- Email: [email protected]
A riboswitch sensor to determine vitamin B12 in fermented foods
Accepted Manuscript Analytical Methods A Riboswitch Sensor to Determine Vitamin B12 in Fermented Foods Xuan Zhu, Xiaofeng Wang, Chen Zhang, Xiaoqi Wang, Qing Gu PII: DOI: Reference:
S0308-8146(14)01901-3 http://dx.doi.org/10.1016/j.foodchem.2014.11.163 FOCH 16848
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
1 September 2014 23 November 2014 29 November 2014
Please cite this article as: Zhu, X., Wang, X., Zhang, C., Wang, X., Gu, Q., A Riboswitch Sensor to Determine Vitamin B12 in Fermented Foods, Food Chemistry (2014), doi: http://dx.doi.org/10.1016/j.foodchem.2014.11.163
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
A Riboswitch Sensor to Determine Vitamin B12 in Fermented Foods
2
Xuan Zhu, Xiaofeng Wang, Chen Zhang, Xiaoqi Wang, Qing Gu*
3
Zhejiang Gongshang University, Key Laboratory for Food Microbial Technology of Zhejiang Province,
4
No. 18 Xuezheng Str., Hangzhou, Zhejiang Province, 310018 China
5
*corresponding author ( Phone: +86-571-28008902; Fax: +86-571-28008900; e-mail:
6
[email protected])
7
8
Abstract
9
We describe a sensitive and selective method for determination of vitamin B12
10
content in fermented foods using riboswitch sensor. A riboswitch amplicon from
11
Propionibacterium freudenreichii was cloned in p519NGFP vector in Escherichia coli
12
BL21 (DE3). The expression of green fluorescence protein was revers correlated to
13
the concentrations of adenosylcobalamin. Adenosylcobalamin directly binds to
14
riboswitch region leading to conformational changes in the secondary structure of
15
mRNA, thus inhibiting expression. After various examinations, a standard curve was
16
obtained from 10 to 1000 ng/mL of cyanocobalamin. The limit of determination is 10
17
ng/mL. The inter-assay coefficients of variation were 7.5% for the range of 10-1000
18
ng/mL. The recovery of this method was 92.3%. This method has no or less responses
19
to nucleic acid, pseudovitamin B12, vitamin B12 bound to intrinsic factor and
20
haptocorrin. The riboswitch sensor results were similar with HPLC, but they were Ca.
21
24% lower than the microbiological assay results.
22
Keywords: riboswitch; green fluorescence protein; cobalamin
23 1
24
25
1. Introduction
26
Vitamin B12, one of water-soluble vitamins, is the general name for natural occurring
27
cobalt organometallic compounds containing substances and is involved in wind
28
range of biochemical processes such as DNA synthesis and regulation, fatty acid
29
synthesis, amino acid metabolism as well as energy production. Vitamin B12
30
deficiency causesmitotic disorder, neuropathy, nervous system disease, and pernicious
31
anaemia(Allen, 2010). To prevent such a fatal deficiency disease, daily intake of 2.4
32
µg vitamin B12 is advised (Rucker, Suttie, McCormick, & Machilin, 2001).
33
Vitamin B12 is exclusively synthesized by some bacteria and archaea and is
34
accumulated in animal bodies by rumen bacteria (Martens, Barg, Warren, & Jahn,
35
2002, Perlman, 1959). Therefore, animal based foods and some fermented plant based
36
foods are considered as the main dietary sources of vitamin B12 for human. (Keuth &
37
Bisping, 1994). Most of the data of Vitamin B12content in foods were acquired by a
38
microbiological assay (MA). The AOAC reference analytical method excels in its
39
sensitivity,
40
consuming.Moreover, results obtained by MA overestimate Vitamin B12 content in
41
foods due to deoxyribonucleutide and other corrinoids (Denter & Bisping, 1994). In
42
addition to MA, HPLC based determination methods, typicallyReversed-phase HPLC
43
using UV and fluorescence, are also developed for analyzing vitamin B12 in fortified
44
foods, vitamin supplement tablets, and infant formula.(Campos-Gimenez, Fontannaz,
45
Trisconi, Kilinc, Gimenez, Andrieux, et al., 2012; Kirchner, Degenhardt, Raffler, &
46
Nelson, 2012; Schimpf, Spiegel, Thompson, & Dowell, 2012; Vyas, O'Kane, &
but
suffers
from
poor
selectivity,
labor
intensity
and
time
2
47
Dowell, 2012). However, most of these methods are only applied for vitamin tablets
48
or milk while none was reported to be used in analyzing low concentrations of
49
vitamin B12 in fermentation foods. Lou and colleaguesreported successful detecting
50
of vitamin B12 in food products by HPLC-ESI-MS.(Luo, Chen, Ding, Tang, & Yao,
51
2006) Furthermore, an UPLC method was successfully applied to detecting vitamin
52
B12 in fermented food withhigh sensitivity (15 ng/mL) and selectivityof active
53
vitamin B12 againstpseudovitamin B12 (Chamlagain, Edelmann, Kariluoto,
54
Ollilainen, & Piironen, 2014). However,
55
complicated sample preparation.
56
Riboswtich is a 5’-untranslated leader sequence of the correspondent mRNA, which
57
regulates translation initiation and gene expression by binding to a specific
58
molecule(Winkler & Breaker, 2005). Expression of cobalamin biosynthetic cob
59
operon and transporter btuB gene was repressed by presence of vitamin B12,
60
particular Adenosyl-cobalamin (ADCBL) (Fowler, Brown, & Li, 2008). It has been
61
reported that ADCBL directly binds to riboswitch region leading to a conformational
62
change in the secondary structure of mRNA which masks the ribosome-binding site
63
(RBS), and thus inhibiting gene expression (Nahvi, Sudarsan, Ebert, Zou, Brown, &
64
Breaker, 2002). Vitreschak et al have summarized almost 200 B12 elements from 66
65
bacterial genomes by computer multiple alignment (Vitreschak, Rodionov, Mironov,
66
& Gelfand, 2003a). A fragment in front of cbiB in Propionibactiumshermaniiwas
67
predicted
68
frombtuBriboswitch
69
btuBriboswitchbased onE. coliriboswitchesin detecting low concentrations vitamin
70
B12 in vitro (Fowler, Brown, & Li, 2008). But the expression of green fluorescence
as
a
chromatography methods require
cobalaminriboswitch,
which
in
al
E.
coli.
Li
et
have
has
different
reported
an
sequences engineered
3
71
protein (GFP) was also inhibited by other corrinoids.
72
In this study we aim to develop a sensitive riboswitch based method (RB) for the
73
analysis of vitamin B12 in various foods. The new method requires simple sample
74
preparation and takes 5 hours. A 2 nd polynomial curve range of the assay is 10 ng/mL
75
to 1000 ng/nL. Compared with microbiological assay (MA), results from this method
76
were minimally influenced by deoxyribonucleutide and corrinoids.
77
2. Materials and methods
78
2.1 Bacteria culture
79
E. coli DH5α and E. coli BL21 (DE3) were grown in Luria-Bertnai (LB) medium at
80
37 °C and 200 rpm. Transformants of E. coli were cultivated in LB medium
81
supplemented with Kanamycin (10 mg/mL) (Sigma-Aldrich, USA). Lactobacillus
82
delbrueckii spp. lactisDSM 20355 was used as the indicator strain in the
83
microbiological assay of vitamin B12.
84
Propionibacteriumfreudenreichiispp.
85
reuteriDSM 20016 and Acetobaterpasteurianus DSM 3509 were cultured in de Man,
86
Rogosa, and Sharpe (MRS) broth (Luqiao, China) at 37 °C. Plasmid p519ngfp was
87
used for the construction of recombinant plasmid carrying cobalamin riboswitch gene.
88
All the bacteria strains and their source were listed in Table 1.
89
2.2. Riboswtich sensor construction and validation
90
The whole chromosome was extracted and purified with QIAamp DNA Mini Kit
91
(QIAGEN, Hilden, Germany) according to the instruction by the manufacturer.
92
Isolated DNA was used as template for PCR. The 213bp fragment containing a
shermanii
DSM
20270,
Lactobacillus
4
93
riboswitch and RBS from P. freudenreichii spp. shermanii DSM 20270 was amplified
94
by the primersdesigned based on the sequence in Gene Bank database (NC_014215.1
95
from 1368045 to 1368257) (Table 1). PCR condition was set as follows: 5 min at
96
94 °C, followed by 30 cycles of 30s at 94 °C, 30s at 50 °C, and 30sat 72 °C. Final
97
extension is 10 min at 72 °C. Fragment was subcloned into p519ngfp vector. The
98
resulting plasmid was amplified in E. Coli DH5α.E. coli BL21 (DE3) was used for
99
expression. All procedures were used as described previously (Sambrook J., Fritsh E.
100
F., & Maniatis T., 2001).
101
E. coli containing p519-switch-ngfp plasmid was incremented in LB media at 37 °C
102
overnight and centrifuged at 4,000 g for 5 min (Biofugepico, Heraeus Instruments,
103
Hanau, Germany), followed by wash in 0.9% sodium chloride for three times. 10 7
104
bacteria were inoculated into 10 mL of vitamin B12 test broth (Merck, Germany)
105
supplemented with 0.01, 1, and 10 mg/L of adenosylcobalamin at 37 °C for 5 hours.
106
Expression of GFP was examined by fluorescence microscope (Leica, Germany) at
107
1000x.
108
2.3. Sample preparation and vitamin B12 determination
109
Vitamin B12 was extracted from 10 g samples in 100 mL sodium acetate buffer (pH
110
6.0) with presence of KCN and heated in a water bath for 30 min at 70 °C. CNCBL
111
was used as standard substance since it is the most stable form of cobalamins.
112
10 6cfu/mL E. coli containing p519-switch-ngfp were inoculated in 9 mL of vitamin
113
B12 test broth supplemented with 1 mL of 0.01, 0.25, 0.5, 0.75, 1, and 1.25 mg/L of
114
cyanocobalamin (CNCBL) at 37 °C for 5 hours. GFP intensity was measured by
115
Cary Eclipse Fluorescence Spectrophotometer (G9800A, Agilent Technologies, US) 5
116
with excitation emission set at 420 nm/470 nm. The values were normalized by
117
absorbance at 600nm (OD600) for all samples. The standard curve was decided by
118
plotting concentrations of CNCBL on the abscissas and fluorescence intensity on
119
ordinates. The response was calculated as a 2 nd order polynomial regression. The
120
residue fluorescence intensity was expressed as a percentage activity of that with 0.01
121
mg/L cyanocobalamin. Hydroxycobalamin (HOCBL), methylcobalamin (MCBL),
122
adenosylcobalamin (ADCBL), intrinsic factor protein binding CNCBL (IF-CNCBL),
123
haptocorrin binding CNCBL (HC-CMCBL), deoxyribonucleotide, pseudovitamin B12,
124
and decomposed vitamin B12 by light at different concentrations were utilized to
125
evaluate influence of cobalamin analogues on sensitivity and selectivity of this
126
riboswitch sensor.
127
3. Results
128
3.1. Riboswtich sensor construction and examination
129
To construct a cobalaminribosensor switch, a 213bp fragment containing a
130
predictedriboswitch and RBS from P. freudenreichiispp. shermanii DSM 20270 was
131
cloned into p519ngfp between pnpt2 promoter and GFP (Fig. 1). The conversed
132
regions of this fragment shared 100% similarity with the sequencereported by
133
vitreschak et al. (Vitreschak, Rodionov, Mironov, & Gelfand, 2003b). For
134
heterologous expression and riboswitch examination, the recombinant plasmids were
135
transferred into E. coli BL21 (DE3) strain. Large-scale cultures were conducted with
136
adenosylcobalamin at 0.01, 1, and 10 mg/L. The fluorescence intensity
137
diminishedwith increasingadenosylcobalamin concentrations (Fig. 2 C D E). In
138
contrast, fluorescence intensity of E. coli with p519ngfp was stable during the 6
139
increase of adenosylcobalamin concentrations (data not shown). These results
140
indicated
141
p519ngfpplasmid and being ableto inhibit GFP expression as we assumed.
142
3.2 Determination of vitamin B12 using the riboswitch sensor
143
With the result that GFP intensity decreases with the presence of vitamin
144
B12,wenexttry to determine the standard curve of this assay. We analyzed the suitable
145
cell density for the determination of vitamin B12. When cells were seeded at a low
146
density (Ca. 103 CFU/mL), the slope was steep and sensitivity of the assay was high.
147
But it took more than 48 hours to perform a test. In contrast, when more than 10 7
148
CFU/mL cells were seeded, the range of the vitamin B12 to be measured is narrow.
149
Considering the importance of the measurement accuracy and variation, 106 CFU/mL
150
cells were used for vitamin B12 measurement of 10-1000 ng/mL.
151
3.3. Measurement of vitamin B12by the riboswitch sensor
152
E. coli cannot produce vitamin B12 de novo(Fowler, Sugiman-Marangos, Junop,
153
Brown, & Li, 2012).Thus it cannot survive in vitamin B12 test broth without extra
154
supplementation. We determined that the bacteria grew with 5 ng/mL of vitamin B12
155
in culture.Therefore, the average and the standard deviation of the GFP intensity were
156
calculated at 5 ng/mL vitamin B12. The limit of detection of vitamin B12 (i.e. 10
157
ng/mL) was considered when the concentration of vitamin B12 was three times
158
greater than the standard deviation of GFP intensity at 5 ng/mL vitamin B12. When
159
the amount of vitamin B12 was greater than 1200 ng/mL, the curve was not fitted for
160
a 2 nd order polynomial curve. Thus, the 2 nd order polynomial curve can be used in the
161
region between 10 and 1000 ng/mL.
that this predictedriboswitch has been successfully cloned into
7
162
The accuracy of measurement was evaluated by calculating inter and intra-assay
163
coefficients of variation of the results. The inter-assay coefficients of variation were
164
7.5% for the range of 10-1000 ng/mL. The intra-assay coefficients of variation were
165
calculated by analyzing pig livers (n=4), stinky tofu (n=4), and other fermented media
166
(n=4). The coefficients of variation were 4.1% for pig livers, 5.0% for stinky tofus,
167
5.4% for Vitamin B12 test broth fermented by P. freudenreichii, and 4.7% for soymilk
168
fermented by P. freudenterichii and L. reuteri respectively. There is no relationship
169
between the magnitude of the coefficient of variation and the nature of the samples.
170
The recovery of this method was 92.3%, which indicated that the extraction and
171
measurement procedures were qualified.
172
The responses of the riboswitch sensor to corrinoids and other complexes were also
173
investigated. As shown in Fig. 3 B, MCBL and OHCBL inhibited GFP expression in a
174
similar manner with a gradual slope. On the other hand, ADCBL inhibited the GFP
175
expression more than CNCBL. Therefore, according to the different responses, all
176
bioactive cobalamin should be extracted by heating in actate buffer with KCN to form
177
CNCBL before assay, when a sample is assayed by the riboswitch assay. In addition,
178
CNCBL is the most thermal stable form.
179
As shown in the panel C of Fig. 3, deoxyribonucleotide and decomposed vitamin B12
180
cannot inhibit GFP expression by shutting down the riboswitch sensor. In contrast,
181
they were detected as vitamin B12 in the microbiological assay. Vitamin B12 in foods
182
and living materials is normally bound to proteins such as intrinsic factor, haptcorrin
183
and other vitamin B12 dependent enzymes. The complex forms of cobalamin were
184
also tested. As shown in Fig. 3 C, only the high concentration of IF-CNCBL (750
185
ng/mL) and HC-CNCBL (750 ng/mL) can turn the riboswitch sensor off. No standard 8
186
curve can be obtained. Thus, releasing vitamin B12 from some proteins before
187
measurement is essential for this riboswitch sensor method.
188
3.3. Measurement of vitamin B12 in samples in comparison with other methods
189
We then compared our assay with other established methods. The results of the P.
190
freudenreichii cells via MA, HPLC, and RB were similar. When we analyze pig livers,
191
result obtained by RB showed 22.3% lower vitamin B12 content than that by the MA,
192
whilesimilar result was obtained by HPLC. In fermented soymilk, the vitamin B12
193
contents determined by the RB were respectively 92.2% of HPLC and 77.0% of MA.
194
The results of vitamin B12 contents in stinky tofu determined by RB were 103.9% of
195
HPLC and 50.8% of MA. For the fermented media by A. pasteurianus and
196
decomposed cyanocobalamin, no detectable level of cyanocobalamin was found by
197
HPLC and RB. However, the results by MA were still high. This phenomenon could
198
be due to the inherent drawbacks associated with the MA.
199
4. Discussion
200
In this study, we describe a sensitive and selective method for determination of
201
vitamin B12 content in fermented foods using riboswitch sensor. Vitreschak et al
202
found 200 of B12 riboswitch elements from 66 bacterial genomes by computer
203
multiple alignment (Vitreschak, Rodionov, Mironov, & Gelfand, 2003a). Moreover, in
204
the website of GenomeNet, a B12 riboswitch element responsible for expression of
205
ABC transporter substrate-binding protein was found in the genome of P. propionicum
206
F0230a (from 16287 to16468 bp). Fowler et al. used a FACS-based method to achieve
207
engineering artificial riboswitches (Campos-Gimenez, et al., 2012). Furthermore,
208
these riboswitch tools were used to explore intermolecular interactions of a vitamin 9
209
B12 binding protein (Fowler, Sugiman-Marangos, Junop, Brown, & Li, 2012). Even a
210
very low ADCBL concentration resulted in strong repression of reporter expression.
211
These artificial elements were so sensitive that it has very narrow detection range,
212
limiting theirapplication in a quantitative method.
213
E. coli is an idea host of B12 riboswitch sensor, as it cannot produce vitamin B12 de
214
novo. In our work, the B12 riboswitch element from P. freudenreichii spp. shermanii
215
DSM 20270 has a definitely different structure with riboswitches in E. coli (Fowler,
216
Brown, & Li, 2008), although the riboswitchis regulated by similar mechanism which
217
covers RBS section with an antisequence (Fig. 2 B). Our riboswitch has a short right
218
arm and “CCCC” sequence head (Fig. 2 A), which is responsible for folding of RNA
219
structure. The reporter gene repression as the result of presence vitamin B12 (Fig. 2)
220
was also related to transport protein efficiency of the ADCBL or precursors into
221
cytoplasm (Kirchner, Degenhardt, Raffler, & Nelson, 2012). The extent of GFP
222
repression was affected by transport proteins, as they have various affinities with
223
different cobamindes (Fig. 3 B). ADCBL precursors or cobamindes are quickly
224
converted to ADCBL by metabolic enzymes following transport (Kirchner,
225
Degenhardt, Raffler, & Nelson, 2012). Thus these substances can also be measured. In
226
addition, the affinity of riboswitch was also a crucial factor for GFP repression. Some
227
of riboswitches recognized MCBL and OHCBL with more than 500-fold higher
228
affinity than ADCBL (Johnson, Reyes, Polaski, & Batey, 2012). This structure
229
containing a short right arm has a high sensitivity with derivatives with small
230
upper-axial moieties (Johnson, Reyes, Polaski, & Batey, 2012). Unlike the E. coli
231
btuB riboswitch that selectively binds adenosylcobalamin (Kirchner, Degenhardt,
232
Raffler, & Nelson, 2012), this riboswitch responsesto various bioactive cobalamin 10
233
such as CNCBL, OHCBL, and MCBL besides ADCBL (Fig. 3B). However, this
234
riboswitch sensor has few responses on pseudovitamin B12 and light-decomposed
235
vitamin B12, as their upper-moieties cannot bind this section.
236
The GFP expression was partly inhibited by vitamin B12 bound to IF. In contrast, HC
237
almost blocked the determination completely (Fig. 3C). Some researchers
238
demonstrated that vitamin B12 binds with IF via two sites (Schneider & Stroinski,
239
1987). One is dimethylbenzimidazole, which was bound with cobalt as lower-axis.
240
The other is a part of the A and B ring of porphyrin ring. As a result, the upper ligand
241
of the corrin ring was not affected by IF. In contrast, HC binds to cobalamin via only
242
one site in porphyrin ring. Although the upper and lower ligands were not covered by
243
HC, a steric hindrance occurs between a large group and the HC molecule. A
244
distortion of bond may be the reason for the less response from HCCBL in RB
245
method.
246
The MA is recognized as an official standard method for determination of low levels
247
of vitamin B12 in foods and biological material. However, it was consistently reported
248
to yield higher vitamin B12 results than HPLC and other methods. Thisdeviation
249
appeared due to their response to corrinoids which are inactive for human such as
250
pseudovitamin B12 and nucleic acid (Santos, Vera, Lamosa, de Valdez, de Vos, Santos,
251
et al., 2007; Wongyai, 2000). In our previous work, we found that A. pasteuriamus
252
could produce corrinoids by which L. delbrueckii could survive in vitamin B12 test
253
broth. But no detected bioactive vitamin B12 could be found via HPLC method. As
254
shown in table 2, Our RB method had no response to pseudovitamin B12 from A.
255
pasteruiamus. Moreover, RB method has shown no response to nucleic acid in Fig.
256
3C. L. derlbrueckii can utilize the nucleic acid pool to survive in the media without 11
257
vitamin B12, as vitamin B12 wasinvolved in the nucleic acid synthesisas coenzyme
258
(Schneider & Stroinski, 1987). Some scientists reported a 20% higher vitamin B12
259
content in foods by MA compared to HPLC (Heudi, Kilinc, Fontannaz, & Marley,
260
2006). Poor accuracy of the MA was further shown in analysis of shellfish, meat and
261
milk, where the results from MA were up to 4-5 times higher than HPLC and other
262
methods (Watanabe, Yabuta, Tanioka, & Bito, 2014).
263
In present work, results from the RB method were similar with that of HPLC (table 2),
264
of which no analogues of cobalamin could be detected. However, for the samples of
265
stinky tofu, the results of RB were still 13% higher than HPLC. HPLC method,
266
particularly UPLC method, is capable of sensitive quantification of the vitamin B12
267
content in microbial cells, and fermented matrices.(Chamlagain, Edelmann, Kariluoto,
268
Ollilainen, & Piironen, 2014). However, this method requires a complicated
269
preparation procedure to collect free vitamin B12 released from protein. Thus, in
270
some cases, vitamin B12 bound to protein cannot be detected. For example, in
271
fermented tofu that contains lots of protein, the recovery of vitamin B12 was less than
272
75% (Zhu & Bisping, 2012). RB method was less influencedby bound protein.
273
5. Conclusion
274
In this study, riboswitch sensor was adopted as a sensitive and specific method of
275
vitamin B12 quantification in fermented foods with short incubation. The method
276
allowed for selective determination of bioactive vitamin B12, thus avoiding the
277
influence from nucleic acid and other inactive corrinoids as in the MA. This method
278
also allowed for eliminating the influence of proteins as in the HPLC method.
279 12
280
6. Acknowledgments
281
This project was supported by the National High Technology Research and
282
Development Program ("863" Program) of China (2014AA022210), Scientific
283
Research Foundation of Zhejiang Gongshang University, and Scientific Research
284
Foundation of Education Department of Zhejiang Province (1110kz0414207).
285
7. References
286
Allen, L. H. (2010). Bioavailability of vitamin B12. Int J Vitam Nutr Res, 80(4-5), 330-335.
287 288
Barry, T. N., Hoskin, S. O., & Wilson, P. R. (2002). Novel forages for growth and health in farmed deer. N Z Vet J, 50(6), 244-251.
289 290 291 292
Campos-Gimenez, E., Fontannaz, P., Trisconi, M. J., Kilinc, T., Gimenez, C., Andrieux, P., & Nelson, M. (2012). Determination of vitamin B12 in infant formula and adult nutritionals by liquid chromatography/UV detection with immunoaffinity extraction: First Action 2011.08. J AOAC Int, 95(2), 307-312.
293 294 295
Chamlagain, B., Edelmann, M., Kariluoto, S., Ollilainen, V., & Piironen, V. (2014). Ultra-high performance liquid chromatographic and mass spectrometric analysis of active vitamin B12 in cells of Propionibacterium and fermented cereal matrices. Food Chem, 166, 630-638.
296 297
Denter, J., & Bisping, B. (1994). Formation of B-vitamins by bacteria during the soaking process of soybeans for tempe fermentation. Int J Food Microbiol, 22(1), 23-31.
298 299
Fowler, C. C., Brown, E. D., & Li, Y. (2008). A FACS-based approach to engineering artificial riboswitches. Chembiochem, 9(12), 1906-1911.
300 301 302
Fowler, C. C., Sugiman-Marangos, S., Junop, M. S., Brown, E. D., & Li, Y. (2012). Exploring intermolecular interactions of a substrate binding protein using a riboswitch-based sensor. Chem Biol, 20(12), 1502-1512.
303 304 305
Heudi, O., Kilinc, T., Fontannaz, P., & Marley, E. (2006). Determination of Vitamin B12 in food products and in premixes by reversed-phase high performance liquid chromatography and immunoaffinity extraction. J Chromatogr A, 1101(1-2), 63-68.
306 307
Johnson, J. E., Jr., Reyes, F. E., Polaski, J. T., & Batey, R. T. (2012). B12 cofactors directly stabilize an mRNA regulatory switch. Nature, 492(7427), 133-137.
308 309 310
Keuth, S., & Bisping, B. (1994). Vitamin B12 production by Citrobacter freundii or Klebsiella pneumoniae during tempeh fermentation and proof of enterotoxin absence by PCR. App Environ Microb, 60(5), 1495-1499.
311 312
Kirchner, U., Degenhardt, K., Raffler, G., & Nelson, M. (2012). Determination of vitamin B12 in infant formula and adult nutritionals using HPLC after purification on an immunoaffinity column: 13
313
first action 2011.09. J AOAC Int, 95(4), 933-936.
314 315
Luo, X., Chen, B., Ding, L., Tang, F., & Yao, S. Z. (2006). HPLC-ESI-MS analysis of vitamin B12 in food products and in multivitamins-multimineral tablets. Analytica Chimica Acta, 562(2), 185-189.
316 317
Martens, J. H., Barg, H., Warren, M. J., & Jahn, D. (2002). Microbial production of vitamin B12. Appl Microbiol Biotechnol, 58(3), 275-285.
318 319
Nahvi, A., Sudarsan, N., Ebert, M. S., Zou, X., Brown, K. L., & Breaker, R. R. (2002). Genetic control by a metabolite binding mRNA. Chem Biol, 9(9), 1043.
320
Perlman, D. (1959). Microbial synthesis of cobamides. Adv Appl Microbiol, 1, 87-122.
321 322
Rucker, B. R., Suttie, J. W., McCormick, B. D., & Machilin, L. J. (2001). Handbook of vitamin. New York: Marcel Dekker, Inc.
323 324
Sambrook J., Fritsh E. F., & Maniatis T. (2001). Molecular cloning: a laboratory manual. Cold spring harbor laboratory In). New York: Cold Spring Harbor.
325 326 327
Santos, F., Vera, J. L., Lamosa, P., de Valdez, G. F., de Vos, W. M., Santos, H., Sesma, F., & Hugenholtz, J. (2007). Pseudovitamin is the corrinoid produced by Lactobacillus reuteri CRL1098 under anaerobic conditions. FEBS Letters, 581(25), 4865-4870.
328 329
Schimpf, K., Spiegel, R., Thompson, L., & Dowell, D. (2012). Determination of vitamin B12 in infant formula and adult nutritionals by HPLC: First Action 2011.10. J AOAC Int, 95(2), 313-318.
330 331
Schneider, Z., & Stroinski, A. (1987). Comprehensive B12: chemistry, biochemistry, nutrition, ecology, medicine. Berlin: Verlag Walter de Gruyter.
332 333 334
Vitreschak, A. G., Rodionov, D. A., Mironov, A. A., & Gelfand, M. S. (2003a). Regulation of the vitamin B12 metabolism and transport in bacteria by a conserved RNA structural element. Rna, 9(9), 1084-1097.
335 336 337
Vitreschak, A. G., Rodionov, D. A., Mironov, A. A., & Gelfand, M. S. (2003b). Regulation of the vitamin B12 metabolism and transport in bacteria by a conserved RNA structural element. Rna, 9(9), 1084-1097.
338 339 340
Vyas, P., O'Kane, A. A., & Dowell, D. (2012). Determination of vitamin B12 in infant formula and adult nutritionals by surface plasmon resonance: First Action 2011.16 (test kit method). J AOAC Int, 95(2), 329-334.
341
Watanabe, F. (2007). Vitamin B12 sources and bioavailability. Exp Biol Med, 232(10), 1266-1274.
342 343 344
Watanabe, F., Yabuta, Y., Tanioka, Y., & Bito, T. (2014). Biologically active vitamin B12 compounds in foods for preventing deficiency among vegetarians and elderly subjects. J Agric Food Chem, 61(28), 6769-6775.
345 346
Winkler, W. C., & Breaker, R. R. (2005). Regulation of bacterial gene expression by riboswitches. Annu Rev Microbiol, 59, 487-517.
347 348
Wongyai, S. (2000). Determination of vitamin B12 in multivitamin tablets by multimode high-performance liquid chromatography. J Chromatogr A, 870(1-2), 217-220. 14
349 350
Zhu, X., & Bisping, B. (2012). Determination of Vitamin B12 in Fermented Soybean Products by High-Performance Liquid Chromatography. J biotech, 150, 330.
351 352 353 354 355 356 357 358 359
Figure Caption
360 361 362 363 364
Fig. 1 Genetic maps of p519ngfp and p519-Switch-ngfp showing the insert of a riboswitch. The insert riboswitch was obtained by PCR amplication of the leader mRNA of cbib in P. freudenreichii. The restricted enzyme sites used in cloning experiments are shown.
365 366 367 368 369 370 371 372
Fig. 2 Mechanism of vitamin B12 dependantriboswitch (A and B) and expressions of GFP in E. coli BL21 with recombinant plasmid p519-Switch-ngfp under various supplementation of adenosylcobalamin (0.01, 1, and 10 mg/L) (C, D, and E). A and B: AGGAG is a RBS (ribosome binding site). The antisequestor region with P3 (CCCC) aptamer domain forms a pseudoknot under a high concentration of adenosylcobalamin. The complement antisequestor region with RBS forms a hairloop to attenuate translation.
373
379
Fig. 3 A: a standard curve for determination of vitamin B12 by the riboswitch method. 1 mL of vitamin B12 test broth with E. coli BL21 (10 6 cfu/mL) supplemented respectively with 0, 0.25, 0.5, 0.75, 1, and 1.25 µg/mL CNCBL were incubated at 37 ℃ for 5 hours. B: Expression inhibition of GFP by CNCBL (□), ADCBL (▲), MCBL (●), HOCBL (△), and Nothing (◆). C: Expression inhibition of GFP by CNCBL (□), IF-CNCBL (■), HC-CNCBL (▲), deoxyribonucleotide (◇), pseudovitamin B12 (◆),
380
and decomposed vitamin B12 by light (△).
374 375 376 377 378
381 15
382 383
Table 1 Strains, plasmids and primer used in this study
Stains and plasmid
Relevant characteristics/genotype
Source or reference
Lactobacillus delbrueckii spp. lactis DSM 20355
Indicator organism
Deutsche Sammlung von Mikroorganismenun d Zellkulturen (DSMZ)
Propionibacterium freudenreichii spp. shermanii DSM 20270
Organism including vitamin B 12
DSMZ
Strains
riboswitches
Lactobacillus reuteri DSM 20016
Vitamin B12 synthesis organism
DSMZ
Acetobacter pasteurianus DSM 3509
Pseudovitamin B12 produce organism
DSMZ
E. coli DH5α
F-,φ80dlacZ ∆M15, ∆( lacZYA -argF )U169, deoR , recA1 , endA1 , hsdR17 (rK -, mK+), phoA , supE44 , λ-, thi -1 , gyrA96 , relA1
Zhejiang Gongshang Uni, China
E. coli BL21 (DE3)
hsdSB(rB- mB -)λ(DE3[lacI lacUV5-T7 gene 1 ind1 sam7 nin5])
Zhejiang Gongshang Uni, China
p519ngfp
nptII promoter in front of gfp; IncQ replicon; Kmr
Zhejiang Gongshang Uni, China
Primer (5’-3’)
Restriction enzyme sites are underlined
Restriction enzyme
Riboswitch-F
GCTCTAGATGTACTAGGGTCAATGTGCTGG
XbaI
Riboswitch-R
GGAATTCCATATGGGACAAGAACCTCAAATCCACG
NdeI
Plasmid
384 385 386 387 16
388 389 390 391 392 393 394 395 396 397 398
Table 2 Vitamin B12 content of Pig liver, vitamin B12 test broth fermented by P. freudenreichii, soymilk fermented by P. freudenterichii and L.reuteri, vitamin B12 test broth fermented byA. pasteurianus, stinky tofu, and photolysis of cyanocobalamin (1000 ng/mL) (n=4)
Sample
HPLC(ng/g)
Pig liver
905.23±40.67
Vitamin B12 test broth fermented by P. freudenreichii
500.57±5.67
Soymilk fermented by P. freudenterichii and L. reuteri
332.98±2.33
Vitamin B12 test broth fermented by A. pasteurianus
NA
Stinky tofu
499.36±12.65
Decomposed cyanocobalamin by light
NA
a
a
MA(ng/g)
1200.54±60.34 514.55±37.01
a
RB(ng/g)
398.27±16.34
b
a
507.11±27.66
b
306.46±14.70
105.08±20.33 a
1022.36±40.37 566.36±20.66
932.89±37.89
519.45±26.04
RB/
HPLC
MA
a
1.03
0.78
a
1.01
0.99
c
0.92
0.77
/
/
1.04
0.51
/
/
NA b
RB/
a
NA
399
1.Values are means±SD
400
2. NA= not detected
401
3. Analytical replicates=2
402
4. MA means microbiological assay; RS means Riboswitch sensor.
403
5. Values with dissimilar superscript letters (a b and c) along each row indicate significant difference (p< 0.05)
404 405 406
17
407 408
18
409 410
19
411 412
20
413
Highlights:
414
1. A riboswitch sensor method for vitamin B12 determination in foods was developed. 2. The riboswitch sensor results were similar with HPLC. 3. The results of riboswitch were 23% lower than the microbiological assay results. 4. This method has no responses to nucleic acid and pseudovitamin B12. 5. The inter-assay coefficients of variation were 7% for the range of 10-1000 ng/mL.
415 416 417 418 419 420 421 422
21
S0308-8146(14)01901-3 http://dx.doi.org/10.1016/j.foodchem.2014.11.163 FOCH 16848
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
1 September 2014 23 November 2014 29 November 2014
Please cite this article as: Zhu, X., Wang, X., Zhang, C., Wang, X., Gu, Q., A Riboswitch Sensor to Determine Vitamin B12 in Fermented Foods, Food Chemistry (2014), doi: http://dx.doi.org/10.1016/j.foodchem.2014.11.163
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
A Riboswitch Sensor to Determine Vitamin B12 in Fermented Foods
2
Xuan Zhu, Xiaofeng Wang, Chen Zhang, Xiaoqi Wang, Qing Gu*
3
Zhejiang Gongshang University, Key Laboratory for Food Microbial Technology of Zhejiang Province,
4
No. 18 Xuezheng Str., Hangzhou, Zhejiang Province, 310018 China
5
*corresponding author ( Phone: +86-571-28008902; Fax: +86-571-28008900; e-mail:
6
[email protected])
7
8
Abstract
9
We describe a sensitive and selective method for determination of vitamin B12
10
content in fermented foods using riboswitch sensor. A riboswitch amplicon from
11
Propionibacterium freudenreichii was cloned in p519NGFP vector in Escherichia coli
12
BL21 (DE3). The expression of green fluorescence protein was revers correlated to
13
the concentrations of adenosylcobalamin. Adenosylcobalamin directly binds to
14
riboswitch region leading to conformational changes in the secondary structure of
15
mRNA, thus inhibiting expression. After various examinations, a standard curve was
16
obtained from 10 to 1000 ng/mL of cyanocobalamin. The limit of determination is 10
17
ng/mL. The inter-assay coefficients of variation were 7.5% for the range of 10-1000
18
ng/mL. The recovery of this method was 92.3%. This method has no or less responses
19
to nucleic acid, pseudovitamin B12, vitamin B12 bound to intrinsic factor and
20
haptocorrin. The riboswitch sensor results were similar with HPLC, but they were Ca.
21
24% lower than the microbiological assay results.
22
Keywords: riboswitch; green fluorescence protein; cobalamin
23 1
24
25
1. Introduction
26
Vitamin B12, one of water-soluble vitamins, is the general name for natural occurring
27
cobalt organometallic compounds containing substances and is involved in wind
28
range of biochemical processes such as DNA synthesis and regulation, fatty acid
29
synthesis, amino acid metabolism as well as energy production. Vitamin B12
30
deficiency causesmitotic disorder, neuropathy, nervous system disease, and pernicious
31
anaemia(Allen, 2010). To prevent such a fatal deficiency disease, daily intake of 2.4
32
µg vitamin B12 is advised (Rucker, Suttie, McCormick, & Machilin, 2001).
33
Vitamin B12 is exclusively synthesized by some bacteria and archaea and is
34
accumulated in animal bodies by rumen bacteria (Martens, Barg, Warren, & Jahn,
35
2002, Perlman, 1959). Therefore, animal based foods and some fermented plant based
36
foods are considered as the main dietary sources of vitamin B12 for human. (Keuth &
37
Bisping, 1994). Most of the data of Vitamin B12content in foods were acquired by a
38
microbiological assay (MA). The AOAC reference analytical method excels in its
39
sensitivity,
40
consuming.Moreover, results obtained by MA overestimate Vitamin B12 content in
41
foods due to deoxyribonucleutide and other corrinoids (Denter & Bisping, 1994). In
42
addition to MA, HPLC based determination methods, typicallyReversed-phase HPLC
43
using UV and fluorescence, are also developed for analyzing vitamin B12 in fortified
44
foods, vitamin supplement tablets, and infant formula.(Campos-Gimenez, Fontannaz,
45
Trisconi, Kilinc, Gimenez, Andrieux, et al., 2012; Kirchner, Degenhardt, Raffler, &
46
Nelson, 2012; Schimpf, Spiegel, Thompson, & Dowell, 2012; Vyas, O'Kane, &
but
suffers
from
poor
selectivity,
labor
intensity
and
time
2
47
Dowell, 2012). However, most of these methods are only applied for vitamin tablets
48
or milk while none was reported to be used in analyzing low concentrations of
49
vitamin B12 in fermentation foods. Lou and colleaguesreported successful detecting
50
of vitamin B12 in food products by HPLC-ESI-MS.(Luo, Chen, Ding, Tang, & Yao,
51
2006) Furthermore, an UPLC method was successfully applied to detecting vitamin
52
B12 in fermented food withhigh sensitivity (15 ng/mL) and selectivityof active
53
vitamin B12 againstpseudovitamin B12 (Chamlagain, Edelmann, Kariluoto,
54
Ollilainen, & Piironen, 2014). However,
55
complicated sample preparation.
56
Riboswtich is a 5’-untranslated leader sequence of the correspondent mRNA, which
57
regulates translation initiation and gene expression by binding to a specific
58
molecule(Winkler & Breaker, 2005). Expression of cobalamin biosynthetic cob
59
operon and transporter btuB gene was repressed by presence of vitamin B12,
60
particular Adenosyl-cobalamin (ADCBL) (Fowler, Brown, & Li, 2008). It has been
61
reported that ADCBL directly binds to riboswitch region leading to a conformational
62
change in the secondary structure of mRNA which masks the ribosome-binding site
63
(RBS), and thus inhibiting gene expression (Nahvi, Sudarsan, Ebert, Zou, Brown, &
64
Breaker, 2002). Vitreschak et al have summarized almost 200 B12 elements from 66
65
bacterial genomes by computer multiple alignment (Vitreschak, Rodionov, Mironov,
66
& Gelfand, 2003a). A fragment in front of cbiB in Propionibactiumshermaniiwas
67
predicted
68
frombtuBriboswitch
69
btuBriboswitchbased onE. coliriboswitchesin detecting low concentrations vitamin
70
B12 in vitro (Fowler, Brown, & Li, 2008). But the expression of green fluorescence
as
a
chromatography methods require
cobalaminriboswitch,
which
in
al
E.
coli.
Li
et
have
has
different
reported
an
sequences engineered
3
71
protein (GFP) was also inhibited by other corrinoids.
72
In this study we aim to develop a sensitive riboswitch based method (RB) for the
73
analysis of vitamin B12 in various foods. The new method requires simple sample
74
preparation and takes 5 hours. A 2 nd polynomial curve range of the assay is 10 ng/mL
75
to 1000 ng/nL. Compared with microbiological assay (MA), results from this method
76
were minimally influenced by deoxyribonucleutide and corrinoids.
77
2. Materials and methods
78
2.1 Bacteria culture
79
E. coli DH5α and E. coli BL21 (DE3) were grown in Luria-Bertnai (LB) medium at
80
37 °C and 200 rpm. Transformants of E. coli were cultivated in LB medium
81
supplemented with Kanamycin (10 mg/mL) (Sigma-Aldrich, USA). Lactobacillus
82
delbrueckii spp. lactisDSM 20355 was used as the indicator strain in the
83
microbiological assay of vitamin B12.
84
Propionibacteriumfreudenreichiispp.
85
reuteriDSM 20016 and Acetobaterpasteurianus DSM 3509 were cultured in de Man,
86
Rogosa, and Sharpe (MRS) broth (Luqiao, China) at 37 °C. Plasmid p519ngfp was
87
used for the construction of recombinant plasmid carrying cobalamin riboswitch gene.
88
All the bacteria strains and their source were listed in Table 1.
89
2.2. Riboswtich sensor construction and validation
90
The whole chromosome was extracted and purified with QIAamp DNA Mini Kit
91
(QIAGEN, Hilden, Germany) according to the instruction by the manufacturer.
92
Isolated DNA was used as template for PCR. The 213bp fragment containing a
shermanii
DSM
20270,
Lactobacillus
4
93
riboswitch and RBS from P. freudenreichii spp. shermanii DSM 20270 was amplified
94
by the primersdesigned based on the sequence in Gene Bank database (NC_014215.1
95
from 1368045 to 1368257) (Table 1). PCR condition was set as follows: 5 min at
96
94 °C, followed by 30 cycles of 30s at 94 °C, 30s at 50 °C, and 30sat 72 °C. Final
97
extension is 10 min at 72 °C. Fragment was subcloned into p519ngfp vector. The
98
resulting plasmid was amplified in E. Coli DH5α.E. coli BL21 (DE3) was used for
99
expression. All procedures were used as described previously (Sambrook J., Fritsh E.
100
F., & Maniatis T., 2001).
101
E. coli containing p519-switch-ngfp plasmid was incremented in LB media at 37 °C
102
overnight and centrifuged at 4,000 g for 5 min (Biofugepico, Heraeus Instruments,
103
Hanau, Germany), followed by wash in 0.9% sodium chloride for three times. 10 7
104
bacteria were inoculated into 10 mL of vitamin B12 test broth (Merck, Germany)
105
supplemented with 0.01, 1, and 10 mg/L of adenosylcobalamin at 37 °C for 5 hours.
106
Expression of GFP was examined by fluorescence microscope (Leica, Germany) at
107
1000x.
108
2.3. Sample preparation and vitamin B12 determination
109
Vitamin B12 was extracted from 10 g samples in 100 mL sodium acetate buffer (pH
110
6.0) with presence of KCN and heated in a water bath for 30 min at 70 °C. CNCBL
111
was used as standard substance since it is the most stable form of cobalamins.
112
10 6cfu/mL E. coli containing p519-switch-ngfp were inoculated in 9 mL of vitamin
113
B12 test broth supplemented with 1 mL of 0.01, 0.25, 0.5, 0.75, 1, and 1.25 mg/L of
114
cyanocobalamin (CNCBL) at 37 °C for 5 hours. GFP intensity was measured by
115
Cary Eclipse Fluorescence Spectrophotometer (G9800A, Agilent Technologies, US) 5
116
with excitation emission set at 420 nm/470 nm. The values were normalized by
117
absorbance at 600nm (OD600) for all samples. The standard curve was decided by
118
plotting concentrations of CNCBL on the abscissas and fluorescence intensity on
119
ordinates. The response was calculated as a 2 nd order polynomial regression. The
120
residue fluorescence intensity was expressed as a percentage activity of that with 0.01
121
mg/L cyanocobalamin. Hydroxycobalamin (HOCBL), methylcobalamin (MCBL),
122
adenosylcobalamin (ADCBL), intrinsic factor protein binding CNCBL (IF-CNCBL),
123
haptocorrin binding CNCBL (HC-CMCBL), deoxyribonucleotide, pseudovitamin B12,
124
and decomposed vitamin B12 by light at different concentrations were utilized to
125
evaluate influence of cobalamin analogues on sensitivity and selectivity of this
126
riboswitch sensor.
127
3. Results
128
3.1. Riboswtich sensor construction and examination
129
To construct a cobalaminribosensor switch, a 213bp fragment containing a
130
predictedriboswitch and RBS from P. freudenreichiispp. shermanii DSM 20270 was
131
cloned into p519ngfp between pnpt2 promoter and GFP (Fig. 1). The conversed
132
regions of this fragment shared 100% similarity with the sequencereported by
133
vitreschak et al. (Vitreschak, Rodionov, Mironov, & Gelfand, 2003b). For
134
heterologous expression and riboswitch examination, the recombinant plasmids were
135
transferred into E. coli BL21 (DE3) strain. Large-scale cultures were conducted with
136
adenosylcobalamin at 0.01, 1, and 10 mg/L. The fluorescence intensity
137
diminishedwith increasingadenosylcobalamin concentrations (Fig. 2 C D E). In
138
contrast, fluorescence intensity of E. coli with p519ngfp was stable during the 6
139
increase of adenosylcobalamin concentrations (data not shown). These results
140
indicated
141
p519ngfpplasmid and being ableto inhibit GFP expression as we assumed.
142
3.2 Determination of vitamin B12 using the riboswitch sensor
143
With the result that GFP intensity decreases with the presence of vitamin
144
B12,wenexttry to determine the standard curve of this assay. We analyzed the suitable
145
cell density for the determination of vitamin B12. When cells were seeded at a low
146
density (Ca. 103 CFU/mL), the slope was steep and sensitivity of the assay was high.
147
But it took more than 48 hours to perform a test. In contrast, when more than 10 7
148
CFU/mL cells were seeded, the range of the vitamin B12 to be measured is narrow.
149
Considering the importance of the measurement accuracy and variation, 106 CFU/mL
150
cells were used for vitamin B12 measurement of 10-1000 ng/mL.
151
3.3. Measurement of vitamin B12by the riboswitch sensor
152
E. coli cannot produce vitamin B12 de novo(Fowler, Sugiman-Marangos, Junop,
153
Brown, & Li, 2012).Thus it cannot survive in vitamin B12 test broth without extra
154
supplementation. We determined that the bacteria grew with 5 ng/mL of vitamin B12
155
in culture.Therefore, the average and the standard deviation of the GFP intensity were
156
calculated at 5 ng/mL vitamin B12. The limit of detection of vitamin B12 (i.e. 10
157
ng/mL) was considered when the concentration of vitamin B12 was three times
158
greater than the standard deviation of GFP intensity at 5 ng/mL vitamin B12. When
159
the amount of vitamin B12 was greater than 1200 ng/mL, the curve was not fitted for
160
a 2 nd order polynomial curve. Thus, the 2 nd order polynomial curve can be used in the
161
region between 10 and 1000 ng/mL.
that this predictedriboswitch has been successfully cloned into
7
162
The accuracy of measurement was evaluated by calculating inter and intra-assay
163
coefficients of variation of the results. The inter-assay coefficients of variation were
164
7.5% for the range of 10-1000 ng/mL. The intra-assay coefficients of variation were
165
calculated by analyzing pig livers (n=4), stinky tofu (n=4), and other fermented media
166
(n=4). The coefficients of variation were 4.1% for pig livers, 5.0% for stinky tofus,
167
5.4% for Vitamin B12 test broth fermented by P. freudenreichii, and 4.7% for soymilk
168
fermented by P. freudenterichii and L. reuteri respectively. There is no relationship
169
between the magnitude of the coefficient of variation and the nature of the samples.
170
The recovery of this method was 92.3%, which indicated that the extraction and
171
measurement procedures were qualified.
172
The responses of the riboswitch sensor to corrinoids and other complexes were also
173
investigated. As shown in Fig. 3 B, MCBL and OHCBL inhibited GFP expression in a
174
similar manner with a gradual slope. On the other hand, ADCBL inhibited the GFP
175
expression more than CNCBL. Therefore, according to the different responses, all
176
bioactive cobalamin should be extracted by heating in actate buffer with KCN to form
177
CNCBL before assay, when a sample is assayed by the riboswitch assay. In addition,
178
CNCBL is the most thermal stable form.
179
As shown in the panel C of Fig. 3, deoxyribonucleotide and decomposed vitamin B12
180
cannot inhibit GFP expression by shutting down the riboswitch sensor. In contrast,
181
they were detected as vitamin B12 in the microbiological assay. Vitamin B12 in foods
182
and living materials is normally bound to proteins such as intrinsic factor, haptcorrin
183
and other vitamin B12 dependent enzymes. The complex forms of cobalamin were
184
also tested. As shown in Fig. 3 C, only the high concentration of IF-CNCBL (750
185
ng/mL) and HC-CNCBL (750 ng/mL) can turn the riboswitch sensor off. No standard 8
186
curve can be obtained. Thus, releasing vitamin B12 from some proteins before
187
measurement is essential for this riboswitch sensor method.
188
3.3. Measurement of vitamin B12 in samples in comparison with other methods
189
We then compared our assay with other established methods. The results of the P.
190
freudenreichii cells via MA, HPLC, and RB were similar. When we analyze pig livers,
191
result obtained by RB showed 22.3% lower vitamin B12 content than that by the MA,
192
whilesimilar result was obtained by HPLC. In fermented soymilk, the vitamin B12
193
contents determined by the RB were respectively 92.2% of HPLC and 77.0% of MA.
194
The results of vitamin B12 contents in stinky tofu determined by RB were 103.9% of
195
HPLC and 50.8% of MA. For the fermented media by A. pasteurianus and
196
decomposed cyanocobalamin, no detectable level of cyanocobalamin was found by
197
HPLC and RB. However, the results by MA were still high. This phenomenon could
198
be due to the inherent drawbacks associated with the MA.
199
4. Discussion
200
In this study, we describe a sensitive and selective method for determination of
201
vitamin B12 content in fermented foods using riboswitch sensor. Vitreschak et al
202
found 200 of B12 riboswitch elements from 66 bacterial genomes by computer
203
multiple alignment (Vitreschak, Rodionov, Mironov, & Gelfand, 2003a). Moreover, in
204
the website of GenomeNet, a B12 riboswitch element responsible for expression of
205
ABC transporter substrate-binding protein was found in the genome of P. propionicum
206
F0230a (from 16287 to16468 bp). Fowler et al. used a FACS-based method to achieve
207
engineering artificial riboswitches (Campos-Gimenez, et al., 2012). Furthermore,
208
these riboswitch tools were used to explore intermolecular interactions of a vitamin 9
209
B12 binding protein (Fowler, Sugiman-Marangos, Junop, Brown, & Li, 2012). Even a
210
very low ADCBL concentration resulted in strong repression of reporter expression.
211
These artificial elements were so sensitive that it has very narrow detection range,
212
limiting theirapplication in a quantitative method.
213
E. coli is an idea host of B12 riboswitch sensor, as it cannot produce vitamin B12 de
214
novo. In our work, the B12 riboswitch element from P. freudenreichii spp. shermanii
215
DSM 20270 has a definitely different structure with riboswitches in E. coli (Fowler,
216
Brown, & Li, 2008), although the riboswitchis regulated by similar mechanism which
217
covers RBS section with an antisequence (Fig. 2 B). Our riboswitch has a short right
218
arm and “CCCC” sequence head (Fig. 2 A), which is responsible for folding of RNA
219
structure. The reporter gene repression as the result of presence vitamin B12 (Fig. 2)
220
was also related to transport protein efficiency of the ADCBL or precursors into
221
cytoplasm (Kirchner, Degenhardt, Raffler, & Nelson, 2012). The extent of GFP
222
repression was affected by transport proteins, as they have various affinities with
223
different cobamindes (Fig. 3 B). ADCBL precursors or cobamindes are quickly
224
converted to ADCBL by metabolic enzymes following transport (Kirchner,
225
Degenhardt, Raffler, & Nelson, 2012). Thus these substances can also be measured. In
226
addition, the affinity of riboswitch was also a crucial factor for GFP repression. Some
227
of riboswitches recognized MCBL and OHCBL with more than 500-fold higher
228
affinity than ADCBL (Johnson, Reyes, Polaski, & Batey, 2012). This structure
229
containing a short right arm has a high sensitivity with derivatives with small
230
upper-axial moieties (Johnson, Reyes, Polaski, & Batey, 2012). Unlike the E. coli
231
btuB riboswitch that selectively binds adenosylcobalamin (Kirchner, Degenhardt,
232
Raffler, & Nelson, 2012), this riboswitch responsesto various bioactive cobalamin 10
233
such as CNCBL, OHCBL, and MCBL besides ADCBL (Fig. 3B). However, this
234
riboswitch sensor has few responses on pseudovitamin B12 and light-decomposed
235
vitamin B12, as their upper-moieties cannot bind this section.
236
The GFP expression was partly inhibited by vitamin B12 bound to IF. In contrast, HC
237
almost blocked the determination completely (Fig. 3C). Some researchers
238
demonstrated that vitamin B12 binds with IF via two sites (Schneider & Stroinski,
239
1987). One is dimethylbenzimidazole, which was bound with cobalt as lower-axis.
240
The other is a part of the A and B ring of porphyrin ring. As a result, the upper ligand
241
of the corrin ring was not affected by IF. In contrast, HC binds to cobalamin via only
242
one site in porphyrin ring. Although the upper and lower ligands were not covered by
243
HC, a steric hindrance occurs between a large group and the HC molecule. A
244
distortion of bond may be the reason for the less response from HCCBL in RB
245
method.
246
The MA is recognized as an official standard method for determination of low levels
247
of vitamin B12 in foods and biological material. However, it was consistently reported
248
to yield higher vitamin B12 results than HPLC and other methods. Thisdeviation
249
appeared due to their response to corrinoids which are inactive for human such as
250
pseudovitamin B12 and nucleic acid (Santos, Vera, Lamosa, de Valdez, de Vos, Santos,
251
et al., 2007; Wongyai, 2000). In our previous work, we found that A. pasteuriamus
252
could produce corrinoids by which L. delbrueckii could survive in vitamin B12 test
253
broth. But no detected bioactive vitamin B12 could be found via HPLC method. As
254
shown in table 2, Our RB method had no response to pseudovitamin B12 from A.
255
pasteruiamus. Moreover, RB method has shown no response to nucleic acid in Fig.
256
3C. L. derlbrueckii can utilize the nucleic acid pool to survive in the media without 11
257
vitamin B12, as vitamin B12 wasinvolved in the nucleic acid synthesisas coenzyme
258
(Schneider & Stroinski, 1987). Some scientists reported a 20% higher vitamin B12
259
content in foods by MA compared to HPLC (Heudi, Kilinc, Fontannaz, & Marley,
260
2006). Poor accuracy of the MA was further shown in analysis of shellfish, meat and
261
milk, where the results from MA were up to 4-5 times higher than HPLC and other
262
methods (Watanabe, Yabuta, Tanioka, & Bito, 2014).
263
In present work, results from the RB method were similar with that of HPLC (table 2),
264
of which no analogues of cobalamin could be detected. However, for the samples of
265
stinky tofu, the results of RB were still 13% higher than HPLC. HPLC method,
266
particularly UPLC method, is capable of sensitive quantification of the vitamin B12
267
content in microbial cells, and fermented matrices.(Chamlagain, Edelmann, Kariluoto,
268
Ollilainen, & Piironen, 2014). However, this method requires a complicated
269
preparation procedure to collect free vitamin B12 released from protein. Thus, in
270
some cases, vitamin B12 bound to protein cannot be detected. For example, in
271
fermented tofu that contains lots of protein, the recovery of vitamin B12 was less than
272
75% (Zhu & Bisping, 2012). RB method was less influencedby bound protein.
273
5. Conclusion
274
In this study, riboswitch sensor was adopted as a sensitive and specific method of
275
vitamin B12 quantification in fermented foods with short incubation. The method
276
allowed for selective determination of bioactive vitamin B12, thus avoiding the
277
influence from nucleic acid and other inactive corrinoids as in the MA. This method
278
also allowed for eliminating the influence of proteins as in the HPLC method.
279 12
280
6. Acknowledgments
281
This project was supported by the National High Technology Research and
282
Development Program ("863" Program) of China (2014AA022210), Scientific
283
Research Foundation of Zhejiang Gongshang University, and Scientific Research
284
Foundation of Education Department of Zhejiang Province (1110kz0414207).
285
7. References
286
Allen, L. H. (2010). Bioavailability of vitamin B12. Int J Vitam Nutr Res, 80(4-5), 330-335.
287 288
Barry, T. N., Hoskin, S. O., & Wilson, P. R. (2002). Novel forages for growth and health in farmed deer. N Z Vet J, 50(6), 244-251.
289 290 291 292
Campos-Gimenez, E., Fontannaz, P., Trisconi, M. J., Kilinc, T., Gimenez, C., Andrieux, P., & Nelson, M. (2012). Determination of vitamin B12 in infant formula and adult nutritionals by liquid chromatography/UV detection with immunoaffinity extraction: First Action 2011.08. J AOAC Int, 95(2), 307-312.
293 294 295
Chamlagain, B., Edelmann, M., Kariluoto, S., Ollilainen, V., & Piironen, V. (2014). Ultra-high performance liquid chromatographic and mass spectrometric analysis of active vitamin B12 in cells of Propionibacterium and fermented cereal matrices. Food Chem, 166, 630-638.
296 297
Denter, J., & Bisping, B. (1994). Formation of B-vitamins by bacteria during the soaking process of soybeans for tempe fermentation. Int J Food Microbiol, 22(1), 23-31.
298 299
Fowler, C. C., Brown, E. D., & Li, Y. (2008). A FACS-based approach to engineering artificial riboswitches. Chembiochem, 9(12), 1906-1911.
300 301 302
Fowler, C. C., Sugiman-Marangos, S., Junop, M. S., Brown, E. D., & Li, Y. (2012). Exploring intermolecular interactions of a substrate binding protein using a riboswitch-based sensor. Chem Biol, 20(12), 1502-1512.
303 304 305
Heudi, O., Kilinc, T., Fontannaz, P., & Marley, E. (2006). Determination of Vitamin B12 in food products and in premixes by reversed-phase high performance liquid chromatography and immunoaffinity extraction. J Chromatogr A, 1101(1-2), 63-68.
306 307
Johnson, J. E., Jr., Reyes, F. E., Polaski, J. T., & Batey, R. T. (2012). B12 cofactors directly stabilize an mRNA regulatory switch. Nature, 492(7427), 133-137.
308 309 310
Keuth, S., & Bisping, B. (1994). Vitamin B12 production by Citrobacter freundii or Klebsiella pneumoniae during tempeh fermentation and proof of enterotoxin absence by PCR. App Environ Microb, 60(5), 1495-1499.
311 312
Kirchner, U., Degenhardt, K., Raffler, G., & Nelson, M. (2012). Determination of vitamin B12 in infant formula and adult nutritionals using HPLC after purification on an immunoaffinity column: 13
313
first action 2011.09. J AOAC Int, 95(4), 933-936.
314 315
Luo, X., Chen, B., Ding, L., Tang, F., & Yao, S. Z. (2006). HPLC-ESI-MS analysis of vitamin B12 in food products and in multivitamins-multimineral tablets. Analytica Chimica Acta, 562(2), 185-189.
316 317
Martens, J. H., Barg, H., Warren, M. J., & Jahn, D. (2002). Microbial production of vitamin B12. Appl Microbiol Biotechnol, 58(3), 275-285.
318 319
Nahvi, A., Sudarsan, N., Ebert, M. S., Zou, X., Brown, K. L., & Breaker, R. R. (2002). Genetic control by a metabolite binding mRNA. Chem Biol, 9(9), 1043.
320
Perlman, D. (1959). Microbial synthesis of cobamides. Adv Appl Microbiol, 1, 87-122.
321 322
Rucker, B. R., Suttie, J. W., McCormick, B. D., & Machilin, L. J. (2001). Handbook of vitamin. New York: Marcel Dekker, Inc.
323 324
Sambrook J., Fritsh E. F., & Maniatis T. (2001). Molecular cloning: a laboratory manual. Cold spring harbor laboratory In). New York: Cold Spring Harbor.
325 326 327
Santos, F., Vera, J. L., Lamosa, P., de Valdez, G. F., de Vos, W. M., Santos, H., Sesma, F., & Hugenholtz, J. (2007). Pseudovitamin is the corrinoid produced by Lactobacillus reuteri CRL1098 under anaerobic conditions. FEBS Letters, 581(25), 4865-4870.
328 329
Schimpf, K., Spiegel, R., Thompson, L., & Dowell, D. (2012). Determination of vitamin B12 in infant formula and adult nutritionals by HPLC: First Action 2011.10. J AOAC Int, 95(2), 313-318.
330 331
Schneider, Z., & Stroinski, A. (1987). Comprehensive B12: chemistry, biochemistry, nutrition, ecology, medicine. Berlin: Verlag Walter de Gruyter.
332 333 334
Vitreschak, A. G., Rodionov, D. A., Mironov, A. A., & Gelfand, M. S. (2003a). Regulation of the vitamin B12 metabolism and transport in bacteria by a conserved RNA structural element. Rna, 9(9), 1084-1097.
335 336 337
Vitreschak, A. G., Rodionov, D. A., Mironov, A. A., & Gelfand, M. S. (2003b). Regulation of the vitamin B12 metabolism and transport in bacteria by a conserved RNA structural element. Rna, 9(9), 1084-1097.
338 339 340
Vyas, P., O'Kane, A. A., & Dowell, D. (2012). Determination of vitamin B12 in infant formula and adult nutritionals by surface plasmon resonance: First Action 2011.16 (test kit method). J AOAC Int, 95(2), 329-334.
341
Watanabe, F. (2007). Vitamin B12 sources and bioavailability. Exp Biol Med, 232(10), 1266-1274.
342 343 344
Watanabe, F., Yabuta, Y., Tanioka, Y., & Bito, T. (2014). Biologically active vitamin B12 compounds in foods for preventing deficiency among vegetarians and elderly subjects. J Agric Food Chem, 61(28), 6769-6775.
345 346
Winkler, W. C., & Breaker, R. R. (2005). Regulation of bacterial gene expression by riboswitches. Annu Rev Microbiol, 59, 487-517.
347 348
Wongyai, S. (2000). Determination of vitamin B12 in multivitamin tablets by multimode high-performance liquid chromatography. J Chromatogr A, 870(1-2), 217-220. 14
349 350
Zhu, X., & Bisping, B. (2012). Determination of Vitamin B12 in Fermented Soybean Products by High-Performance Liquid Chromatography. J biotech, 150, 330.
351 352 353 354 355 356 357 358 359
Figure Caption
360 361 362 363 364
Fig. 1 Genetic maps of p519ngfp and p519-Switch-ngfp showing the insert of a riboswitch. The insert riboswitch was obtained by PCR amplication of the leader mRNA of cbib in P. freudenreichii. The restricted enzyme sites used in cloning experiments are shown.
365 366 367 368 369 370 371 372
Fig. 2 Mechanism of vitamin B12 dependantriboswitch (A and B) and expressions of GFP in E. coli BL21 with recombinant plasmid p519-Switch-ngfp under various supplementation of adenosylcobalamin (0.01, 1, and 10 mg/L) (C, D, and E). A and B: AGGAG is a RBS (ribosome binding site). The antisequestor region with P3 (CCCC) aptamer domain forms a pseudoknot under a high concentration of adenosylcobalamin. The complement antisequestor region with RBS forms a hairloop to attenuate translation.
373
379
Fig. 3 A: a standard curve for determination of vitamin B12 by the riboswitch method. 1 mL of vitamin B12 test broth with E. coli BL21 (10 6 cfu/mL) supplemented respectively with 0, 0.25, 0.5, 0.75, 1, and 1.25 µg/mL CNCBL were incubated at 37 ℃ for 5 hours. B: Expression inhibition of GFP by CNCBL (□), ADCBL (▲), MCBL (●), HOCBL (△), and Nothing (◆). C: Expression inhibition of GFP by CNCBL (□), IF-CNCBL (■), HC-CNCBL (▲), deoxyribonucleotide (◇), pseudovitamin B12 (◆),
380
and decomposed vitamin B12 by light (△).
374 375 376 377 378
381 15
382 383
Table 1 Strains, plasmids and primer used in this study
Stains and plasmid
Relevant characteristics/genotype
Source or reference
Lactobacillus delbrueckii spp. lactis DSM 20355
Indicator organism
Deutsche Sammlung von Mikroorganismenun d Zellkulturen (DSMZ)
Propionibacterium freudenreichii spp. shermanii DSM 20270
Organism including vitamin B 12
DSMZ
Strains
riboswitches
Lactobacillus reuteri DSM 20016
Vitamin B12 synthesis organism
DSMZ
Acetobacter pasteurianus DSM 3509
Pseudovitamin B12 produce organism
DSMZ
E. coli DH5α
F-,φ80dlacZ ∆M15, ∆( lacZYA -argF )U169, deoR , recA1 , endA1 , hsdR17 (rK -, mK+), phoA , supE44 , λ-, thi -1 , gyrA96 , relA1
Zhejiang Gongshang Uni, China
E. coli BL21 (DE3)
hsdSB(rB- mB -)λ(DE3[lacI lacUV5-T7 gene 1 ind1 sam7 nin5])
Zhejiang Gongshang Uni, China
p519ngfp
nptII promoter in front of gfp; IncQ replicon; Kmr
Zhejiang Gongshang Uni, China
Primer (5’-3’)
Restriction enzyme sites are underlined
Restriction enzyme
Riboswitch-F
GCTCTAGATGTACTAGGGTCAATGTGCTGG
XbaI
Riboswitch-R
GGAATTCCATATGGGACAAGAACCTCAAATCCACG
NdeI
Plasmid
384 385 386 387 16
388 389 390 391 392 393 394 395 396 397 398
Table 2 Vitamin B12 content of Pig liver, vitamin B12 test broth fermented by P. freudenreichii, soymilk fermented by P. freudenterichii and L.reuteri, vitamin B12 test broth fermented byA. pasteurianus, stinky tofu, and photolysis of cyanocobalamin (1000 ng/mL) (n=4)
Sample
HPLC(ng/g)
Pig liver
905.23±40.67
Vitamin B12 test broth fermented by P. freudenreichii
500.57±5.67
Soymilk fermented by P. freudenterichii and L. reuteri
332.98±2.33
Vitamin B12 test broth fermented by A. pasteurianus
NA
Stinky tofu
499.36±12.65
Decomposed cyanocobalamin by light
NA
a
a
MA(ng/g)
1200.54±60.34 514.55±37.01
a
RB(ng/g)
398.27±16.34
b
a
507.11±27.66
b
306.46±14.70
105.08±20.33 a
1022.36±40.37 566.36±20.66
932.89±37.89
519.45±26.04
RB/
HPLC
MA
a
1.03
0.78
a
1.01
0.99
c
0.92
0.77
/
/
1.04
0.51
/
/
NA b
RB/
a
NA
399
1.Values are means±SD
400
2. NA= not detected
401
3. Analytical replicates=2
402
4. MA means microbiological assay; RS means Riboswitch sensor.
403
5. Values with dissimilar superscript letters (a b and c) along each row indicate significant difference (p< 0.05)
404 405 406
17
407 408
18
409 410
19
411 412
20
413
Highlights:
414
1. A riboswitch sensor method for vitamin B12 determination in foods was developed. 2. The riboswitch sensor results were similar with HPLC. 3. The results of riboswitch were 23% lower than the microbiological assay results. 4. This method has no responses to nucleic acid and pseudovitamin B12. 5. The inter-assay coefficients of variation were 7% for the range of 10-1000 ng/mL.
415 416 417 418 419 420 421 422
21