CORRESPONDENCE
npg
© 2015 Nature America, Inc. All rights reserved.
Chinese government reaffirms backing for GM products To the Editor: This year’s ‘No. 1 Central Document’ (NCD)—a document issued every January by the Chinese Communist Party’s Central Committee and the State Council—states explicitly that China will put more resources into R&D on genetically modified (GM) products and into public education surrounding these products so as to regain the country’s position as a global leader in the area. Genetic modification has been discussed six times in the annual NCD between 2007 and 2015. Here we survey NCDs from 2007 to 2015 and infer that the top leadership’s opinion shifted like a pendulum toward GM technology, and then away in the face of public opposition—and now back again. In 2007, the NCD announced an intention to implement a GM food labeling system. It was followed by the “Mandatory Labeling of GM Food” policy released by China’s Ministry of Agriculture in 2007 (ref. 1). In 2008, the NCD started national research projects to cultivate new varieties of transgenic organisms. This was also followed by a National Science and Technology Major Project (NSTMP) program (released by China’s Ministry of Science and Technology in 2008) to invest about 24 billion yuan (~$3.9 billion) in developing GM technologies from 2008 to 2020. In 2009 and 2010, the two NCDs proposed to accelerate the transgenic NSTMP program2. And in 2009, the Biosafety Committee of China’s Ministry of Agriculture issued biosafety licenses for two crops, rice and corn3. This policy evolution—from asserting the need for labeling of GM products to concerted investment in the GM NSTMP program and regulatory approvals raised expectations that China might become a leader in commercializing GM rice and corn1,2. However, in the years following 2009, public resistance to GM products grew. And in China, public opinion often has a powerful influence in steering government policy in areas that are less politically controversial. For example, in the late 2000s, public environmental protests were instrumental in persuading local government officials to cancel construction of a billion-dollar chemical plant and a trash incinerator project4. Such environmental activism has also induced officials to disclose more environmental information5. As in the West, public
concerns about GM food safety are widespread. According to the Blue Book of Public Opinion Survey in China, only 0.9% of the respondents fully accept GM food6. In this context, the political neutrality of GM policy has meant that public safety concerns have trumped GM policy to the extent that a protechnology stance in the early 2000s was almost entirely reversed by 2010. Thus, from 2011 to 2014, the NCD failed to mention GM products at all. And the NCD in 2014 even neglected to use the term “GM,” preferring instead the euphemism “molecular breeding.” We contend that this political stalemate regarding GM products—reflected in the NCDs—has set back transgenic R&D in China and turned the country from being one of the world’s proponents of the technology to one of the most conservative developers. Between 1997 and 2001, China encouraged the planting of insect-resistant GM cotton and was the fourth largest GM crop grower globally. In 2014, China, with 3.9 million hectares of GM crops, ranked sixth in the world behind India (with 11.6 million hectares of GM crops)7. And in 2013, research funding for GM seed cultivation programs fell to around 400 million yuan (~$65 million), down from as much as 2 billion yuan in 2010 (ref. 8). In this context, we believe that the 2015 NCD represents an important
shift in government policy. It indicates a restatement of interest by the Central Committee and State Council in GM technology, which augurs well for R&D projects on GM products. It remains to be seen how fast this renewed political backing can be translated into a return of national funding into this important research area. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests.
Rongrong Li1, Qiang Wang2,3 & Alan McHughen4 1School of Business Administration, Xinjiang
University of Finance and Economics, Urumqi, Xinjiang, P.R. China. 2Xinjiang Institute of Ecology and Geography, CAS, Urumqi, P.R. China. 3China University of Petroleum (East China), Qingdao, P.R. China. 4Department of Botany & Plant Sciences, University of California-Riverside, California, USA. e-mail: [email protected] Qiu, J. Natl. Sci. Rev. 1, 466–470 (2014). Wang, Q. Nature 519, 7 (2015). Qiu, J. Nature 455, 850–852 (2008). Wang, Q. Science 328, 824 (2010). Wang, Q. Nature 497, 159 (2013). Xie, G.-Y. Blue Book of Public Opinion Surveys in China (Social Science Academic Press, Shanghai, China, 2014). 7. Clive, J. ISAAA brief No. 49 (2014). 8. Shuping, N. & Stanway, D. Reuters http://www.reuters. com/article/2014/03/10/china-parliament-g moidUSL3N0M41FR20140310 (10 March 2015). 1. 2. 3. 4. 5. 6.
Increased bioavailable vitamin B6 in field-grown transgenic cassava for dietary sufficiency To the Editor: Vitamin B6 deficiency is associated with cardiovascular disease, diabetes, neurological diseases1 and nodding syndrome (NS, a childhood condition in eastern Africa2). Pyridoxal 5’-phosphate (PLP), the cofactor form of vitamin B6, is required for multiple essential enzymatic reactions and other forms of vitamin B6, such as pyridoxine and pyridoxamine, inhibit oxygen radical production, pathogenic glycation and lipid peroxidation3,4. Notably, investigation of the causes of NS has revealed a high prevalence
NATURE BIOTECHNOLOGY VOLUME 33 NUMBER 10 OCTOBER 2015
of vitamin B6 deficiencies in affected African populations2,5,6. Humans cannot synthesize vitamin B6 de novo, so this micronutrient is obtained from food by intestinal absorption7. In this Correspondence, we report biofortification of cassava Manihot esculenta Crantz with vitamin B6 using transgenes from Arabidopsis thaliana. Cassava is a staple root crop for more than 250 million people in Africa8, including regions in which vitamin B6 deficiency is endemic6. In addition to the plant’s starchy roots, cassava leaves are also consumed as a 1029
Biosynthesis de novo
Salvage pathway and glycosylation
–
–
se T-a
NH2 O NH2
P-ase
Gln
PDX2 NH3
PLR PN
SOS4
PL
PD
S4
SO
X3
–
se
a
P-
PLP O
PM
–
PDX1
OH
O
SOS4
PMP
3
X PD
R5P
O
P-ase
L
–
NH2
G3P
Glu
PNG
PNP
WT 35S-5 Standards
800 600
PNG
400 200 PMP PM PNP
0
2
c
PL
PN
PLP
4 6 8 Retention time (min)
10
Phosphorylated B6 vitamers (PLP, PNP, PMP)
12
400 300
PNG
100 0
PMP PM PNP
0
2
PL
PN
PLP
4 6 8 Retention time (min)
Unphosphorylated B6 vitamers (PL, PN, PM)
80
**
50
**
40
*
30
**
20 10
*
0
Vitamin B6 content in roots (μg/g FW)
60
**
15
10
**
5 ** ** *
**
*
PA T-2
5 S35
S3 35
PA T-1 2
PA T-2
5 S35
S3 35
12
*
0
WT
10
PNG
20
70
WT PAT-12 Standards
200
WT
0
500
PA T-1 2
1,000
Fluorescence intensity
Fluorescence intensity
b
Vitamin B6 content in leaves (μg/g FW)
npg
a
G
1030
variable for nine field-grown, six-month-old plants of the transgenic lines, when compared with the wild-type control (Supplementary Table 3). Leaves and roots were collected from three independent plants in the first field experiment to test the stability of the improved vitamin B6 trait in the transgenic lines using three biological replicates. Transgenic 35S and PAT lines had greater total vitamin B6 content compared with the wild-type control (3.9- to 48.2-fold increase in transgenic leaves and a 1.9- to 5.8-fold
To evaluate the stability of the engineered vitamin B6 trait, we selected nine transgenic 35S and PAT lines with increased total vitamin B6 content (Supplementary Fig. 3) and evaluated them in the field. Wild-type and transgenic cassava lines propagated in vitro were first hardened in the greenhouse and then transferred to an experimental field in Shanghai for multiplication. Stem cuttings from the multiplied plants were then evaluated further in two field experiments. Plant height and storage root yield were
T
vegetable. Based on micronutrient data from raw cassava root, it has been suggested that a 2.3-fold increase in vitamin B6 content would be sufficient to meet the recommended dietary allowance for populations relying heavily on cassava in their diet9. Biosynthesis of vitamin B6 de novo in planta requires two enzymes, a synthase, PDX1, and a glutaminase, PDX2 (ref. 10 and Fig. 1a). Salvage and glycosylation pathways interconvert the B6 vitamers and their corresponding derivatives11–13 (Fig. 1a). We constructed cassettes for ubiquitous (CaMV35S promoter) and root-enhanced (Patatin promoter) heterologous expression of the A. thaliana PDX1.1 and PDX2 genes (AtPDX1.1 and AtPDX2, respectively; Supplementary Fig. 1). We hypothesized that this strategy would reduce the risk of gene silencing associated with overexpression of endogenous genes14. A subset of independent, single-insertion transgenic cassava lines generated with each of these constructs for either ubiquitous (five lines) or root-enhanced (four lines) expression (hereafter named 35S and PAT lines, respectively; Supplementary Fig. 2), were selected and transferred to the greenhouse. Plant height, as well as above- and belowground fresh weight, revealed no significant differences between wild-type, transgenic control–harboring empty pCAMBIA vector, and the transgenic 35S and PAT lines, except for lines 35S-1 and PAT-7, which had a significant but not unusual increase in plant height that is sometimes observed for cassava plants grown from tissue culture (Supplementary Table 1). Vitamin B6 measurements made using a microbial assay (Supplementary Note 1) showed that total vitamin B6 content was several-fold higher in the leaves and roots of four transgenic 35S and four PAT cassava lines (Supplementary Fig. 3) and correlated with the expression levels of AtPDX1.1 and AtPDX2 as measured by qPCR (Supplementary Fig. 4, Supplementary Table 2). We identified three orthologs of AtPDX1 and one ortholog of AtPDX2 in the cassava genome (Supplementary Fig. 5). Expression profiles of these orthologs were similar in wildtype and transgenic cassava lines with the exception of the M. esculenta PDX1.3a gene, which had a slightly reduced expression level in 35S-2 leaves (Supplementary Fig. 6). Together, these results indicate that expression of the AtPDX1.1 and AtPDX2 transgenes had no significant effect on the expression of the endogenous cassava genes but increased total vitamin B6 content in leaves and roots.
G
© 2015 Nature America, Inc. All rights reserved.
CORRESPONDENCE
Figure 1 Genetic engineering of vitamin B6. (a) Biosynthesis de novo, salvage pathways and glycosylation of vitamin B6. Glu, glutamate; Gln, glutamine; G3P, glyceraldehyde 3-phosphate; R5P, ribose 5-phosphate; PN, pyridoxine; PNG, pyridoxine 5’-b-d-glucoside; PL, pyridoxal; PM, pyridoxamine; PNP, pyridoxine 5’-phosphate; PLP, pyridoxal 5’-phosphate; PMP, pyridoxamine 5’-phosphate; T-ase, transaminase; P-ase, phosphatase; PLR, PL reductase; GT, glycosyl transferase; GL, glucosidase. (b) HPLC chromatogram of field-grown wild-type cassava and transgenic cassava expressing AtPDX1.1 and AtPDX2 under either the 35S promoter (transgenic line 35S-5) or Patatin promoter (transgenic line PAT-12). (c) Analysis of unphosphorylated, phosphorylated B6 vitamers and PNG in leaves and roots of wild-type and selected transgenic cassava lines grown under field conditions. Mean ± s.d. of three biological replicates. Student’s t-test (wild-type vs. transgenic line), *P < 0.05, **P < 0.01. Error bars, mean ± s.d. of three biological replicates.
VOLUME 33 NUMBER 10 OCTOBER 2015 NATURE BIOTECHNOLOGY
CORRESPONDENCE Table 1 Caco-2 cell assays with extracts from processed cassava leaves and roots a Compound
Pyridoxine (0 nM)
Pyridoxine (100 nM)
Pyridoxine (1,000 nM)
Wild-type leaves (boiled)
35S-5 leaves (boiled)
Wild-type roots (boiled)
PAT-12 roots (boiled)
Apical compartment at 0 h (nM) Pyridoxal
NA
NA
NA
115.1
77.1
178.5
25.1
Pyridoxine
NA
NA
NA
58.1
3,042.5
0
33.2
Pyridoxine
0
100
1,000
142.6
1,295.1
264.8
146.3
PMP
NA
NA
NA
73.6
108.4
50.5
47.8
PNP
NA
NA
NA
ND
ND
ND
ND
PNG
NA
NA
NA
398.3
4,634.5
243.5
703.1
60.5 (±9.3)
70.4 (±4.4)
41.9 (±5.5)
66.0 (±8.0)
50.8 (±4.6)
60.9 (±0.6)
Intracellular at 48 h (nM) PLP
43.6 (±10.1)
npg
© 2015 Nature America, Inc. All rights reserved.
Basolateral compartment after 48 h (nM) Pyridoxal
ND
13.1 (±1.9)
90.3 (±12.4)
11.1 (±0.5)
88.3 (±13.6)
13.3 (±0.3)
44.1 (±1.7)
Pyridoxine
ND
ND
12.1 (±2.1)
ND
6.2 (±5.4)
ND
ND
Pyridoxine
ND
ND
3.5 (±0.6)
ND
7.8 (±1.7)
ND
4.8 (±0.2)
PMP
ND
ND
ND
ND
ND
ND
ND
PNP
ND
ND
ND
ND
ND
ND
ND
PNG
ND
ND
1.7 (±2.8)
ND
26.1 (±8.2)
ND
3.7 (±6.3)
aAnalysis
of PLP in Caco-2 cells in the apical compartment and B6 vitamers and PNG in the apical (0 h) and basolateral (48 h) compartments. Equal amounts (10 g) of powder from leaves and roots were extracted in 20 ml of buffer, diluted tenfold and supplemented to the apical compartment. Mean ± s.d. of three biological replicates. NA, not applicable; ND, not detected.
increase in transgenic roots), which was consistent with the amounts previously measured in greenhouse-grown plants (Supplementary Fig. 7). These data confirm that the engineered vitamin B6 trait was stable over two cycles of vegetative propagation. Plants accumulate a substantial fraction of vitamin B6 as pyridoxine-5'-b-d-glucoside (PNG)15. Forms of vitamin B6 accumulating in wild-type control and transgenic cassava were analyzed using an established highperformance liquid chromatography (HPLC) method16. Comparison of chromatograms revealed significant differences between wildtype control and four selected transgenic lines (Fig. 1b). Our analyses revealed that the increased vitamin B6 content of transgenic lines was mostly due to an increase in unphosphorylated forms (4.6- to 14.8-fold increase in leaves and 1.7- to 5.8-fold in roots) and of PNG (6.1- to 9.7-fold increase in leaves and 5.4- to 30.7-fold increase in roots) (Fig. 1b,c and Supplementary Table 4). We observed no increase of phosphorylated forms in leaves and roots of transgenic lines, with the exception of the roots of the PAT-12 line (threefold increase). Cassava leaves and storage roots are usually boiled in water before consumption, a step that removes toxic cyanogens. This process could affect vitamin B6 content due to the instability of this vitamin at high temperature as well as its solubility in water. Vitamin B6 could, therefore, be reduced in a cassava meal. In all cases, boiling decreased the relative vitamin B6 content of leaves (65%, 15% and 34% decrease for tested wild-type line, transgenic line 35S-5 and
transgenic line 35S-3, respectively) and roots (51%, 71% and 75% for tested wild-type line, transgenic line PAT-12 and transgenic line PAT-2, respectively) based on fresh weight (Supplementary Tables 4 and 5). Nonetheless, transgenic leaves and roots retained significantly (P value < 0.01) higher amounts of unphosphorylated B6 forms over wild-type control (ninefold increase in transgenic leaves and fourfold increase in transgenic roots; see Supplementary Table 5) and PNG (40-fold increase in transgenic leaves and 14-fold increase in transgenic roots: see Supplementary Table 5) even after boiling. B6 vitamers and PNG from food sources have different bioavailability in humans15. Inhibition of pyridoxine uptake by pyridoxamine in human intestinal epithelial Caco-2 (colorectal adenocarcinoma) cells17, as well as the competitive inhibition of pyridoxine uptake by PNG in rat hepatocytes18, suggest that B6 vitamer composition might also influence the bioavailability of total vitamin B6 from food sources. We used a Caco-2 cell assay to test the bioavailability of vitamin B6 from wild-type and transgenic cassava. Leaves and roots were first boiled for 30 min as in routine cassava preparation. Leaf and root extracts were then supplied to the apical compartment of the assay (Supplementary Fig. 8) and Caco-2 cells were incubated for 48 h. According to current models for intestinal metabolism of vitamin B6, PL, PM and PN are readily absorbed, converted into PLP through intracellular salvage pathway enzymes (Fig. 1a) and subsequently
NATURE BIOTECHNOLOGY VOLUME 33 NUMBER 10 OCTOBER 2015
dephosphorylated to be excreted as PL in the portal vein. PNG is converted into PN before uptake, transformed into PLP intracellularly and also excreted into the portal vein as PL7. Therefore, measurement of pyridoxal concentrations in the basolateral compartment (Supplementary Fig. 8) provides a robust estimation of vitamin B6 bioavailability in the intestine. Incubation of Caco-2 cells with extracts from cassava leaves and storage roots resulted in an eightfold and fourfold, respectively, increase of pyridoxal levels (transgenic line versus wild type) (Table 1), showing there were higher levels of bioavailable vitamin B6 in the transgenic cassava. Incubation of Caco-2 cells with extracts from transgenic cassava or an excess amount of pyridoxine (1 mM) also resulted in trace accumulation of pyridoxine and pyridoxamine in the basolateral compartment (Table 1 and Supplementary Fig. 8). When the maximum capacity of the intestine to convert pyridoxine and pyridoxamine into PLP is exceeded, these B6 vitamers are excreted without intracellular conversion7,20. The intracellular PLP levels were consistent with a higher conversion of precursor B6 vitamers into PLP by Caco-2 cells incubated with leaf and root extracts from transgenic cassava (Table 1). Using bioavailable ‘vitamin B6 equivalents’21, we calculated that the vitamin B6 recommended dietary allowance for an adult person (1.3 mg/day) would be reached with 51 g of boiled 35S-5 leaves or 505 g (~1.7 lb) of boiled PAT12 storage roots (Supplementary Note 1). This study shows that cassava can be biofortified using the AtPDX-based 1031
CORRESPONDENCE engineering strategy to increase vitamin B6. The engineered trait was stable under field conditions and the engineered vitamin B6 was bioavailable. Because cassava is a staple crop in those African regions where vitamin B6 deficiency is endemic, AtPDX transgenic vitamin B6-enhanced cassava could be included in a portfolio of dietary and supplementation approaches to help reduce this important nutritional deficiency. Further characterization of the vitamin B6–enhanced transgenic lines, including location-specific agronomic field performance tests, will be the next step before introgressing, or engineering, the increased vitamin B6 trait into cassava cultivars preferred by local farmers and consumers22.
ACKNOWLEDGMENTS We acknowledge support from ETH Zurich, the University of Geneva, the Swiss National Science Foundation (project 31003A_140911) and the Earmarked Fund for the China Agriculture Research System (CARS-12-shzp). We thank S. Klarer and I. Zurkirchen (ETH Zurich) for special care of the
COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests.
Kuan-Te Li1, Michael Moulin2, Nathalie Mangel1, Monique Albersen3, Nanda M Verhoeven-Duif3, Qiuxiang Ma4, Peng Zhang4, Teresa B Fitzpatrick2, Wilhelm Gruissem1,4 & Hervé Vanderschuren1,5 1Department of Biology, ETH Zurich, Zurich,
Switzerland. 2Department of Botany and Plant Biology, University of Geneva, Geneva, Switzerland. 3Department of Medical Genetics, University Medical Center, Utrecht, the Netherlands. 4SIBS-ETH Shanghai Center for Cassava Biotechnology, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China. 5AgroBioChem Department, Gembloux Agro-Bio Tech, University of Liège, Liège, Belgium. e-mail: ([email protected], [email protected]) or ([email protected]) or ([email protected]).
1. Hellmann, H. & Mooney, S. Molecules 15, 442–459 (2010). 2. Dowell, S.F. et al. Emerg. Infect. Dis. 19, 1374–1384 (2013). 3. Voziyan, P.A. & Hudson, B.G. Cell. Mol. Life Sci. 62, 1671–1681 (2005). 4. Kannan, K. & Jain, S.K. Free Radic. Biol. Med. 36, 423–428 (2004). 5. Vogel, G. Science 336, 144–146 (2012). 6. Foltz, J.L. et al. PLoS One 8, e66419 (2013). 7. Albersen, M. et al. PLoS One 8, e54113 (2013). 8. Sayre, R. et al. Annu. Rev. Plant Biol. 62, 251–272 (2011). 9. Fitzpatrick, T.B. et al. Plant Cell 24, 395–414 (2012). 10. Tambasco-Studart, M. et al. Proc. Natl. Acad. Sci. USA 102, 13687–13692 (2005). 11. Vanderschuren, H. et al. Front. Plant Sci. 4, 143 (2013). 12. Raschke, M. et al. Plant J. 66, 414–432 (2011). 13. Chen, H. & Xiong, L. Plant Biotechnol. J. 7, 673–681 (2009). 14. Fagard, M. & Vaucheret, H. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51, 167–194 (2000). 15. Gregory, J.F. III. Annu. Rev. Nutr. 18, 277–296 (1998). 16. Szydlowski, N. et al. Plant J. 75, 40–52 (2013). 17. Said, H.M., Ortiz, A. & Ma, T.Y. Am. J. Physiol. Cell Physiol. 285, C1219–C1225 (2003). 18. Zhang, Z., Gregory, J.F. III & McCormick, D.B. J. Nutr. 123, 85–89 (1993). 19. Mackey, A.D., McMahon, R.J., Townsend, J.H. & Gregory, J.F. III. J. Nutr. 134, 842–846 (2004). 20. Sakurai, T., Asakura, T. & Matsuda, M. J. Nutr. Sci. Vitaminol. (Tokyo) 34, 179–187 (1988). 21. Gregory, J.F. III. Food Nutr. Res. 56, 5809 (2012). 22. Nyaboga, E. et al. Front. Plant Sci. 4, 526 (2013).
npg
© 2015 Nature America, Inc. All rights reserved.
Note: Any Supplementary Information and Source Data files are available in the online version of the paper (doi:10.1038/nbt.3318).
cassava plants. We are grateful to M. de Meyer (University of Geneva) and M. Bosma (University of Utrecht) for assistance with the microbial assays and the Caco-2 cell assay, respectively.
1032
VOLUME 33 NUMBER 10 OCTOBER 2015 NATURE BIOTECHNOLOGY
npg
© 2015 Nature America, Inc. All rights reserved.
Chinese government reaffirms backing for GM products To the Editor: This year’s ‘No. 1 Central Document’ (NCD)—a document issued every January by the Chinese Communist Party’s Central Committee and the State Council—states explicitly that China will put more resources into R&D on genetically modified (GM) products and into public education surrounding these products so as to regain the country’s position as a global leader in the area. Genetic modification has been discussed six times in the annual NCD between 2007 and 2015. Here we survey NCDs from 2007 to 2015 and infer that the top leadership’s opinion shifted like a pendulum toward GM technology, and then away in the face of public opposition—and now back again. In 2007, the NCD announced an intention to implement a GM food labeling system. It was followed by the “Mandatory Labeling of GM Food” policy released by China’s Ministry of Agriculture in 2007 (ref. 1). In 2008, the NCD started national research projects to cultivate new varieties of transgenic organisms. This was also followed by a National Science and Technology Major Project (NSTMP) program (released by China’s Ministry of Science and Technology in 2008) to invest about 24 billion yuan (~$3.9 billion) in developing GM technologies from 2008 to 2020. In 2009 and 2010, the two NCDs proposed to accelerate the transgenic NSTMP program2. And in 2009, the Biosafety Committee of China’s Ministry of Agriculture issued biosafety licenses for two crops, rice and corn3. This policy evolution—from asserting the need for labeling of GM products to concerted investment in the GM NSTMP program and regulatory approvals raised expectations that China might become a leader in commercializing GM rice and corn1,2. However, in the years following 2009, public resistance to GM products grew. And in China, public opinion often has a powerful influence in steering government policy in areas that are less politically controversial. For example, in the late 2000s, public environmental protests were instrumental in persuading local government officials to cancel construction of a billion-dollar chemical plant and a trash incinerator project4. Such environmental activism has also induced officials to disclose more environmental information5. As in the West, public
concerns about GM food safety are widespread. According to the Blue Book of Public Opinion Survey in China, only 0.9% of the respondents fully accept GM food6. In this context, the political neutrality of GM policy has meant that public safety concerns have trumped GM policy to the extent that a protechnology stance in the early 2000s was almost entirely reversed by 2010. Thus, from 2011 to 2014, the NCD failed to mention GM products at all. And the NCD in 2014 even neglected to use the term “GM,” preferring instead the euphemism “molecular breeding.” We contend that this political stalemate regarding GM products—reflected in the NCDs—has set back transgenic R&D in China and turned the country from being one of the world’s proponents of the technology to one of the most conservative developers. Between 1997 and 2001, China encouraged the planting of insect-resistant GM cotton and was the fourth largest GM crop grower globally. In 2014, China, with 3.9 million hectares of GM crops, ranked sixth in the world behind India (with 11.6 million hectares of GM crops)7. And in 2013, research funding for GM seed cultivation programs fell to around 400 million yuan (~$65 million), down from as much as 2 billion yuan in 2010 (ref. 8). In this context, we believe that the 2015 NCD represents an important
shift in government policy. It indicates a restatement of interest by the Central Committee and State Council in GM technology, which augurs well for R&D projects on GM products. It remains to be seen how fast this renewed political backing can be translated into a return of national funding into this important research area. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests.
Rongrong Li1, Qiang Wang2,3 & Alan McHughen4 1School of Business Administration, Xinjiang
University of Finance and Economics, Urumqi, Xinjiang, P.R. China. 2Xinjiang Institute of Ecology and Geography, CAS, Urumqi, P.R. China. 3China University of Petroleum (East China), Qingdao, P.R. China. 4Department of Botany & Plant Sciences, University of California-Riverside, California, USA. e-mail: [email protected] Qiu, J. Natl. Sci. Rev. 1, 466–470 (2014). Wang, Q. Nature 519, 7 (2015). Qiu, J. Nature 455, 850–852 (2008). Wang, Q. Science 328, 824 (2010). Wang, Q. Nature 497, 159 (2013). Xie, G.-Y. Blue Book of Public Opinion Surveys in China (Social Science Academic Press, Shanghai, China, 2014). 7. Clive, J. ISAAA brief No. 49 (2014). 8. Shuping, N. & Stanway, D. Reuters http://www.reuters. com/article/2014/03/10/china-parliament-g moidUSL3N0M41FR20140310 (10 March 2015). 1. 2. 3. 4. 5. 6.
Increased bioavailable vitamin B6 in field-grown transgenic cassava for dietary sufficiency To the Editor: Vitamin B6 deficiency is associated with cardiovascular disease, diabetes, neurological diseases1 and nodding syndrome (NS, a childhood condition in eastern Africa2). Pyridoxal 5’-phosphate (PLP), the cofactor form of vitamin B6, is required for multiple essential enzymatic reactions and other forms of vitamin B6, such as pyridoxine and pyridoxamine, inhibit oxygen radical production, pathogenic glycation and lipid peroxidation3,4. Notably, investigation of the causes of NS has revealed a high prevalence
NATURE BIOTECHNOLOGY VOLUME 33 NUMBER 10 OCTOBER 2015
of vitamin B6 deficiencies in affected African populations2,5,6. Humans cannot synthesize vitamin B6 de novo, so this micronutrient is obtained from food by intestinal absorption7. In this Correspondence, we report biofortification of cassava Manihot esculenta Crantz with vitamin B6 using transgenes from Arabidopsis thaliana. Cassava is a staple root crop for more than 250 million people in Africa8, including regions in which vitamin B6 deficiency is endemic6. In addition to the plant’s starchy roots, cassava leaves are also consumed as a 1029
Biosynthesis de novo
Salvage pathway and glycosylation
–
–
se T-a
NH2 O NH2
P-ase
Gln
PDX2 NH3
PLR PN
SOS4
PL
PD
S4
SO
X3
–
se
a
P-
PLP O
PM
–
PDX1
OH
O
SOS4
PMP
3
X PD
R5P
O
P-ase
L
–
NH2
G3P
Glu
PNG
PNP
WT 35S-5 Standards
800 600
PNG
400 200 PMP PM PNP
0
2
c
PL
PN
PLP
4 6 8 Retention time (min)
10
Phosphorylated B6 vitamers (PLP, PNP, PMP)
12
400 300
PNG
100 0
PMP PM PNP
0
2
PL
PN
PLP
4 6 8 Retention time (min)
Unphosphorylated B6 vitamers (PL, PN, PM)
80
**
50
**
40
*
30
**
20 10
*
0
Vitamin B6 content in roots (μg/g FW)
60
**
15
10
**
5 ** ** *
**
*
PA T-2
5 S35
S3 35
PA T-1 2
PA T-2
5 S35
S3 35
12
*
0
WT
10
PNG
20
70
WT PAT-12 Standards
200
WT
0
500
PA T-1 2
1,000
Fluorescence intensity
Fluorescence intensity
b
Vitamin B6 content in leaves (μg/g FW)
npg
a
G
1030
variable for nine field-grown, six-month-old plants of the transgenic lines, when compared with the wild-type control (Supplementary Table 3). Leaves and roots were collected from three independent plants in the first field experiment to test the stability of the improved vitamin B6 trait in the transgenic lines using three biological replicates. Transgenic 35S and PAT lines had greater total vitamin B6 content compared with the wild-type control (3.9- to 48.2-fold increase in transgenic leaves and a 1.9- to 5.8-fold
To evaluate the stability of the engineered vitamin B6 trait, we selected nine transgenic 35S and PAT lines with increased total vitamin B6 content (Supplementary Fig. 3) and evaluated them in the field. Wild-type and transgenic cassava lines propagated in vitro were first hardened in the greenhouse and then transferred to an experimental field in Shanghai for multiplication. Stem cuttings from the multiplied plants were then evaluated further in two field experiments. Plant height and storage root yield were
T
vegetable. Based on micronutrient data from raw cassava root, it has been suggested that a 2.3-fold increase in vitamin B6 content would be sufficient to meet the recommended dietary allowance for populations relying heavily on cassava in their diet9. Biosynthesis of vitamin B6 de novo in planta requires two enzymes, a synthase, PDX1, and a glutaminase, PDX2 (ref. 10 and Fig. 1a). Salvage and glycosylation pathways interconvert the B6 vitamers and their corresponding derivatives11–13 (Fig. 1a). We constructed cassettes for ubiquitous (CaMV35S promoter) and root-enhanced (Patatin promoter) heterologous expression of the A. thaliana PDX1.1 and PDX2 genes (AtPDX1.1 and AtPDX2, respectively; Supplementary Fig. 1). We hypothesized that this strategy would reduce the risk of gene silencing associated with overexpression of endogenous genes14. A subset of independent, single-insertion transgenic cassava lines generated with each of these constructs for either ubiquitous (five lines) or root-enhanced (four lines) expression (hereafter named 35S and PAT lines, respectively; Supplementary Fig. 2), were selected and transferred to the greenhouse. Plant height, as well as above- and belowground fresh weight, revealed no significant differences between wild-type, transgenic control–harboring empty pCAMBIA vector, and the transgenic 35S and PAT lines, except for lines 35S-1 and PAT-7, which had a significant but not unusual increase in plant height that is sometimes observed for cassava plants grown from tissue culture (Supplementary Table 1). Vitamin B6 measurements made using a microbial assay (Supplementary Note 1) showed that total vitamin B6 content was several-fold higher in the leaves and roots of four transgenic 35S and four PAT cassava lines (Supplementary Fig. 3) and correlated with the expression levels of AtPDX1.1 and AtPDX2 as measured by qPCR (Supplementary Fig. 4, Supplementary Table 2). We identified three orthologs of AtPDX1 and one ortholog of AtPDX2 in the cassava genome (Supplementary Fig. 5). Expression profiles of these orthologs were similar in wildtype and transgenic cassava lines with the exception of the M. esculenta PDX1.3a gene, which had a slightly reduced expression level in 35S-2 leaves (Supplementary Fig. 6). Together, these results indicate that expression of the AtPDX1.1 and AtPDX2 transgenes had no significant effect on the expression of the endogenous cassava genes but increased total vitamin B6 content in leaves and roots.
G
© 2015 Nature America, Inc. All rights reserved.
CORRESPONDENCE
Figure 1 Genetic engineering of vitamin B6. (a) Biosynthesis de novo, salvage pathways and glycosylation of vitamin B6. Glu, glutamate; Gln, glutamine; G3P, glyceraldehyde 3-phosphate; R5P, ribose 5-phosphate; PN, pyridoxine; PNG, pyridoxine 5’-b-d-glucoside; PL, pyridoxal; PM, pyridoxamine; PNP, pyridoxine 5’-phosphate; PLP, pyridoxal 5’-phosphate; PMP, pyridoxamine 5’-phosphate; T-ase, transaminase; P-ase, phosphatase; PLR, PL reductase; GT, glycosyl transferase; GL, glucosidase. (b) HPLC chromatogram of field-grown wild-type cassava and transgenic cassava expressing AtPDX1.1 and AtPDX2 under either the 35S promoter (transgenic line 35S-5) or Patatin promoter (transgenic line PAT-12). (c) Analysis of unphosphorylated, phosphorylated B6 vitamers and PNG in leaves and roots of wild-type and selected transgenic cassava lines grown under field conditions. Mean ± s.d. of three biological replicates. Student’s t-test (wild-type vs. transgenic line), *P < 0.05, **P < 0.01. Error bars, mean ± s.d. of three biological replicates.
VOLUME 33 NUMBER 10 OCTOBER 2015 NATURE BIOTECHNOLOGY
CORRESPONDENCE Table 1 Caco-2 cell assays with extracts from processed cassava leaves and roots a Compound
Pyridoxine (0 nM)
Pyridoxine (100 nM)
Pyridoxine (1,000 nM)
Wild-type leaves (boiled)
35S-5 leaves (boiled)
Wild-type roots (boiled)
PAT-12 roots (boiled)
Apical compartment at 0 h (nM) Pyridoxal
NA
NA
NA
115.1
77.1
178.5
25.1
Pyridoxine
NA
NA
NA
58.1
3,042.5
0
33.2
Pyridoxine
0
100
1,000
142.6
1,295.1
264.8
146.3
PMP
NA
NA
NA
73.6
108.4
50.5
47.8
PNP
NA
NA
NA
ND
ND
ND
ND
PNG
NA
NA
NA
398.3
4,634.5
243.5
703.1
60.5 (±9.3)
70.4 (±4.4)
41.9 (±5.5)
66.0 (±8.0)
50.8 (±4.6)
60.9 (±0.6)
Intracellular at 48 h (nM) PLP
43.6 (±10.1)
npg
© 2015 Nature America, Inc. All rights reserved.
Basolateral compartment after 48 h (nM) Pyridoxal
ND
13.1 (±1.9)
90.3 (±12.4)
11.1 (±0.5)
88.3 (±13.6)
13.3 (±0.3)
44.1 (±1.7)
Pyridoxine
ND
ND
12.1 (±2.1)
ND
6.2 (±5.4)
ND
ND
Pyridoxine
ND
ND
3.5 (±0.6)
ND
7.8 (±1.7)
ND
4.8 (±0.2)
PMP
ND
ND
ND
ND
ND
ND
ND
PNP
ND
ND
ND
ND
ND
ND
ND
PNG
ND
ND
1.7 (±2.8)
ND
26.1 (±8.2)
ND
3.7 (±6.3)
aAnalysis
of PLP in Caco-2 cells in the apical compartment and B6 vitamers and PNG in the apical (0 h) and basolateral (48 h) compartments. Equal amounts (10 g) of powder from leaves and roots were extracted in 20 ml of buffer, diluted tenfold and supplemented to the apical compartment. Mean ± s.d. of three biological replicates. NA, not applicable; ND, not detected.
increase in transgenic roots), which was consistent with the amounts previously measured in greenhouse-grown plants (Supplementary Fig. 7). These data confirm that the engineered vitamin B6 trait was stable over two cycles of vegetative propagation. Plants accumulate a substantial fraction of vitamin B6 as pyridoxine-5'-b-d-glucoside (PNG)15. Forms of vitamin B6 accumulating in wild-type control and transgenic cassava were analyzed using an established highperformance liquid chromatography (HPLC) method16. Comparison of chromatograms revealed significant differences between wildtype control and four selected transgenic lines (Fig. 1b). Our analyses revealed that the increased vitamin B6 content of transgenic lines was mostly due to an increase in unphosphorylated forms (4.6- to 14.8-fold increase in leaves and 1.7- to 5.8-fold in roots) and of PNG (6.1- to 9.7-fold increase in leaves and 5.4- to 30.7-fold increase in roots) (Fig. 1b,c and Supplementary Table 4). We observed no increase of phosphorylated forms in leaves and roots of transgenic lines, with the exception of the roots of the PAT-12 line (threefold increase). Cassava leaves and storage roots are usually boiled in water before consumption, a step that removes toxic cyanogens. This process could affect vitamin B6 content due to the instability of this vitamin at high temperature as well as its solubility in water. Vitamin B6 could, therefore, be reduced in a cassava meal. In all cases, boiling decreased the relative vitamin B6 content of leaves (65%, 15% and 34% decrease for tested wild-type line, transgenic line 35S-5 and
transgenic line 35S-3, respectively) and roots (51%, 71% and 75% for tested wild-type line, transgenic line PAT-12 and transgenic line PAT-2, respectively) based on fresh weight (Supplementary Tables 4 and 5). Nonetheless, transgenic leaves and roots retained significantly (P value < 0.01) higher amounts of unphosphorylated B6 forms over wild-type control (ninefold increase in transgenic leaves and fourfold increase in transgenic roots; see Supplementary Table 5) and PNG (40-fold increase in transgenic leaves and 14-fold increase in transgenic roots: see Supplementary Table 5) even after boiling. B6 vitamers and PNG from food sources have different bioavailability in humans15. Inhibition of pyridoxine uptake by pyridoxamine in human intestinal epithelial Caco-2 (colorectal adenocarcinoma) cells17, as well as the competitive inhibition of pyridoxine uptake by PNG in rat hepatocytes18, suggest that B6 vitamer composition might also influence the bioavailability of total vitamin B6 from food sources. We used a Caco-2 cell assay to test the bioavailability of vitamin B6 from wild-type and transgenic cassava. Leaves and roots were first boiled for 30 min as in routine cassava preparation. Leaf and root extracts were then supplied to the apical compartment of the assay (Supplementary Fig. 8) and Caco-2 cells were incubated for 48 h. According to current models for intestinal metabolism of vitamin B6, PL, PM and PN are readily absorbed, converted into PLP through intracellular salvage pathway enzymes (Fig. 1a) and subsequently
NATURE BIOTECHNOLOGY VOLUME 33 NUMBER 10 OCTOBER 2015
dephosphorylated to be excreted as PL in the portal vein. PNG is converted into PN before uptake, transformed into PLP intracellularly and also excreted into the portal vein as PL7. Therefore, measurement of pyridoxal concentrations in the basolateral compartment (Supplementary Fig. 8) provides a robust estimation of vitamin B6 bioavailability in the intestine. Incubation of Caco-2 cells with extracts from cassava leaves and storage roots resulted in an eightfold and fourfold, respectively, increase of pyridoxal levels (transgenic line versus wild type) (Table 1), showing there were higher levels of bioavailable vitamin B6 in the transgenic cassava. Incubation of Caco-2 cells with extracts from transgenic cassava or an excess amount of pyridoxine (1 mM) also resulted in trace accumulation of pyridoxine and pyridoxamine in the basolateral compartment (Table 1 and Supplementary Fig. 8). When the maximum capacity of the intestine to convert pyridoxine and pyridoxamine into PLP is exceeded, these B6 vitamers are excreted without intracellular conversion7,20. The intracellular PLP levels were consistent with a higher conversion of precursor B6 vitamers into PLP by Caco-2 cells incubated with leaf and root extracts from transgenic cassava (Table 1). Using bioavailable ‘vitamin B6 equivalents’21, we calculated that the vitamin B6 recommended dietary allowance for an adult person (1.3 mg/day) would be reached with 51 g of boiled 35S-5 leaves or 505 g (~1.7 lb) of boiled PAT12 storage roots (Supplementary Note 1). This study shows that cassava can be biofortified using the AtPDX-based 1031
CORRESPONDENCE engineering strategy to increase vitamin B6. The engineered trait was stable under field conditions and the engineered vitamin B6 was bioavailable. Because cassava is a staple crop in those African regions where vitamin B6 deficiency is endemic, AtPDX transgenic vitamin B6-enhanced cassava could be included in a portfolio of dietary and supplementation approaches to help reduce this important nutritional deficiency. Further characterization of the vitamin B6–enhanced transgenic lines, including location-specific agronomic field performance tests, will be the next step before introgressing, or engineering, the increased vitamin B6 trait into cassava cultivars preferred by local farmers and consumers22.
ACKNOWLEDGMENTS We acknowledge support from ETH Zurich, the University of Geneva, the Swiss National Science Foundation (project 31003A_140911) and the Earmarked Fund for the China Agriculture Research System (CARS-12-shzp). We thank S. Klarer and I. Zurkirchen (ETH Zurich) for special care of the
COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests.
Kuan-Te Li1, Michael Moulin2, Nathalie Mangel1, Monique Albersen3, Nanda M Verhoeven-Duif3, Qiuxiang Ma4, Peng Zhang4, Teresa B Fitzpatrick2, Wilhelm Gruissem1,4 & Hervé Vanderschuren1,5 1Department of Biology, ETH Zurich, Zurich,
Switzerland. 2Department of Botany and Plant Biology, University of Geneva, Geneva, Switzerland. 3Department of Medical Genetics, University Medical Center, Utrecht, the Netherlands. 4SIBS-ETH Shanghai Center for Cassava Biotechnology, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China. 5AgroBioChem Department, Gembloux Agro-Bio Tech, University of Liège, Liège, Belgium. e-mail: ([email protected], [email protected]) or ([email protected]) or ([email protected]).
1. Hellmann, H. & Mooney, S. Molecules 15, 442–459 (2010). 2. Dowell, S.F. et al. Emerg. Infect. Dis. 19, 1374–1384 (2013). 3. Voziyan, P.A. & Hudson, B.G. Cell. Mol. Life Sci. 62, 1671–1681 (2005). 4. Kannan, K. & Jain, S.K. Free Radic. Biol. Med. 36, 423–428 (2004). 5. Vogel, G. Science 336, 144–146 (2012). 6. Foltz, J.L. et al. PLoS One 8, e66419 (2013). 7. Albersen, M. et al. PLoS One 8, e54113 (2013). 8. Sayre, R. et al. Annu. Rev. Plant Biol. 62, 251–272 (2011). 9. Fitzpatrick, T.B. et al. Plant Cell 24, 395–414 (2012). 10. Tambasco-Studart, M. et al. Proc. Natl. Acad. Sci. USA 102, 13687–13692 (2005). 11. Vanderschuren, H. et al. Front. Plant Sci. 4, 143 (2013). 12. Raschke, M. et al. Plant J. 66, 414–432 (2011). 13. Chen, H. & Xiong, L. Plant Biotechnol. J. 7, 673–681 (2009). 14. Fagard, M. & Vaucheret, H. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51, 167–194 (2000). 15. Gregory, J.F. III. Annu. Rev. Nutr. 18, 277–296 (1998). 16. Szydlowski, N. et al. Plant J. 75, 40–52 (2013). 17. Said, H.M., Ortiz, A. & Ma, T.Y. Am. J. Physiol. Cell Physiol. 285, C1219–C1225 (2003). 18. Zhang, Z., Gregory, J.F. III & McCormick, D.B. J. Nutr. 123, 85–89 (1993). 19. Mackey, A.D., McMahon, R.J., Townsend, J.H. & Gregory, J.F. III. J. Nutr. 134, 842–846 (2004). 20. Sakurai, T., Asakura, T. & Matsuda, M. J. Nutr. Sci. Vitaminol. (Tokyo) 34, 179–187 (1988). 21. Gregory, J.F. III. Food Nutr. Res. 56, 5809 (2012). 22. Nyaboga, E. et al. Front. Plant Sci. 4, 526 (2013).
npg
© 2015 Nature America, Inc. All rights reserved.
Note: Any Supplementary Information and Source Data files are available in the online version of the paper (doi:10.1038/nbt.3318).
cassava plants. We are grateful to M. de Meyer (University of Geneva) and M. Bosma (University of Utrecht) for assistance with the microbial assays and the Caco-2 cell assay, respectively.
1032
VOLUME 33 NUMBER 10 OCTOBER 2015 NATURE BIOTECHNOLOGY
