Transcript IB496-April 18

Metabolomics, spring 06
Metabolomics Essentiality
Hans Bohnert
ERML 196
[email protected]
265-5475
333-5574
http://www.life.uiuc.edu/bohnert/
class April 18
One important aspect of plant genomics,
probably the most important one,
will be to define cell-specific
transcript profiles under a variety of conditions
Chris Somerville, April 2006
This will be equally important for
metabolite concentrations, the flux through pathways
and changes in photosynthesis during the day,
under stresses and in sinks.
Today’s discussion topic (single cell profiles):
Fan TWM, Bandura LL, Higashi RM, Lane AN (2005) Metabolomics-edited
transcriptomics of Se anticancer action in human lung cancer cells.
Metabolomics 1, 325.
Pvalue
0.00
0.11
citric acid
malic acid
p-hydroxybenzoic
acid
Total
0.73
0.77
0.05
0.10
0.55
0.86
0.09
maltose
trehalose
Total
1.55
0.80
1.08
0.09
0.31
0.53
mannitol
glycerol
inositol
Total
1.33
0.87
0.96
0.97
0.11
0.23
0.84
0.74
glucose-6-P
0.55
0.10
phosphate
0.71
0.24
0.02*
Amino
acids
0.39
0.61
0.56
Organic
acids
glutamic acid
threonine
Total
Absolute change and
Confidence
0.07*
Sugars
Col 21 CO2/
ambient
fold change
Alcohols
Metabolites
Thimm et al. (2004) Mapman: a userdriven tool to display genomics data
sets onto diagrams of metabolic
pathways and other biological
Processes. Plant Journal 37, 914.
Usadel et al. (2005) Extension of the
visualization tool MapMan to allow
statistical analysis of arrays,
display of corresponding genes,
and comparison with known
responses. Plant Physiol. 138, 1195.
In total
~60 metabolites
could be scored
60
• all metabolites
40
F2 16.92 %
20
• biological repeats
Col June 27 A
Col June 21 C
• scale according to
factors analyzing
similarity in the
response
Col June 21 A
0
Cvi June 21 C
Col June 27 C
Cvi June 21 A
Cvi June 27 A
-20
Cvi June 27 C
The chosen example
explains >90% of the
variability
-40
-60
-60
-40
-20
0
20
40
60
F1 77.71 %
Supplemental Figure 5.
Linear discriminant analysis, LDA
(Catchpole et al., 2005)
June 27
2
Organic acid
6 carbon sugars
12 carbon sugars
Sugars
18 carbon sugars
1.5
1
up
down early,
up late
0
-1.5
-1
0
-0.5
0.5
-0.5
down
Amino acid
-2
Figure 6.
1.5
Metabolite changes
(major categories)
over time points
up early
down late
-1
-1.5
1
June 21
0.5
Polyols
Col-0
Cvi-0
Cook et al. (2004) A prominent role for the CBF
cold response pathway in configuring the low-temperature
metabolome of Arabidopsis. PNAS 101, 15243.
Effect of cold treatment on the
Arabidopsis metabolome
CBF3 confers metabolic
signatures in non-acclimated
plants similar to that in
acclimated plants
Low temperature and Cvi
Cvi – cold treatment and the raffinose pathway
Cvi – deficiency in metabolites that increase in WS
Catchpole et al. (2005) Hierarchical metabolomics demonstrates substantial
compositional similarity between genetically modified and conventional potato
crops. PNAS 102, 14458.
We show that, apart from targeted changes, these GM potatoes in
this study appear substantially equivalent to traditional cultivars.
PCA
LDA (Df1 vs. Df2
Df2 vs. Df3)
• metabolite fingerprints
GC-TOF-MS
• potato cultivars
• GMOs (SST & SST/FFT)
fructan 1-fructosyltransferase
• matrix presentation of
sample frequency in
predicted/observed
diagram fashion
• most differences in
fructan polymerization
• “substantial
equivalence”
Overlaid single-ion chromatograms
potato diagrams vs. sample
of different degree of polymerization
of fructans
(fructose
polymers,
n >10)
GC-TOF-MS of m/z 217 in cultivars
Desiree (= precursor for GMO)
lacks kestose –
which is present in GMO
SST & SST/FFT
NMR
• Absorption by nuclei [not electrons] of electromagnetic radiation (up to ~900 MHz)
• Certain nuclei with spin and magnetic moment split energy levels in a field
• The split is characteristic of the nucleus and the bonds in which it is involved
• Continuous wave (CW) and pulsed (Fourier-transformed, FT) spectrometers
• http://en.wikipedia.org/wiki/Nuclear_magnetic_resonance
What NMR signals mean
1,3-butanediol
chemical shift
imprinted by
neighboring nuclei
characteristic for each bond
two-dimensional
(change field by 90o
repeated scans at
different frequencies)
(1)
(2)
(3)
(4)
quartet
doublet
triplet
triplet
one-dimensional
compare signals with a
library of known signals
Chemical shift is usually expressed in parts per million (ppm) by frequency,
because it is calculated from:
Since the numerator is usually in hertz, and the denominator in megahertz,
delta is expressed in ppm.
The detected frequencies (in Hz) for 1H, 13C, and 29Si nuclei are usually referenced
against TMS (tetramethylsilane), which is assigned the chemical shift of zero.
Other standard materials are used for setting the chemical shift for other nuclei.
The operating frequency of a magnet is calculate from the Larmor equation:
Flarmor = γ * B0, where B0 is the actual strength of the magnet
in units like teslas or gauss, and
γ is the gyromagnetic ratio of the nucleus being tested.
Not only 13C or 1H – other atoms as well can be seen
Isotope
Occurre
nce
in nature
(%)
H
99.984
1/2
2.79628
0.016
1
0.85739
2.8 x 10
B
18.8
3
1.8005
7.4 x 10
B
81.2
3/2
2.6880
2.6 x 10
98.9
0
1.1
1/2
0.70220
99.64
1
0.40358
0.37
1/2
−0.28304
99.76
0
0.0317
5/2
1
2
H
10
11
12
C
13
C
14
N
15
N
16
O
17
O
spin
number l
Magnetic
moment
μ
(A·m²)
−1.8930
Electric
quadrupole
moment
-24
2
(e×10 cm )
7.1 x 10
-3
-2
-2
-2
−4.0 x 10
-3
Frequen
cy at 7 T
(MHz)
Relative
sensitivit
y
300.13
1
46.07
0.0964
32.25
0.0199
96.29
0.165
75.47
0.0159
21.68
0.00101
30.41
0.00104
40.69
0.0291
What NMR signals mean
Fan, Bandura, Higashi & Lane (2005) Metabolomics 1, 325-339
Metabolomics-edited transcriptomics analysis of
Se anticancer action in human lung cancer cells
(META)
Transcriptomic analysis is an essential tool for systems biology but it has been stymied by a lack of global
understanding of genomic functions, resulting in the inability to link functionally disparate gene expression
events. Using the anticancer agent selenite and human lung cancer A549 cells as a model system, we
demonstrate that these difficulties can be overcome by a progressive approach which harnesses the emerging
power of metabolomics for transcriptomic analysis. We have named the approach Metabolomics-edited transcriptomic
analysis (META). The main analytical engine was 13C isotopomer profiling using a combination of multi-nuclear 2-D
NMR and GC-MS techniques. Using 13C-glucose as a tracer, multiple disruptions to the central metabolic network in
A549 cells induced by selenite were defined. META was then achieved by coupling the metabolic dysfunctions
to altered gene expression profiles to: (1) provide new insights into the regulatory network underlying the metabolic
dysfunctions; (2) enable the assembly of disparate gene expression events into functional pathways that was not
feasible by transcriptomic analysis alone. This was illustrated in particular by the connection of mitochondrial
dysfunctions to perturbed lipid metabolism via the AMP-AMPK pathway. Thus, META generated both extensive and
highly specific working hypotheses for further validation, thereby accelerating the resolution of complex biological
problems such as the anticancer mechanism of selenite.
Key words
(3-6) two-dimensional NMR; GC-tandem MS; 13C isotopomer profiling; selenite; lung adenocarcinoma A549 cells.
Abbreviations
1H–13C HMBC: 1H–13C heteronuclear multiple bond correlation spectroscopy;
1H–13C HSQC: 1H–13C heteronuclear single quantum coherence spectroscopy;
2-D 1H TOCSY: two dimensional 1H total correlation spectroscopy;
[U)13C]-glucose: uniformly 13C-labeled glucose;
MSn: mass spectrometry to the nth dimension;
MTBSTFA: N-methyl-N-[tert-butyldimethylsilyl]trifluoroacetamide;
P-choline or PC: phosphorylcholine;
PDA: photodiode array;
TCA: trichloroacetic acid.
Knowledge:
Se is an essential atom, high amounts affect (cancer) growth, Se in
proteins is related to ROS homeostasis (somehow!)
Experiment:
The addition of Se to lung cells affects growth – what is the basis?
Use genomics platforms (transcript analysis), GC-MS & esp. NMR
Hypothesis:
gene expression is altered, and metabolite analysis can be
correlated with transcript changes – can it, is the question!
Approaches
Microscopy, NMR, GC-MS, transcripts
Se interferes with the cytoskeleton and mitochondrial activity
Selenite effects proliferating cells;
Selenite-rich diets may have anti-cancer
applications.
Se leads to degradation of DNA
TUNEL assay?
control
Se treated
High resolution 2D NMR spectra of control and Se-treated cells
*
*
“1H chemical shift”
Metabolites with chemical shift indicative of changes 12C/13C and 1H connectivity
Se-cells (13C-glc)
spectral differences
control/Se
GC-MS + NMR
absolute amount
labeled positions (12C-13C)
high
resolution
Control 1D
Control 2D
down
up
*depletion 13C