A general review on metabolic engineering of valuable plants as good sources of bioactive compounds.

1. Metabolic engineering approaches
in medicinal plants
By: Naghmeh Poorinmohammad
May 2015

2. Introduction
• Plants are the richest with secondary metabolites among different organisms (5.000
– 25.000 per plant).
• Plants provide the source material for over half the drugs currently prescribed.
• Often, bioactive compounds in plants are produced in very small quantities.
10,000 kg of Pacific yew bark yields less than 1 kg of the potent anti-cancer
compound paclitaxel!
Chandra et al. Biotechnology for Medicinal Plants. Springer, 2014: P275 1

3. Introduction
• Shikimic acid pathway, non-mevalonate (MEP) pathway and mevalonate (MVA)
pathway lead to diverse classes of compounds, which include the terpenoids,
monoterpene indole alkaloids, isoquinoline alkaloids, flavonoids and anthocyanins.
Wilson et al. Current opinion in biotechnology 26 (2014): 174-182. 2

4. Introduction
Challenges
Small
quantities
Undesirable
properties
Low
number of
candidates
3

5. Introduction
A good
Solution!
Metabolic
Engineering
(ME)
Enhancement
Suppression
Sequestration or
diversification
4

6. ME approaches theory
Farré et al. Annual review of plant biology 65 (2014): 187-223. 5

7. ME approaches theory
STOP!
We must first gain knowledge about the plant’s
metabolism.
Remember!
The use of term “engineering” implies that there
is some precise understanding of the system that
is being modified.
6

8. Understanding plant metabolism:
A systems biological approach
• To be able to manipulate plant metabolism, one must first create a metabolic model.
• Constructing a metabolic model using a systems biological approach requires the
four steps:
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
7

9. Understanding plant metabolism:
A systems biological approach
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
Step 1  Find enzymes/reactions:
- Literature search and databases
- Gene annotation
- Experimental procedures
Step 2  Identification of inter-compartmental transport
reactions:
- The most challenging step as information in this
area remains, for the most part, uncharacterized.
Step 3  Determine the reversibility of reactions
- There are some obvious reactions
- Some others can be concluded thermodynamically
- Some algorithms do exist
8

10. Major Databases Containing Metabolic Information on Plants
Baghalian et al. The Plant Cell Online 26.10 (2014): 3847-3866. 9

11. metacyc.org
10

12. http://omictools.com/13c-fluxomics-c1414-p1.html
11

13. Compartmentation makes plant genome-scale
reconstruction challenging: even for Arabidopsis,
the most intensively studied plant species, this
knowledge is far from complete.
de Oliveira et al. Current opinion in biotechnology24.2 (2013): 271-277. 12

14. Understanding plant metabolism:
A systems biological approach
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
• While step one provides a static view of the metabolic
network, mathematical methods are required in order
to process and integrate heterogeneous omics data and
to build a comprehensive metabolic model.
• Computational platforms have been developed to
make the mathematical analysis and visualization
convenient.
• Mathematical modeling methods:
- Network Models
- Stoichiometric Models
- Genome-Scale Metabolic Models
- Metabolic flux analysis (MFA) (moghayesash ba
gene essentiality data)
- Kinetic Models
13

15. Definitions of Terms Used in Metabolic Modeling
Baghalian et al. The Plant Cell Online 26.10 (2014): 3847-3866. 14

16. Stoichiometric Models
Baghalian et al. The Plant Cell Online 26.10 (2014): 3847-3866. 15

17. Understanding plant metabolism:
A systems biological approach
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
• A valid model faithfully reflects the biologically realistic
behavior of the metabolic network.
• Lack of concordance between observed and predicted
behavior requires re-elaboration of the model to
remove inconsistencies.
• A model becomes acceptable as soon as the outcome
of future experiments can be predicted.
• It is also an overriding priority to continuously update
the validated model by reference to new findings.
• 70 to 90% similarity between experiment and
prediction is a good result.
16

18. Understanding plant metabolism:
A systems biological approach
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
• A valid model can be regarded as a virtual laboratory,
so predictions can be made much faster and more
cheaply than by conducting the necessary wet lab
experiments.
• The intention is to manipulate a given pathway to
produce a specific outcome, applying the model can
potentially generate a variety of alternative strategies,
which can lead to a directed experimental validation.
• NFA is one of the important part of systems biology and
has the potential to make predictions on the basis of a
functional understanding.
17

19. flux balance analysis of
large networks
de Oliveira et al. Current opinion in biotechnology24.2 (2013): 271-277. 18

20. ME approaches theory
19

21. CASE STUDY 1
20

22. CASE STUDY 1
Hasan et al. Plant cell reports 33.6 (2014): 895-904. 21

23. ME approaches theory
22

24. • GGPP, the precursor for both diterpenoids and tetraterpenoids, was
accumulated by suppressing the pathway to carotenoids.
• Two genes for the pathway, phytoene synthase (PSY) and phytoene
desaturase (PDS) in the first and second step for the pathway of carotenoid
synthesis respectively, were silenced
• How?
Partial products of PSY and PDS were prepared by PCR and used to
produce the pTRV2:PSY and pTRV2:PDS plasmid.
Transformation of A.tumefaciens strain GV2260 with the
plasmids was acomplished by the freez-thaw method.
To confirm successful silencing of the PSY and PDS gene, total
RNA was isolated and visualized on agarose gel.
CASE STUDY 1
Hasan et al. Plant cell reports 33.6 (2014): 895-904. 23

25. • Identification and quantification of the amounts of taxadiene were
conducted by GC–MS.
• Metabolic pathway shunting by suppression of the phytoene synthase gene
expression which resulted in accumulation of increased taxadiene
accumulation by 1.4- or 1.9- fold, respectively. In
CASE STUDY 1
Hasan et al. Plant cell reports 33.6 (2014): 895-904. 24

26. 1. Knocking out gene function by targeted RNA degradation.
2. Interfering with protein function using specific inhibitors
or antibodies.
Other methods for gene silencing??
Wurtzel et al. Encyclopedia of Chemical Processing (2006): 2191-2200. 25

27. CASE STUDY 2
26

28. CASE STUDY 2
• The medicinal plant Catharanthus roseus is of enormous pharmaceutical interest
because it contains more than 120 terpenoid indole alkaloids (TIAs):
Ajmalicine  antihypertensive
vinblastine and vincristine  antineoplastic
They are produced only in very low amounts in C. roseus plants (2) and,
despite significant efforts, cell cultures are not yet a valid alternative
for production.
Rischer et al. Proceedings of the National Academy of Sciences 103.14 (2006): 5614-5619. 27

29. CASE STUDY 2
Metabolite Profiling Transcipt profiling
Identify the conditions where
differential accumulation of
the desired metabolites can be
observed.
TIA-targeted metabolite
profiling
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
LC/MS  Peak filtering  178
peaks
Using an internal library of
masses and retention times,
TIAs were identified
Rischer et al. Proceedings of the National Academy of Sciences 103.14 (2006): 5614-5619. 28

30. CASE STUDY 2
Metabolite Profiling Transcipt profiling
Identify the conditions where
differential accumulation of
the desired metabolites can be
observed.
TIA-targeted metabolite
profiling
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
LC/MS  Peak filtering  178
peaks
Using an internal library of
masses and retention times,
TIAs were identified
cDNA-AFLP technique
Sequencing  BLAST with a
library (European Molecular
Biology Laboratory)
Less than 10% gave a perfect
match. Thus, the vast majority
of the tags
are undescribed.
Rischer et al. Proceedings of the National Academy of Sciences 103.14 (2006): 5614-5619. 28

31. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
• The accumulation profiles of the 178 metabolite
peaks retained were combined with the expression
profiles of the 417 transcripts for integrated
analysis.
• principal component analysis (PCA) was performed
first to explore the variability structure of the data.
Rischer et al. Proceedings of the National Academy of Sciences 103.14 (2006): 5614-5619. 29

32. Rischer et al. Proceedings of the National Academy of
Sciences 103.14 (2006): 5614-5619.
30

33. 31

34. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
• Correlation network analysis was used to establish
gene-to-gene and gene-to-metabolite co-regulation
patterns
• The Pearson correlation coefficient between each
pair of variables (either gene or metabolite) across
the profiles, including all time points and
conditions, was calculated.
• TOM SAWYER VISUALIZATION 6.0  visualization
Rischer et al. Proceedings of the National Academy of Sciences 103.14 (2006): 5614-5619. 32

35. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
Rischer et
al. Proceedings of the
National Academy of
Sciences 103.14
(2006): 5614-5619.
33

36. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
Rischer et
al. Proceedings of the
National Academy of
Sciences 103.14
(2006): 5614-5619.
34

37. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
Rischer et al. Proceedings of the National Academy of Sciences 103.14 (2006): 5614-5619. 35

38. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
Rischer et al. Proceedings of the National Academy of Sciences 103.14 (2006): 5614-5619. 36

39. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
Rischer et al. Proceedings of the National Academy of Sciences 103.14 (2006): 5614-5619. 37

40. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
• The power of this approach was also successfully
demonstrated by other studies of taxol biosynthesis
and by the author’s own study of tobacco nicotine
biosynthesis.
38

41. CASE STUDY 2
Establishment of
the Metabolic
Network
Convert
reconstruction to
Mathematical
Model and
Visualization
Validation
Analytical
Investigation
A publishable full research data
provided
39

42. CASE STUDY 2
40

43. CASE STUDY 2
Zhou et al. Biotechnology and applied biochemistry 52.4 (2009): 313-323. 41

44. CASE STUDY 2
• Genes encoding rate-limiting enzymes and some key transcription factors can be
used to improve TIA production by overexpressing them in transgenic C. roseus
cultures.
• Tdc cDNA driven by CaMV 35S (cauliflower-mosaic-virus 35 S promoter) was
introduced into C. roseus by means of A. tumefaciens-mediated gene
transformation.
.
Zhou et al. Biotechnology and applied biochemistry 52.4 (2009): 313-323. 42

45. CASE STUDY 2
Zhou et al. Biotechnology and applied biochemistry 52.4 (2009): 313-323. 43

46. CASE STUDY 2
• Genes encoding rate-limiting enzymes and some key transcription factors can
be used to improve TIA production by overexpressing them in transgenic C.
roseus cultures.
• Tdc cDNA driven by CaMV 35S (cauliflower-mosaic-virus 35 S promoter) was
introduced into C. roseus by means of A. tumefaciens-mediated gene
transformation.
• Overexpression of Tdc was not necessary to achieve high levels of alkaloid
accumulation, but only enhanced tryptamine levels, and constitutively high
TDC activity seemed to be detrimental to C. roseus growth.
• The Str-overexpressing cultures showed a 10-fold higher STR activity than
wildtype, resulting in higher TIA accumulation.
Zhou et al. Biotechnology and applied biochemistry 52.4 (2009): 313-323. 44

47. Other methods for increasing expression??
1. Introduce genes into the plant.
2. Promoters to direct gene expression in the appropriate spatial
and temporal landscape.
Wurtzel et al. Encyclopedia of Chemical Processing (2006): 2191-2200. 45

48. • Although most early examples of metabolic engineering involved single-gene
interventions, this approach suffers from limitations such as the inability to
increase the activity of multimeric enzymes with heterologous subunits, and the
inability to target multiple metabolites simultaneously.
Multigene transfer
Farré et al. Annual review of plant biology 65 (2014): 187-223. 46

49. Multigene transfer
Farré et al. Annual review of plant biology 65 (2014): 187-223. 47

50. Increasing metabolic diversity
• The common building blocks of the diterpenoid pathways include various
diterpene synthases, cytochrome P450 monooxygenases,
glycosyltransferases, acyltransferases, methyltransferases, and
oxidoreductases that act specifically on different parts of the skeleton.
• Combining these enzymes in a common background should make it possible
to generate novel forms of decoration (or at least novel combinations of
decoration) on a single diterpene skeleton.
• Similar building blocks are required for the synthesis of triterpenes.
• For example, the production of avenacin 1 requires many enzymes with
different roles, but only five of the corresponding genes have been isolated.
• These genes have been expressed in different combinations in heterologous
plants—e.g., the expression of AsMT1, AsUGT74H5, and AsSCPL showed that it
was possible to achieve the acetylation of the triterpene backbone.
Farré et al. Annual review of plant biology 65 (2014): 187-223. 48

51. Challenges
Farré et al. Annual review of plant biology 65 (2014): 187-223. 49

52. Summary
 Metabolic engineering is in principle a simple process of enhancing the capacity
of existing pathways, controlling the distribution of flux, and, if necessary,
bolting on additional functionalities so that novel compounds can be produced.
 In practice, there are at least four major challenges that need to be overcome:
- Gaining enough knowledge of the endogenous pathways to know the
best intervention points.
- Identifying and sourcing the genes that can be used to modify the
targeted metabolic pathway.
- Expressing those genes in such a way as to produce a functional enzyme
in a relevant context (such as the correct subcellular compartment).
- Achieving the primary goals of the metabolic engineering strategy
without affecting endogenous metabolism to the extent that the plant
cannot grow and develop normally.
50

53. Summary
 The valuable pharmacological and nutritional properties of many
secondary metabolites combined with their low levels of production in
nature mean that active research into novel strategies for metabolic
engineering will become increasingly important.
 The evolution of metabolic engineering from single- to multistep
approaches, along with high-throughput methods for gene discovery and
functional analysis and novel platforms for combinatorial testing of
heterologous pathways in plants, will increase the predictive accuracy of
early development stages.
 Thus helping to refine engineering strategies and reduce the need for
trial-and-error testing.
51

54. Thank You