linked by 1,4-α glycosidic bonds. Compared to glucose, 1,4-α glycosidically linked di- and
oligosaccharides undergo different degradation reactions.25-28 It can be assumed that the
degradation products formed in icodextrin-based PD fluids differ from those formed in glucose-
based products. Consequently, the content of process contaminants may be underestimated when
only GDPs are quantified.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
69
70
71
72
73
74
Therefore, we conducted targeted screening for the most important polyglucose degradation
products (pGDPs) in PD fluids with reactive carbonyl structures. Thus, it was possible to identify
5-HMF, 3-DG, 3-DGal, 4-deoxyglucosone (4-DG), and 3,4-dideoxypentosone (3,4-DDPS) in
polyglucose-containing PD fluids. The latter two compounds have not been detected before in
glucose-containing products. Additionally, trapping with OPD identified a raw material
contaminant which forms 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one.
75
76
Experimental Section
77
78
79
80
Reagents and Samples. All chemicals and reagents were obtained from Acros (Geel, Belgium),
AppliChem (Darmstadt, Germany), Fluka (Steinheim, Germany), or Sigma-Aldrich (Steinheim,
Germany) and were at least of analytical grade. 3-DG (purity > 95%) was purchased from
Chemos (Regenstauf, Germany) and 5-HMF from SAFC (St. Louis, MO, USA). Water for all
4
ACS Paragon Plus Environment

Page 5 of 31
Analytical Chemistry
1
2
3
81
4
experiments was taken from a labwater system (Synergi-185, Millipore, Schwalbach, Germany).
Typical commercial PD fluids with 7.5% icodextrin were examined. Unheated and sterile-
filtered PD fluids were prepared containing the buffer and electrolytes of a commercial product
with 7.5% icodextrin.
5
6
7
8
82
83
84
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
85
86
Semipreparative HPLC System, UHPLC–DAD–MS/MS System, HR-MS, and NMR
Spectroscopy. Semipreparative high-performance liquid chromatography (HPLC) was carried
out on a Jasco Plus system (Groß-Umstadt, Germany) equipped with type 2057 autosampler, two
2087 pumps, 980-50 degasser, and 2077 ultraviolet–visible (UV/Vis)-detector. The system was
controlled by Chrompass 1.7 software. Separation of compounds was achieved on a Nucleodur
C18ec column (250 x 10 mm Macherey-Nagel, Dueren, Germany), which was connected to a
precolumn of the same material (8 x 10 mm, Macherey-Nagel). Fractions were collected by
Advantec fraction collector CHF 122 SB.
87
88
89
90
91
92
93
The Dionex UltiMate 3000RS ultrahigh-performance liquid chromatography with hyphenated
94
diode array tandem mass spectrometry (UHPLC–DAD–MS/MS) system (Thermo Fisher
95
Scientific, Dreieich, Germany) consisted of a pump with degasser, autosampler, column
compartment, and diode array detector, controlled by Chromeleon 6.80 software.
Chromatographic separation was performed on an ACQUITY UPLC® BEH phenyl column (1.7
µm particle size; 2.1 x 100 mm, Waters, Eschborn, Germany), which was equipped with a
VanGuard ACQUITY UPLC® BEH phenyl precolumn (1.7 µm particle size; 2.1 x 5 mm,
96
97
98
99
100
101
102
103
Waters). For UHPLC
Foster City, CA, USA) fitted with an electrospray ionization (ESI) source (Applied Biosystems,
Foster City, CA, USA) was connected to the UHPLC system. The ESI MS was run in positive
–MS/MS analysis, an API 4000 QTrap mass spectrometer (AB Sciex,
–
mode, operated at 700 °C with a voltage of +2500 V. Nitrogen was used for solvent evaporation
5
ACS Paragon Plus Environment

Analytical Chemistry
Page 6 of 31
1
2
3
4
5
6
7
8
9
104
105
106
107
108
109
110
111
112
as well as for collision-induced dissociation. Product ion spectra were obtained at collision
energy of 30 V and collision energy spread of 10 V in QTrap-enhanced mode. For data
acquisition and processing, Analyst 1.5.1 software was used.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
High-resolution (HR) ESI–
MS spectra were obtained using a maXisTM time-of-flight mass
spectrometer (Bruker Daltonics, Bremen, Germany). NMR spectra were recorded by an Avance
600 system (Bruker Biospin, Rheinstetten, Germany). Signals were assigned based on 1H,1H-
correlated spectroscopy (COSY), hetero nuclear single quantum coherence (HSQC), and hetero
nuclear multiple bond correlation (HMBC). For analysis of the raw material contaminant,
additional hydrogen–deuterium (H/D) exchange experiments were performed.
113
114
Synthesis of the Reference Compounds
3-DGal was prepared as previously reported.29
115
116
117
118
119
120
121
122
123
124
125
4-DG was synthesized according to Broberg et al. with some minor changes.30 Briefly, maltose
(60 mM) was dissolved in degassed aqueous potassium hydroxide (50 mM). The reaction
mixture was kept light-protected under nitrogen and was stirred constantly at room temperature
(RT) for 24 h. Afterward the solution was neutralized by adding Dowex 50W (H+) ion
exchanger, followed by filtration and evaporation of the solvent to yield a brown raw product. 4-
DG was then isolated by column chromatography. For this purpose, methanol (100 mL) and
silica gel (5 g) were added to the raw product. Subsequently, the solvent was removed under
reduced pressure. The resulting dry powder was loaded onto a column (5 x 17 cm, silica gel). 4-
DG was eluted with chloroform/methanol 17:3, controlled by thin-layer chromatography (TLC)
(Rf 0.36, visualization with thymol/ethanol/sulfuric acid). When TLC indicated the presence of
4-DG, the respective fractions were concentrated under reduced pressure and lyophilized to yield
6
ACS Paragon Plus Environment

Page 7 of 31
Analytical Chemistry
1
2
3
126
4
39 mg of 4-DG.
5
6
7
8
127
128
129
130
131
132
133
The product was characterized by NMR (atom numbering refers to Scheme 1A). NMR spectra
showed two intramolecular hemiacetals of 4-DG, both formed by ring closure: form 1 1H-NMR
(DMSO-d6): δ 2.28 (1H, H-4a); 2.52 (1H, H-4b); 3.35 (1H H-1a); 3.43 (1H H-1b); 3.63 (1H H-
6a); 3.65 (1H H-6b); 4.38 (1H H-5). 13C-NMR (DMSO- d6): δ 37.8 (C-4); 63.2 (C-1); 64.5 (C-6);
74.8 (C-5); 99.4 (C-2); 212.4 (C-3), form 2 1H-NMR (DMSO-d6): δ 2.41 (1H H-4a); 2.51 (1H H-
4b); 3.36 (1H H-1a); 3.48 (1H H-1b); 3.51 (1H H-6a); 3.53 (1H H-6b); 4.41 (1H H-5), 13C-NMR
(DMSO- d6): δ 37.7 (C-4); 62.9 (C-6); 63.9 (C-1); 74.1 (C-5); 99.8 (C-2); 213.4 (C-3).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
(S)-3-(3-(Hydroxymethyl)quinoxalin-2-yl)propane-1,2-diol (4-DGqx). For the preparation of
the quinoxaline derivative of 4-DG, maltose was treated as described above. After neutralization
with Dowex 50W and subsequent filtration, the reaction mixture was concentrated under reduced
pressure to about 100 mL. Then 20 mL of OPD (0.2% in water) was added to convert 4-DG into
4-DGqx. The reaction mixture was stirred at RT for about 20 h prior to isolation of 4-DGqx by
semipreparative HPLC with the following parameters: eluent A formic acid (0.1 %), eluent B
methanol; flow rate 2.0 mL/min; (time (min)/%B) 0/10, 15/90, 20/90, 20.1/10, 25/10. Aliquots of
400 µL of the reaction mixture were injected and chromatograms were recorded at 316 nm.
Fractions 17.1–17.8 min were collected, combined, concentrated under reduced pressure, and
lyophilized yielding 50 mg of 4-DGqx.
The product was characterized by NMR (atom numbering refers to Scheme 1A) and by
UHPLC–DAD–MS/MS. A representative product ion spectrum is shown in Figure 1B. MS/MS:
235.0 Da [M+H]+, 171.1 Da, and 217.0 Da; 1H-NMR (DMSO-d6): δ 3.11 (dd, J = 14.2, J = 5.4
1H, H-4a); 3.18 (dd, J = 14.2 J = 4.3 1H, H-4b); 3.45 (m, 2H, H-6); 4.07 (m, 1H, H-5); 4.69 (t, J
= 5.7 1H, OH-6); 4.78 (d, J = 5.0 1H, OH-5); 4.84 (dd, J = 13.4 J = 5.2 2H H-1a); 4.92 (dd, J =
7
ACS Paragon Plus Environment

Analytical Chemistry
Page 8 of 31
1
2
3
4
5
6
7
8
9
149
150
151
152
153
13.4 J = 6.1 2H H-1b); 5.40 (t, J = 5.7 1H, OH-1); 7.78 (m, 2H, Ar-H); 8.03 (m, 2H, Ar-H), 13C-
NMR (DMSO-d6): δ 38.3 (C-4); 63.3 (C-1); 65.9 (C-6); 71.6 (C-5); 128.1; 128.3; 129.1; 129.4;
139.5; 140.9; 155.3; 155.7 (Ar-C), UV absorption maxima: 199 nm, 238 nm, 316 nm. Because
NMR spectra of 4-DGqx have not been published before, 1H-NMR- and 13C-NMR spectra are
shown in the Supporting Information, Figures S-1 and S-2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
(5-(1H-Benzo[d]imidazol-2-yl)furan-2-yl)methanol (5-HMFbfm). To prepare this
benzimidazole derivative of 5-HMF, 45 mL of a solution containing 8 mM 5-HMF was mixed
with 5 mL of OPD solution (2% in water). The reaction mixture was stirred in the dark at RT
overnight and then subjected to semipreparative HPLC using the following gradient: eluent A
formic acid (0.075%), eluent B methanol; flow rate: 2.0 mL/min; (time (min)/%B) 0/30, 12/50,
12.1/70, 15/70, 15.1/30, 20/30. Aliquots of 500 µL were injected, chromatograms were recorded
at 316 nm and fractions 12.9–14.1 min were collected and combined. The organic solvent was
evaporated and the product was lyophilized. The product was analyzed by NMR (atom
numbering refers to Scheme 1B), HR-ESI–MS, and UHPLC–DAD–MS/MS. Figure 1D shows a
representative product ion spectrum. MS/MS: 169.0 Da, 215.0 Da [M+H]+, 197.0 Da; 1H-NMR
(DMSO-d6): ): δ 4.54 (s, 2H, H-6); 6.50 (d, J = 3.2 1H, H-4); 7.10 (d, J = 3.3 1H, H-3); 7.12 (m,
2H, Ar-H); 7.54 (m, 2H, Ar-H); 12.90 (s, 1H, N-H). 13C-NMR (DMSO-d6): δ 55.8 (C-6); 109.3;
110.1 (C-3, C-4); 122.1; 143.7; 144.7; 157.1 (Ar-C). HR-ESI–
MS: 215.0817 Da [M+H]+
measured, 215.0815 Da [M+H]+ calculated. UV absorption maxima: 194 nm, 256 nm, 318 nm.
Because NMR spectra of 5-HMFbfm have not been published before, 1H-NMR and 13C-NMR
spectra are shown in the Supporting Information, Figures S-3 and S-4.
170
171
3-(Quinoxalin-2’-yl)propane-1-diol (3,4-DDPSqx). The quinoxaline derivative of 3,4-DDPS
was synthesized according to Mavric and Henle with some modifications.26 Maltose solution (50
8
ACS Paragon Plus Environment

Page 9 of 31
Analytical Chemistry
1
2
3
172
4
mL, 0.2 M) in phosphate buffer (0.5 M; pH 7.0) was mixed with 25 mL of an OPD solution
(0.2% in identical phosphate buffer). The reaction mixture was refluxed for 2 h. After cooling to
RT and membrane filtration, the sample was purified by semipreparative HPLC as follows:
eluent A formic acid (0.1%), eluent B methanol; flow rate 2.0 mL/min; (time (min)/%B) 0/48,
20/80, 25/80, 25.1/48, 30/48. Injection volume was set to 500 µL and chromatograms were
recorded at 316 nm. Fractions 15.8–16.8 min were collected and combined, the organic solvent
was evaporated and the residue was freeze-dried to yield 3.3 mg of 3,4-DDPSqx.
5
6
7
8
173
174
175
176
177
178
179
180
181
182
183
184
185
186
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The product was characterized by NMR (atom numbering refers to Scheme 1C), HR-ESI
–MS,
and UHPLC–DAD–MS/MS. Figure 1F depicts a representative product ion spectrum. MS/MS:
171.1 Da, 143.1 Da, 189.1 Da [M+H]+. 1H-NMR (DMSO-d6): δ 1.96 (m, 2H, H-4); 3.05 (t, J =
8.0 2H, H-5); 3.51 (t, J = 6.3 2H, H-3); 4.55 (s, 1H, OH); 7.81 (m, 2H, Ar-H); 8.04 (m, 2H, Ar-
H); 8.88 (s, 1H, H-1). 13C-NMR (DMSO-d6): δ 31.7 (C-4); 32.1 (C-3); 60.1 (C-5); 128.5; 128.8;
129.0; 130.0; 140.5; 141.4 (Ar-C); 146.5 (C-1); 157.7 (C-2).). HR-ESI–MS: 189.0988 Da
[M+H]+ measured, 189.1022 Da [M+H]+ calculated. UV absorption maxima: 195 nm, 237 nm,
317 nm.
187
188
189
190
191
192
193
194
3-Methylquinoxalin-2-ol was synthesized by the reaction of OPD with pyruvic acid. OPD (452
mg) was dissolved in 20.0 mL of 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)
buffer (1 M pH 7) and 145 µL of pyruvic acid was added. This reaction mixture was stirred in
the dark at RT for 14 h. Then the reaction mixture was fractionated by semipreparative HPLC
with the following parameters: eluent A formic acid (0.1%), eluent B methanol; flow rate 2.0
mL/min; (time (min)/%B) 0/10, 20/90, 25/90, 25.1/10, 30/10. Detection was carried out at 316
nm and fractions 22.0–23.0 min were collected. The organic solvent was removed under reduced
pressure and the product was freeze-dried. The isolated compound was characterized by
9
ACS Paragon Plus Environment

Analytical Chemistry
Page 10 of 31
1
2
3
4
5
6
7
8
9
195
196
197
198
199
UHPLC
–
DAD
–
MS/MS and by NMR (atom numbering refers to Scheme 1D): MS/MS: 161.0 Da
[M+H]+, 133.0 Da, 92.0 Da; 1H-NMR (DMSO- d6): δ 2.40 (s, 3H, H-3); 7.26 (m, 2H, Ar-H);
7.46 (m, 1H, Ar-H); 7.69 (m, 1H, Ar-H), 12.29 (s, 1H, OH), 13C-NMR (DMSO- d6): δ 20.4 (C-
3); 115.1; 122.9; 127.8; 129.1; 131.6; 131.8; 154.8; 159.1 (Ar-C). UV absorption maxima: 198
nm, 227 nm, 339 nm.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
1-(1H-Benzo[d]imidazol-2-yl)ethane-1,2-diol (1-BIME) was synthesized according to Voigt et
al.31 Briefly, 189 mg of glyceraldehyde was dissolved in 20.0 mL of phosphate buffer (0.1 M pH
7.4) containing 208 mM OPD. After 16 h in the dark at RT, 20 mL of water was added. The
reaction mixture was extracted three times with 50 mL of ethyl acetate. The organic phases were
combined and dried over sodium sulfate. The remaining solvent was evaporated to dryness and
the residue purified using a LiChroprep RP-18 column (20 cm x 3 cm, Merck-Millipore,
Schwalbach, Germany). Elution was performed with water/methanol (90:10). Fractions
containing the target substance were combined and further purified by semipreparative HPLC
using the same parameters as for 3-methylquinoxalin-2-ol (see above). Injection volume was set
to 500 µL, chromatograms were recorded at 272 nm and fractions 10.1–11.1 min were collected.
The organic solvent was evaporated and the product was freeze-dried. The product was analyzed
by NMR (atom numbering refers to Scheme 1E) and UHPLC
MS/MS: 1H-NMR (DMSO-d6): δ
–
3.66 (dd, J = 11.0 J = 6.4 1H, H-3a); 3.82 (dd, J = 11.1 J = 4.2 1H, H-3b); 4.78 (t, J = 5.6 1H, H-
2); 7.12 (m, 2H, Ar-H); 7.50 (m, 2H, Ar-H), 13C-NMR (DMSO-d6): δ 65.2 (C-2); 69.3 (C-3);
114.8; 121.0; 138.5; 155.8 (Ar-C). UV absorption maxima: 199 nm, 213 nm, 270 nm.
215
216
217
Isolation of the Raw Material Contamination 1,4-Bis(1H-benzo[d]imidazol-2-yl)-3,4-
dihydroxybutan-1-one. Icodextrin (15.0 g) was dissolved in 100.0 mL of an aqueous solution
containing 5.38 mg/mL sodium chloride, 4.5 mg/mL sodium lactate, 0.36 mg/mL calcium
10
ACS Paragon Plus Environment

Page 11 of 31
Analytical Chemistry
1
2
3
218
4
chloride dihydrate, and 0.12 mg/mL magnesium chloride hexahydrate. Twenty-seven microliters
of a 2% HEPES-buffered (1 M, pH 7) OPD solution was added to 75.0 mL of this icodextrin-
containing solution. The reaction mixture was stirred for 3 h in the dark. Afterward, the solution
was divided into three aliquots, which were separately applied to a Strata-X solid phase
extraction (SPE) column (Phenomenex, Aschaffenburg, Germany, 1 g/20 mL). Each column was
washed with 70 mL of methanol/water (5:95) and 70 mL of methanol/water (50:50). The product
was eluted with 100 mL methanol/water (50:50). The solvents were evaporated to dryness and
the residue was dissolved in 20 mL of water. Then the mixture was fractionated by
5
6
7
8
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
semipreparative HPLC as follows: eluent A formic acid (0.1%), eluent B methanol; flow rate 2.0
mL/min; (time (min)/%B) 0/10, 20/60, 20.1/90, 25/90, 25.1/10 30/10. Injection volume was set
to 500 µL, chromatograms were recorded at 316 nm and fractions 15.1–16.1 min were collected.
The organic solvent was removed under reduced pressure and the product was freeze-dried.
The product was analyzed by NMR (signal assignment based on H/D exchange, COSY, HSQC,
and HMBC, atom numbering refers to Scheme 1F), HR-ESI–MS, and UHPLC–DAD–MS/MS.
A representative product ion spectrum is shown in Figure 2: 337.0 Da [M+H]+, 318.9 Da, 301.0
Da. 1H-NMR (DMSO-d6): δ 3.28 (dd, J = 16.0 J = 3.9 1H, H-2a); 3.50 (dd, J = 16.0 J = 9.1 1H,
H-2b); 4.60 (m, 1H, H-3); 4.83 (t, J = 5.0 1H, H-4); 5.10 (d, J = 6.2 1H, OH-4); 5.94 (d, J = 5.6
1H, OH-3); 7.10-7.81 (m, 8H, Ar-H) 12.26 (s, 1H, NH); 13.27 (s, 1H, NH), UV absorption
maxima: 198 nm, 238 nm, 277 nm, 311 nm.
237
238
239
240
Separation of the Derivatives of 5-HMF, 3-DG, 3-DGal, and 4-DG was achieved by UHPLC
DAD MS/MS applying the following parameters: eluent A 1% methanol and formic acid
–
–
(0.1%), eluent B methanol; flow rate: 0.4 mL/min; (time (min)/%B) –5.0/0, 10.0/0, 30.0/80,
30.1/95, 35.0/95; column temperature was set to 55°C. Aliquots of 10 µL were injected. Data
11
ACS Paragon Plus Environment

Analytical Chemistry
Page 12 of 31
1
2
3
4
5
6
241
242
acquisition started at 0.0 min and chromatograms were monitored at 316 nm. DAD spectra were
recorded from 190 nm to 490 nm.
7
8
9
243
244
245
246
247
Separation of 3,4-DDPSqx and 1,4-Bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxylbutan-1-
one was achieved chromatographically by a different UHPLC method: eluent A acetic acid
(0.075%), eluent B methanol; flow rate: 0.3 mL/min; (time (min)/%B) –5.0/5, 15.0/20, 25.0/50,
25.1/80, 30.0/80; column temperature: 55°C; injection volume: 10 µL. Data acquisition started at
0.0 min. Chromatograms were obtained at 316 nm, DAD spectra from 190 to 490 nm.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
248
249
250
251
252
253
254
255
256
Qualitative Profiling of PD Samples. Twenty microliters of a 1:1 mixture of OPD (4% in
HEPES buffer, 1 M, pH7) and diacetylquinoxaline (internal standard, 25 µg/mL in water) was
added to 80 µL of each PD sample. The solutions were mixed thoroughly, filtered through a
PVDF membrane (0.22 µm, Roth, Karlsruhe, Germany) and rested light-protected for at least 9 h
before UHPLC–DAD–MS/MS analysis. Full DAD spectra were recorded to differentiate
between quinoxaline and non-quinoxaline derivatives, followed by full mass scans and MS/MS
experiments.
As shown in previous studies, the applied derivatization conditions effectively suppressed de
novo formation of various α-dicarbonyl compounds.21, 23
257
258
Results and Discussion
259
260
261
262
Screening for Reactive Icodextrin Degradation Products. The present study intended to
investigate reactive degradation products of the polyglucose icodextrin (pGDPs) in PD fluids.
Because degradation of polyglucose under sterilization conditions was expected to form novel,
as yet unidentified products, we opted for a targeted screening for degradation products. To this
12
ACS Paragon Plus Environment

Page 13 of 31
Analytical Chemistry
1
2
3
263
4
end, OPD was used as derivatizing reagent prior to UHPLC–DAD–MS/MS analysis. In contrast
5
6
7
8
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
to the parent compounds, the derivatization products are stable, UV-active and easily ionized by
ESI. Particularly α-dicarbonyl structures are often trapped by OPD to form quinoxaline
derivatives (Schemes 1A and 1C), which can be analyzed by reversed-phase UHPLC. Because
the concentrations of carbohydrate degradation products in PD fluids are higher than in
biological samples, no derivatization agent leading to fluorescent products was necessary.32 After
derivatization with OPD, UHPLC analysis was performed following a method of Mittelmaier et
al.,23 which had been developed to separate GDPs in glucose-based PD fluids. In the present
study, however, the coelution of several pGDPs was observed indicating that polyglucose-based
PD fluids contain additional and/or different degradation products compared to glucose-based
products. For identification, UHPLC separation of all pGDPs was attempted. Because the
baseline separation of all peaks could not be achieved in one single run, two complementary
UHPLC methods were applied. The first method was optimized to separate the peaks in a
retention time window of 8.5–13.5 min, leading to a separation of four pGDPs at 9.8, 11.2, 12.2,
and 12.8 min (Figure 3). The co-elution of two peaks at a retention time of 17.5 min (Figure 3A,
signal marked with an asterisk), however, could not be avoided. In order to separate both pGDPs,
a second UHPLC method was applied using acetic acid instead of formic acid as eluent buffer
and adjusting the gradient (Figure 3B). Because the concentration of one of the compounds
(later identified as 3,4-DDPSqx) was below the detection limit of the UHPLC-DAD method,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
analysis was performed by UHPLC–MS/MS.
283
284
285
Identification of 3-DG, 3-DGal, 4-DG, 5-HMF, and 3,4-DDPS. At a retention time of 11.2
min (Figure 3A) a signal with the characteristic UV/Vis spectrum of a quinoxaline derivative
was detected. By comparison of retention times, UV/Vis spectra, mass spectra (m/z 235), and
13
ACS Paragon Plus Environment

Analytical Chemistry
Page 14 of 31
1
2
3
4
5
6
7
8
9
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
product ion spectra of the analyte and the reference compound, the signal could be assigned to 3-
DGqx, which is already known to be present in glucose as well as in icodextrin-containing PD
fluids.17, 19, 22, 23
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The peak at 12.2 min (Figure 3A) also showed a specific quinoxaline UV/Vis spectrum with
maxima at 198 nm, 237 nm, and 317 nm. UHPLC–MS/MS analysis revealed identical mass and
product ion spectra as 3-DGqx indicating that this signal could be caused by 3-DGalqx. 3-DGal is
a diastereomer of 3-DG, which is already known to be present in glucose-containing PD fluids.22
By comparison of its chromatographic and spectroscopic behavior with the independently
synthesized reference compound after derivatization with OPD, the structure could be
unequivocally assigned to the quinoxaline derivative of 3-DGal. To date, 3-DGal had not been
detected in icodextrin-based PD fluids and was consequently identified as a novel pGDP.
Another compound, which could not been assigned to any known GDP, eluted at 12.8 min
(Figure 3A). This structure also showed a characteristic UV spectrum for quinoxalines (Figure
S-5A) indicating an α-dicarbonyl structure as parent compound. UHPLC–MS/MS analysis
revealed an apparent molecular weight of m/z 235, equal to 3-DGqx and 3-DGalqx. Therefore, it
was hypothesized that the mass of 235 Da [M+H]+ indicated the presence of 4-DGqx (Scheme
1A), a diastereomer of 3-DGqx and 3-DGalqx. To verify the identity of the unknown GDP, 4-DG
was synthesized as a reference compound.30 Complete consistency of retention times, UV/Vis-,
mass-, and product ion spectra (Figures 1A/B) proved the identity of the compound. 4-DG had
not been detected in PD fluids so far and could be identified in the present study as a novel
pGDP. Because 4-DG has not been reported for glucose-containing PD fluids yet, it can be
assumed that the compound is specific for icodextrin-based products. 4-DG has been detected
before in reaction mixtures of maltose and lysine or as a degradation product of cellobiose or
14
ACS Paragon Plus Environment

Page 15 of 31
Analytical Chemistry
1
2
maltose in alkaline solution.33-35
3
309
4
5
6
7
8
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
The UV/Vis spectrum of the signal at a retention time of 9.8 min did not indicate the presence of
a quinoxaline derivative (Figure S-5B). Consequently, the parent compound of this structure is
probably not an α-dicarbonyl, but must possess another entity that reacts with OPD. To obtain
more information on the structure, HR-MS experiments were performed, resulting in an apparent
molecular mass of m/z 215.0817, which is consistent with the elemental formula of C12H10N2O2.
Based on this knowledge, it was hypothesized that the compound was a reaction product of 5-
HMF with OPD. In order to test this hypothesis, 5-HMF was reacted with OPD, which led to a
derivative showing the same retention time, UV/Vis spectrum and product ion spectrum as the
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
unknown compound when analyzed by UHPLC–DAD–MS/MS (Figures 1C/D). For detailed
characterization of the reaction product of 5-HMF with OPD, the substance was isolated by
semipreparative HPLC. NMR experiments revealed the presence of the benzimidazolyl
derivative of 5-HMF (5-HMFbfm, Scheme 1B). Although 5-HMF is a well-known GDP in PD
fluids,19 5-HMFbfm has not been covered yet by targeted screening of sugar degradation products
in PD fluids using OPD.23 Instead it has been generally analyzed separately by HPLC
–UV
without derivatization.36 A possible derivatization mechanism is displayed in Scheme S-1.
Further, a second chromatographic method was developed to analyze two unknown compounds
that co-eluted under the applied conditions (marked by an asterisk in Figure 3A). Under the
optimized separation conditions, baseline separation of both signals could be achieved. The
compound eluting at 15.1 min (Figure 3B) showed the UV/Vis spectrum of a quinoxaline
structure (Figure S-5C) indicating that the parent compound was an α-dicarbonyl. HR-MS
experiments revealed an apparent exact mass of m/z 189.0988, which is in accordance with the
elemental formula of C11H12N2O. Therefore, 3,4-DDPS (Scheme 1C), which has been reported
15
ACS Paragon Plus Environment

Analytical Chemistry
Page 16 of 31
1
2
3
4
5
6
7
8
9
332
333
334
335
336
by Mavric and Henle as a degradation product of 1,4-linked disaccharides,26 was suggested as
putative structure. To confirm this hypothesis, 3,4-DDPSqx was synthesized and characterized by
UHPLC–DAD–MS/MS. The reference compound 3,4-DDPSqx and the analyte had identical
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
retention times, UV/Vis-, mass-, and product ion spectra (Figures 1E/F). Thus, the unknown
pGDP could be identified as 3,4-DDPS, which was detected for the first time in PD fluids.
337
338
339
340
341
342
343
344
To evaluate if the products were indeed formed during thermal processing, unheated, sterile-
filtered icodextrin-containing PD fluids were analyzed in the same way. In the unheated controls,
3-DG, 3-DGal, 4-DG, 5-HMF and 3,4-DDPS were below the detection limit indicating that the
compounds were neither present in the raw materials nor formed from the raw materials during
sample workup. Further derivatization experiments were performed for different periods of time
between 9 and 24 h. The signals of the five analytes remained stable during prolonged
derivatization, suggesting that artificial formation of those products during derivatization was
very unlikely.
345
346
347
348
349
350
351
352
353
Identification of 1,4-Bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one. Another
unknown product eluted at the retention time of 14.4 min (Figure 3B). The respective UV/Vis
spectrum had four maxima at 198 nm, 238 nm, 277 nm, and 311 nm (Figure S-5D).
Consequently, the precursor could not be assigned to an α-dicarbonyl, but rather contained other
functional groups reacting with OPD. To investigate the source of this product, unheated, sterile-
filtered icodextrin-containing PD fluids were analyzed, where the compound was detected in
similar concentrations (data not shown). Its formation could not be forced by additional heating
of icodextrin, suggesting that it was rather a contamination of raw material formed during
16
ACS Paragon Plus Environment

Page 17 of 31
Analytical Chemistry
1
2
3
354
4
icodextrin production than a thermal degradation product. HR-MS analysis revealed a doubly
charged ion with m/z 169.0697 [M+2H]2+ indicating the elemental formula C18H16N4O3. The four
nitrogen atoms could be inserted by the reaction of 2 moles of OPD with 1 mole of the
degradation product leading to a parent compound with 6 carbon atoms equivalent to the glucose
backbone. Product ion spectra clearly showed a mass difference of 18 Da and of 36 Da compared
to the precursor ion indicating a loss of 2 moles of water (Figure 2). Therefore, the unknown
structure must have at least two aliphatic hydroxyl groups, which eliminates fragmentation
induced during dissociation.
5
6
7
8
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For structure identification, the raw material contaminant was isolated from the polyglucose
matrix after derivatization with OPD by semipreparative HPLC. 1H-NMR as well as COSY
spectra revealed two separated aromatic rings with four protons each. Additionally, the signals of
one of these aromatic rings were split up into four single signals at chemical shifts of 7.30 ppm,
7.37 ppm, 7.54 ppm, and 7.81 ppm, indicating that a substituent exerts a negative mesomeric
effect on the aromatic system. Furthermore, a carbonyl group was indicated by coupling of
protons of a CH2-group to a carbon atom with a chemical shift of 192.4 ppm in the HMBC
spectra. Based on 1H, COSY, HMBC, and HSQC spectra as well as H/D exchange, the
contamination of the raw material was identified to be 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-
dihydroxybutan-1-one (Scheme 1F), which could be derived from 2 moles of OPD and 1 mole
of a contaminant with two reactive functional groups.
Because benzimidazole derivatives may arise from the reaction of OPD with aldehydes or
carboxylic acids, respectively, the parent compound was investigated in more detail. Voigt and
Glomb described the formation of 1-BIME from OPD and glyceraldehyde, which corresponds to
the green part of the molecule in Scheme 1E.31 To confirm the presence of this partial structure
17
ACS Paragon Plus Environment

Analytical Chemistry
Page 18 of 31
1
2
3
4
5
6
7
8
9
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
in the contamination product, OPD was reacted with glyceraldeyde, the reaction product was
isolated and NMR spectra (1H, COSY, HMBC, and HSQC) were recorded. The aromatic protons
of 1-BIME induced signals at 7.12 ppm and 7.50 ppm, which could also be observed for 1,4-
bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one. The signal for H-2 at 4.78 ppm was
also comparable to the corresponding signal of the isolated raw material contamination. Thus, 1-
BIME induced the same signals as the green part of 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-
dihydroxybutan-1-one (Scheme 1E), proving the presence of a glyceraldehyde moiety in the raw
material contamination.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It is unlikely that the second part of 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-
one (the orange section of the molecule in Scheme 1F) was formed by the reaction of OPD and
an aldehyde group as part of an α-dicarbonyl function. In this case, derivatization with OPD
would rather lead to a quinoxaline instead of the observed benzimidazole derivative.23, 29, 37-40 In
order to confirm the structure of the orange part of 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-
dihydroxybutan-1-one, the reference 1-(1H-benzo[d]imidazol-2-yl)ethan-1-one was analyzed by
NMR. This compound showed very characteristic signals for the aromatic protons. Due to the
mesomeric effect of the carbonyl group, the signals of the four aromatic protons appear at 7.31
ppm, 7.38 ppm, 7.56 ppm, and 7.82 ppm. Comparison of NMR spectra (1H-NMR, COSY,
HMBC, and HSQC) of the reference and the isolated compound added up to equal signals for
these protons. Chemical shifts as well as multiplicity were the same for the protons bound to the
carbon atoms numbered with 3’–6’. Additionally, coupling of protons bound to C-2 and the
carbon atom 1 were observed in HMBC spectra for both molecules (Scheme 1F). Thus, these
atoms must be arranged in the same way and, therefore, the partial structure of the contaminant
must be 1-(1H-benzo[d]imidazol-2-yl)ethan-1-one.
18
ACS Paragon Plus Environment

Page 19 of 31
Analytical Chemistry
1
2
3
400
4
A possible precursor of 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one could
contain a carboxylic group at C-1’ (Scheme 1F). To test this hypothesis, OPD was reacted with
pyruvic acid followed by isolation and characterization of the reaction product. The NMR
spectra clearly revealed 3-methylquinoxalin-2-ol as the product of this reaction (Scheme 1D),
which was confirmed by NMR analysis of commercial 3-methylquinoxalin-2-ol. Both substances
resulted in identical NMR spectra, which, in contrast, were not included in the spectra of 1,4-
bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one. Therefore, a partial quinoxalinol
structure of the raw material contamination could be excluded. Consequently, it is unlikely that
C-1’ (Scheme 1F) is derived from a carboxylic group. Because it could be excluded that the
second benzoimidazole derived from an α-keto acid or an α-dicarbonyl, further studies are
required to elucidate the full structure of the parent compound of 1,4-bis(1H-benzo[d]imidazol-
2-yl)-3,4-dihydroxybutan-1-one. A possible explanation could be, for example, that the carbonyl
group is released from a precursor structure after/during the derivatization step.
5
6
7
8
401
402
403
404
405
406
407
408
409
410
411
412
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
413
414
415
416
417
418
419
420
421
422
The compounds 4-DG, 3-DGal, and 3,4-DDPS have not been analyzed in icodextrin-containing
products before. 4-DG and 3,4-DDPS seem to be typical degradation products of icodextrin,
because they have not been detected so far in glucose-containing PD fluids. Mavric and Henle
reported 3,4-DDPS as a degradation product of lactose.26 Recently, Smuda and Glomb described
3,4-DDPS in Maillard model reaction mixtures of maltose and lysine.33 The observation that 3,4-
DDPS has been detected so far only as degradation product of lactose, maltose, and icodextrin
indicates that a 1,4-glycosidic link is required for its formation and that it is, therefore, not
present in glucose-based PD fluids. Smuda and Glomb suggested a pathway of formation via 3-
deoxypentosone, whereas Mavric and Henle proposed formation via 4-DG.26 Since no 3-
deoxypentosone, but 4-DG was detected in the present study, the reaction mechanism described
19
ACS Paragon Plus Environment

Analytical Chemistry
Page 20 of 31
1
2
3
4
5
6
7
8
9
423
424
425
426
by Mavric and Henle seems more likely to occur during sterilization of icodextrin-based PD
fluids. 3,4-DDPS formation via 3-deoxypentosone, however, cannot be excluded, because the
concentration levels of 3,4-DDPS were very low and, thus, many intermediates like 3-
deoxypentosone could be present in quantities below the limit of detection.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
427
428
Conclusion
429
430
431
432
433
434
435
The present study identified 4-DG, 3-DGal, and 3,4-DDPS as novel degradation products in
polyglucose-containing PD fluids. Additionally, a hitherto unknown raw material contaminant,
which is converted to 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one during
derivatization, has been detected in the samples indicating that contamination of the icodextrin
raw material may be an important issue. Furthermore, 3-DG and 5-HMF were analyzed in
icodextrin-based PD fluids and it was shown that 5-HMF can by covered in parallel with the α-
dicarbonyl degradation products after derivatization with OPD yielding 5-HMFbfm.
436
437
438
439
440
441
442
Knowledge on the structures of the predominant reactive carbonyl compounds in polyglucose-
based PD fluids is a prerequisite for structure-specific risk assessment as previously carried out
for GDPs.9 Furthermore, measures to reduce the load of process contaminants can only be
developed if the nature and consequently the formation mechanisms of those contaminants are
known. In the past, for example, the content of glucose-derived degradation products could be
effectively reduced by changing the pH value during sterilization,17, 41 by alternative heating
methods 42 or by removal of GDPs after production using affinity absorption.43
443
20
ACS Paragon Plus Environment

Page 21 of 31
Analytical Chemistry
1
2
3
444
4
Acknowledgment
5
6
7
445
We thank Prof. Dr. Ivana Ivanović-Burmazović, Friedrich-Alexander-Universität Erlangen-
8
9
446
Nürnberg, for providing the maXis® mass spectrometer equipment and the Deutsche
10
11
12
13
447
Forschungsgemeinschaft (DFG) for their contribution to the applied UHPLC
No. INST 90/631-1 FUGG).
–MS/MS unit (grant
448
449
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
450
451
452
453
Supporting Information Available including NMR spectra of 4-DGqx and 5-HMFbfm, UV/Vis
spectra of 4-DGqx, 5-HMFbfm, 3,4-DPPSqx, and 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-
dihydroxybutan-1-one, and a scheme with the proposed formation of 5-HMFbfm. This material is
available free of charge via the Internet at http://pubs.acs.org.
454
21
ACS Paragon Plus Environment

Analytical Chemistry
Page 22 of 31
1
2
3
4
5
6
7
8
9
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
References
(1) Sharma, A.; Blake, P. G., In The Kidney, ed. B. M. Brenner, F. C. Rector. Saunders: Estados
Unidos, 2008, vol. 2, pp 2007-2037.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(2) Frampton, J. E.; Plosker, G. L. Drugs 2003, 63, 2079-2105.
(3) Anet, E. F. L. J. Advan. Carbohydr. Chem. 1964, 19, 181-218.
(4) Kjellstrand, P.; Martinson, E.; Wieslander, A.; Kjellstrand, K.; Jeppsson, E.; Svensson, E.;
Jarkelid, L.; Linden, T.; Olsson, L. F. Perit. Dial. Int. 2001, 21, 338-344.
(5) Nakayama, M.; Kawaguchi, Y.; Yamada, K.; Hasegawa, T.; Takazoe, K.; Katoh, N.;
Hayakawa, H.; Osaka, N.; Yamamoto, H.; Ogawa, A.; Kubo, H.; Shigematsu, T.; Sakai, O.;
Horiuchi, S. Kidney Int. 1997, 51, 182-186.
(6) Tauer, A.; Knerr, T.; Niwa, T.; Schaub, T. P.; Lage, C.; Passlick-Deetjen, J.; Pischetsrieder,
M. Biochem. Biophys. Res. Commun. 2001, 280, 1408-1414.
(7) Tauer, A.; Zhang, X.; Schaub, T. P.; Zimmeck, T.; Niwa, T.; Passlick-Deetjen, J.;
Pischetsrieder, M. Am. J. Kidney Dis. 2003, 41, S57-60.
(8) Mittelmaier, S.; Pischetsrieder, M. Anal. Chem. 2011, 83, 9660-9668.
(9) Distler, L.; Georgieva, A.; Kenkel, I.; Huppert, J.; Pischetsrieder, M. Chem. Res. Toxicol.
2014, 27, 1421-1430.
(10) Honda, K.; Nitta, K.; Horita, S.; Yumura, W.; Nihei, H.; Nagai, R.; Ikeda, K.; Horiuchi, S.
Nephrol. Dial. Transplant. 1999, 14, 1541-1549.
(11) Park, M. S.; Lee, H. A.; Chu, W. S.; Yang, D. H.; Hwang, S. D. Perit. Dial. Int. 2000, 20,
452-460.
22
ACS Paragon Plus Environment

Page 23 of 31
Analytical Chemistry
1
2
3
477
4
(12) Linden, T.; Cohen, A.; Deppisch, R.; Kjellstrand, P.; Wieslander, A. Kidney Int. 2002, 62,
697-703.
5
6
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
7
8
9
(13) Witowski, J.; Korybalska, K.; Wisniewska, J.; Breborowicz, A.; Gahl, G. M.; Frei, U.;
Passlick-Deetjen, J.; Jorres, A. J. Am. Soc. Nephrol. 2000, 11, 729-739.
(14) Wieslander, A. P.; Nordin, M. K.; Kjellstrand, P. T.; Boberg, U. C. Kidney Int. 1991, 40, 77-
79.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(15) Cooker, L. A.; Luneburg, P.; Faict, D.; Choo, C.; Holmes, C. J. Perit. Dial. Int. 1997, 17,
373-378.
(16) Lee, E.; Seo, E. Y.; Kwon, Y.; Ha, H. Anal. Chem. 2011, 83, 1518-1522.
(17) Zimmeck, T.; Tauer, A.; Fuenfrocken, M.; Pischestrieder, M. Perit. Dial. Int. 2002, 22, 350-
356.
(18) Nilsson-Thorell, C. B.; Muscalu, N.; Andren, A. H.; Kjellstrand, P. T.; Wieslander, A. P.
Perit. Dial. Int. 1993, 13, 208-213.
(19) Frischmann, M.; Spitzer, J.; Fünfrocken, M.; Mittelmaier, S.; Deckert, M.; Fichert, T.;
Pischetsrieder, M. Biomed. Chromatogr. 2009, 23, 843-851.
(20) Linden, T.; Forsback, G.; Deppisch, R.; Henle, T.; Wieslander, A. Perit. Dial. Int. 1998, 18,
290-293.
(21) Mittelmaier, S.; Fünfrocken, M.; Fenn, D.; Fichert, T.; Pischetsrieder, M. J. Chromatogr. B
Analyt. Technol. Biomed. Life Sci. 2010, 878, 877-882.
(22) Mittelmaier, S.; Funfrocken, M.; Fenn, D.; Pischetsrieder, M. Anal. Bioanal. Chem. 2011,
399, 1689-1697.
(23) Mittelmaier, S.; Funfrocken, M.; Fenn, D.; Berlich, R.; Pischetsrieder, M. Anal. Bioanal.
Chem. 2011, 401, 1183-1193.
23
ACS Paragon Plus Environment

Analytical Chemistry
Page 24 of 31
1
2
3
4
5
6
7
8
9
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
(24) Ueda, Y.; Miyata, T.; Goffin, E.; Yoshino, A.; Inagi, R.; Ishibashi, Y.; Izuhara, Y.; Saito,
A.; Kurokawa, K.; Van Ypersele De Strihou, C. Kidney Int. 2000, 58, 2518-2524.
(25) Hollnagel, A.; Kroh, L. W. J. Agric. Food Chem. 2000, 48, 6219-6226.
(26) Mavric, E.; Henle, T. Eur. Food Res. Technol. 2006, 223, 803-810.
(27) Kramhöller, B.; Pischetsrieder, M.; Severin, T. J. Agric. Food Chem. 1993, 41, 347-351.
(28) Hollnagel, A.; Kroh, L. W. J. Agric. Food Chem. 2002, 50, 1659-1664.
(29) Gensberger, S.; Mittelmaier, S.; Glomb, M. A.; Pischetsrieder, M. Anal. Bioanal. Chem.
2012, 403, 2923-2931.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(30) Broberg, A.; Kenne, L.; Pedersen, M. Carbohydr. Res. 1998, 306, 171-175.
(31) Voigt, M.; Smuda, M.; Pfahler, C.; Glomb, M. A. J. Agric. Food Chem. 2010, 58, 5685-
5691.
(32) Min, J. Z.; Yamamoto, M.; Yu, H. F.; Higashi, T.; Toyo'oka, T. Anal. Biochem. 2012, 424,
187-194.
(33) Smuda, M.; Glomb, M. A. J. Agric. Food Chem. 2011, 59, 13254-13264.
(34) Whistler, R. L.; BeMiller, J. N. J. Am. Chem. Soc. 1960, 82, 3705-3707.
(35) Morita, N.; Mizutani, M.; Hayashik, K.; Kirihata, M.; Ichimoto, I.; Ueda, H.; Takagi, M.
Bull. Univ. Osaka Pref. Ser. B 1983, 35, 59-70.
(36) Wieslander, A. P.; Deppisch, R.; Svensson, E.; Forsback, G.; Speidel, R.; Rippe, B. Perit.
Dial. Int. 1995, 15, 158-164.
(37) Bednarski, W.; Jedrychowski, L.; Hammond, E.; Nikolov, Z. J. Dairy Sci. 1989, 72, 2474-
2477.
(38) Glomb, M. A.; Tschirnich, R. J. Agric. Food Chem. 2001, 49, 5543-5550.
(39) Degen, J.; Hellwig, M.; Henle, T. J. Agric. Food Chem. 2012, 60, 7071-7079.
24
ACS Paragon Plus Environment

Page 25 of 31
Analytical Chemistry
1
2
3
523
4
(40) Gensberger, S.; Glomb, M. A.; Pischetsrieder, M. J. Agric. Food Chem. 2013, 61, 10238-
10245.
5
6
7
8
524
525
526
527
528
529
(41) Erixon, M.; Wieslander, A.; Linden, T.; Carlsson, O.; Forsback, G.; Svensson, E.; Jonsson,
J. A.; Kjellstrand, P. Perit. Dial. Int. 2006, 26, 490-497.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(42) Shapira, A.; Shazman, A.; Ungar, Y.; Shimoni, E. Mol. Nutr. Food Res. 2007, 51, 473-478.
(43) Ishikawa, N.; Miyata, T.; Ueda, Y.; Inagi, R.; Izuhara, Y.; Yuzawa, H.; Onogi, H.; Nishina,
M.; Nangaku, M.; Van Ypersele De Strihou, C.; Kurokawa, K. Kidney Int. 2003, 63, 331-339.
530
531
25
ACS Paragon Plus Environment

Analytical Chemistry
Page 26 of 31
1
2
3
4
532
Figure Captions
5
6
7
8
9
533
534
535
536
537
Scheme 1. Structures of the degradation products (A) 4-DG, (B) 5-HMF, (C) 3,4-DDPS and
their corresponding OPD derivatives; formation of (D) 3-methylquinoxalin-2-ol and (E) 1-BIME
by the reaction of OPD with pyruvic acid and glyceraldehyde, respectively; (F) structure of the
isolated raw material contamination 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-
one after derivatization with OPD.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
538
539
540
Figure 1. Product ion spectra of compounds after conversion by OPD in PD fluids (a, c, e) and
the respective synthesized references (b, d, f): (A/B) 4-DGqx; (C/D) 5-HMFbfm and (E/F) 3,4-
DDPSqx.
541
542
543
Figure 2. Product ion spectrum of the raw material contamination in icodextrin-based PD
solutions after conversion to 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one by
OPD.
544
545
546
547
548
549
550
551
552
553
Figure 3. (A) UHPLC-DAD chromatogram of a typical icodextrin-containing PD fluid after
adding diacetylquinoxaline as internal standard and derivatization with OPD, recorded at 316
nm. The chromatographic parameters were optimized for the separation of 5-HMFbfm and 3-
DGqx, 3-DGalqx, and 4-DGqx. Signals within the grey rectangle are caused by OPD degradation,
the asterisk indicates two coeluting products 3,4-DDPSqx and the contamination of the raw
material 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one ; (B) UHPLC–MS/MS
chromatogram of a typical icodextrin-containing PD fluid after trapping the degradation products
with OPD. Mass transitions m/z 189–171 (grey) and m/z 337–319 (black) are shown.
Chromatographic separation was optimized for 3,4-DDPSqx and the contamination of the raw
material 1,4-bis(1H-benzo[d]imidazol-2-yl)-3,4-dihydroxybutan-1-one.
26
ACS Paragon Plus Environment

Page 27 of 31
Analytical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1
27
ACS Paragon Plus Environment

Analytical Chemistry
Page 28 of 31
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1
28
ACS Paragon Plus Environment

Page 29 of 31
Analytical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2
29
ACS Paragon Plus Environment