Transcript Plant Physiology Studies Using Positron Emission Imaging

Measuring Dynamic Biological
Responses of Plants to Global
Change using Short-lived
Radioisotopes
Calvin Howell
Duke University Physics
Triangle Universities Nuclear Laboratory
Outline
•
•
•
•
•
•
The TUNL-Phytotron Collaboration
Motivation
Status of Plant Studies with Radioisotopes
Plant Physiology Basics
Demonstration of Technique
Immediate Plans
March 1, 2006
University of Notre Dame
2
TUNL-Phytotron Collaboration
C.R. Howell (Physics)
C. Reid (Biology)
E. Bernhardt (Biology)
A.S. Crowell (Physics Postdoc)
M. Kiser (Physics graduate student)
R. Phillips (Biology Postdoc)
March 1, 2006
University of Notre Dame
3
What is a Phytotron?
• Controlled Environment Facility
• Growth chambers can control many factors:
–
–
–
–
–
–
–
Soil type
Air Temperature
Light levels (total & UV)
Carbon dioxide concentration
Relative humidity
Nutrients
Air pollutants
March 1, 2006
University of Notre Dame
4
Outline
•
•
•
•
•
•
The TUNL-Phytotron Collaboration
Motivation
Status of Plant Studies with Radioisotopes
Plant Physiology Basics
Demonstration of Technique
Immediate Plans
March 1, 2006
University of Notre Dame
5
Motivations
“Industrial Revolution”
March 1, 2006
University of Notre Dame
6
Motivations
Climate models predict
atmospheric CO2 levels
will double by the end
of this century!
How will plants respond?
Intergovernmental Panel on Climate Change (IPCC): Climate Change 2001,
“The
Carbon
Cycle and Atmospheric Carbon
Dioxide” of Notre Dame
March
1, 2006
University
7
Carbon Budget
Intergovernmental Panel on Climate Change (IPCC): Climate Change 2001,
“The Carbon Cycle and Atmospheric Carbon Dioxide”
Sinks in units of billions of metric tons of carbon (GtC)
Fluxes in units of billions of metric tons of carbon per year (GtC/year)
March 1, 2006
University of Notre Dame
8
Carbon Budget
Intergovernmental Panel on Climate Change (IPCC): Climate Change 2001,
“The Carbon Cycle and Atmospheric Carbon Dioxide”
Sinks in units of billions of metric tons of carbon (GtC)
Fluxes in units of billions of metric tons of carbon per year (GtC/year)
March 1, 2006
University of Notre Dame
9
Interesting Aside
• Total tonnage of CO2
produced by vehicles
over 124,000 mile
lifetime
• Assuming ~10 year
lifetime, vehicles emit
more than their own
weight in CO2 per year
13 mpg
18 mpg
22 mpg
36 mpg
65 mpg
March 1, 2006
University of Notre
Dame
10
http://www.sierraclub.org/globalwarming/suvreport/pollution.asp
March 1, 2006
University of Notre Dame
11
Grassland Response
Multiple Factors:
(C) CO2 ; 680 ppm
(T) Temperature; +80 W/m2
(P) Precipitation; +50%
(N) Nitrogen; +7g/m2 year
M. Rebecca Shaw et al., Science
298:1987-1990 (2002) –
Carnegie Institute of
Washington and Stanford Univ.
March 1, 2006
University of Notre Dame
12
FACE Studies
• Free Air CO2 Enrichment (FACE)
experiments
– Large-scale research programs to
study effects of increased CO2 levels
– Many environmental variables
– Difficult to correlate growth
parameters with high precision
• Findings from forest stands
– Initially, carbon stored in wood
– 2 years later, less found in wood, but
more than double in fine roots
– Nearly half of carbon uptake in shortlived tissues, such as foliage
– Increase in net primary production of
25%
– Growth rate increased about 26%
– Limited N  no appreciable change
March 1, 2006
University of Notre Dame
Duke FACTS-I Aerial View
13
FACE Sites
March 1, 2006
University of Notre Dame
14
Outline
•
•
•
•
•
•
The TUNL-Phytotron Collaboration
Motivation
Status of Plant Studies with Radioisotopes
Plant Physiology Basics
Demonstration of Technique
Immediate Plans
March 1, 2006
University of Notre Dame
15
Introduction to Plant Studies with
Radioisotopes
•
14C
used in mid-1940’s
– Long half-life (~5730 years)
– Weak beta emitter
– Tracer measured by destructive harvesting
• Use of 11C for in vivo studies demonstrated in 1963
• 1973 – More and Troughton at the Department of
Scientific and Industrial Research in New Zealand showed
that useful amounts of 11C can be produced using small
van de Graaf accelerators
– Labs in USA, Canada, Scotland, New Zealand, and Germany
start using 11C for mechanistic studies of photosynthate
transport in the mid 1970’s
– Present studies at: Julich, Germany; Univ. Tokyo; BNL;
TUNL-Duke
March 1, 2006
University of Notre Dame
16
Features of using short-lived
radioisotope tracers
• Advantages
• Considerations
– In vivo measurement
– Use same specimen for
numerous experiments
– Conducive to studies
of dynamic phenomena
– Much greater
sensitivity than that of
carbon-14
March 1, 2006
– Experiments must be
performed near
accelerator
– Only observe shortterm phenomena
– For imaging,
sophisticated data
acquisition and data
analysis required
University of Notre Dame
17
Planned Research at the
TUNL-Phytotron Facility
1.
2.
3.
4.
Studies of CO2 uptake and carbon translation under different
environmental conditions
Root exudate measurements
Studies of exchange between plant roots and mycorhhiza
associations; ectomycorrhizal fungi (EMF)
Nutrient uptake and translocation under different environmental
conditions
March 1, 2006
University of Notre Dame
18
Outline
•
•
•
•
•
•
The TUNL-Phytotron Collaboration
Motivation
Status of Plant Studies with Radioisotopes
Plant Physiology Basics
Demonstration of Technique
Immediate Plans
March 1, 2006
University of Notre Dame
19
Plant Physiology 101
• Carbohydrates produced
by photosynthesis
• Sugars produced in
mature leaves and
transported via phloem
tissue
Chloroplasts
trap light
energy
H2O
Light
6H2O + 6CO2 + light  C6H12O6 + 6O2
Sugar
March 1, 2006
University of Notre Dame
CO2
Sugars
From Discover Science, Scott, Foresman,
20& Co., 1993
Plant Physiology 101
a)
b)
c)
d)
e)
Sugars loaded into a sieve tube
Loading of the phloem sets up
water potential gradient that
facilitates movement of water into
dense phloem sap from the
neighboring xylem
As hydrostatic pressure in
phloem sieve tube increases,
pressure flow begins, and sap
moves through the phloem
At the sink, incoming sugars
actively transported out of phloem
and removed as complex
carbohydrates
Loss of solute produces high
water potential in phloem, and
water passes out, returning
eventually to xylem
March 1, 2006
http://home.earthlink.net/~dayvdanls/plant_transport.html
University of Notre Dame
21
Phloem Transport Basics
Sugars from Leaf
Storage
Reproduction
• Stems
• Seeds
• Roots
Growth
• New Shoots
• Roots
March 1, 2006
University of Notre Dame
22
Outline
•
•
•
•
•
•
The TUNL-Phytotron Collaboration
Motivation
Status of Plant Studies with Radioisotopes
Plant Physiology Basics
Demonstration of Technique
Immediate Plans
March 1, 2006
University of Notre Dame
23
Carbon-11 Production
p + 14N  11C + a
+
1
2
3
4
5
1
Produce H- ions in negative ion source
2 Accelerate H- ions toward +5MV terminal
3 Strip off electrons with carbon foil (H-  p)
4
Accelerate protons away from +5MV terminal
5
Bend p in magnet and collide on 14N target
March 1, 2006
University of Notre Dame
24
Production Block Diagram
T > 600ºC
(CuO granules)
Average proton beam current = 1 mA
Total irradiation time = 20 minutes
Gas cell pressure = 100 PSIG
Desired activity = ~10 mCi
March 1, 2006
14N(p,a)11C
University of Notre Dame
25
11C
Production
 CN
 NN

14

11
N
C
11
11
C
14
14
N
 F  5 1010 min1

ln 2

 0.034 min1
t1/ 2
 N t  1
11
C
 14 N
14
A11 C (t )  11 C N 11 C (t )  N 14 N (0)14 N [1  e
March 1, 2006
University of Notre Dame
 11 t
C
]
26
11C
11C
Positrons
 11B + b+ + ne
Qb   [mN (116C )  mN (115B)  me ]c 2
6
mN (116C )c 2  m(11C )c 2  6me c 2   Bi
i 1
5
m ( B)c  m( B)c  5me c   Bi
11
N 5
2
11
2
2
i 1
Qb   [m(11C )  m(11B)  2me ]c 2
b+
Qb   0.96 MeV
March 1, 2006
University of Notre Dame
27
Development Experiments
• Study barley plants grown in ambient (350
PPM) and elevated (700 PPM) levels of
CO2
• Label plants under both conditions
• Analyze differences in carbon uptake and
translocation
March 1, 2006
University of Notre Dame
28
Single Detector Measurements
• Use detectors
collimated for
specific areas of
plant to trace carbon
allocation on a
coarse (source/sink)
scale
• Develop quantitative
flow models to
describe dynamics
March 1, 2006
University of Notre Dame
29
Single Detector Measurements
March 1, 2006
University of Notre Dame
30
Circuit Diagram
BGO Detector
Spect. Amp.
SCA
Scaler
HV
+1300V
March 1, 2006
University of Notre Dame
31
Qualitative Results
Barley
Grown@350PPM
Labeled@350PPM
March 1, 2006
Barley
Grown@700PPM
Labeled@700PPM
University of Notre Dame
Data corrected for half-life and relative detector efficiency
32
Flow Model
Leaf
Source
Shoot
Sink A
Root
Sink B
Total
Sink
Discrete observation times: tk where k = 0,1,2,.…
Yk = counts in Sink B at time tk
Uk = counts in Total Sink at time tk
Input-Output Analysis: (1) Statistical, data-based modeling
(2) No assumptions about mechanism(s) involved
Yk = - a1 Yk-1 - a2 Yk-2 - … - an Yk-n + b0 Uk + b1 Uk-1 + … + bm Uk-m
March 1, 2006
University of Notre Dame
33
Flow Model
Extract Physically Significant Quantities:
(1) Gain – fraction of input that shows up at the output
(2) Average transit time
Best Model: Yk = -a2 Yk-2 + b0 Uk + b2 Uk-2
March 1, 2006
University of Notre Dame
34
Flow Model
Shoot Export
Leaf
Source
Shoot
Sink A
Root
Sink B
Leaf
Sink A’
Shoot
Leaf Export
Sink B’
Total
Sink
Total
Sink
Root
Treat entire plant as Total Sink to probe leaf export
March 1, 2006
University of Notre Dame
35
Modeling Procedure
• Fit data with model using method of least squares
• This gives the model parameters a2, b0, and b2 and the statistical error
in these parameters
• To determine the gain and average transit time, look at the output of
the system with a unit impulse input
Uk 
1 for k=0
Yk  a2Yk 2  b0U k  b2U k 2
0 for k0
N
N
Gain  G   Yk
Avg. Trans. Tim e  t 
k 0
March 1, 2006
University of Notre Dame
k Y
k
k 0
G
36
One Example
a2
b0
b2
G
< t > (min)
Run 1
-0.78443
(0.00015)
-0.15852
(0.00006)
0.29420
(0.00007)
0.6294
(0.0006)
7.7238
(0.0007)
Run 2
-0.69610
(0.00008)
-0.23666
(0.00008)
0.43223
(0.00009)
0.6435
(0.0006)
5.1866
(0.0006)
G  0.63645 0.00997( syst.)  0.0004( stat.)
March 1, 2006
University of Notre Dame
37
Results
Best Model: Yk = -a2 Yk-2 + b0 Uk + b2 Uk-2
[CO2]
Age
(ppm) (days)
Leaf-toLeaf Export Shoot Export
Shoot Transit
Fraction
Fraction
Time (min)
Shoot-toRoot Transit
Time (min)
350
10-12
0.78 ± 0.03
0.28 ± 0.01
20.39 ± 5.02
6.78 ± 2.30
700
10-12
0.90 ± 0.03
0.64 ± 0.01
17.71 ± 1.03
6.45 ± 1.27
350
18-21
0.92 ± 0.05
0.39 ± 0.03
25.03 ± 1.34
6.48 ± 0.01
700
18-21
0.80 ± 0.03
0.52 ± 0.005
22.01 ± 8.41
15.76 ± 2.99
March 1, 2006
University of Notre Dame
38
2D Imaging
CsF detectors
-High stopping power
-High count rate
capability
●
Approximate plant as planar source
●
Build up image through a sequence of exposures
●
Enhanced spatial resolution via coincidence detection
March 1, 2006
University of Notre Dame
39
Then We Have…
Leaf
Source
More
Accurate
Flow Model
Shoot
Total
Sink
Root
Enhanced Resolution
March 1, 2006
Observe Fine Details of
Dynamic Behavior
University of Notre Dame
40
2D Imaging
March 1, 2006
University of Notre Dame
41
Coincidence Circuit
March 1, 2006
University of Notre Dame
42
Efficiency
• Some pixels “see” more of the array than others
• Account for this by simulations
March 1, 2006
University of Notre Dame
43
Efficiency
March 1, 2006
University of Notre Dame
44
Efficiency
March 1, 2006
University of Notre Dame
45
Efficiency
From Side
From Above
March 1, 2006
University of Notre Dame
46
Prototype Efficiency
W
y (cm)
x (cm)
W
March 1, 2006
x (cm)
University of Notre Dame
y (cm)
47
Spatial Probability Distributions
1
13
March 1, 2006
2
14
3
15
4
16
5
17
6
18
7
19
8
20
9
21
10
22
11
23
12
24
University of Notre Dame
48
Prototype Resolution
W
y (cm)
x (cm)
W
x (cm)
March 1, 2006
University of Notre Dame
y (cm)
49
(1) Add SPD for each
coincidence event for a
given exposure time
(2) Subtract off background
events scaled to the
exposure time
(3) Correct for relative
detection efficiency
(4) Correct for 11C half-life
each minute of exposure
y (cm)
Image Reconstruction
x (cm)
March 1, 2006
University of Notre Dame
50
(1) Add SPD for each
coincidence event for a
given exposure time
(2) Subtract off background
events scaled to the
exposure time
(3) Correct for relative
detection efficiency
(4) Correct for 11C half-life
each minute of exposure
y (cm)
Image Reconstruction
x (cm)
March 1, 2006
University of Notre Dame
51
(1) Add SPD for each
coincidence event for a
given exposure time
(2) Subtract off background
events scaled to the
exposure time
(3) Correct for relative
detection efficiency
(4) Correct for 11C half-life
each minute of exposure
y (cm)
Image Reconstruction
x (cm)
March 1, 2006
University of Notre Dame
52
(1) Add SPD for each
coincidence event for a
given exposure time
(2) Subtract off background
events scaled to the
exposure time
(3) Correct for relative
detection efficiency
(4) Correct for 11C half-life
each minute of exposure
y (cm)
Image Reconstruction
x (cm)
March 1, 2006
University of Notre Dame
53
y (cm)
For Example
March 1, 2006
x (cm)
University of Notre Dame
54
Immediate Plans
•
•
•
•
Install radioactive handling system
Develop root exudate experiment
Build high-resolution 2D PET imager
Start full research program
March 1, 2006
University of Notre Dame
55
Radioisotope Production
1. 11CO2 (half life = 20 min.)
+ p  11C + a
Target: gas
+ p  18F + n
Target: 18O enriched water
14N
2.
13NO 3
18O
(half live = 10 min.)
+ p  13N + a
Target: 18O depleted water
16O
March 1, 2006
3. 18F- (half life = 109 min.)
4. H218O (half life = 2 min.)
+ p  15O + d
Target: water
16O
University of Notre Dame
56
14N(p,a)11C
March 1, 2006
Cross Section
EpDame
(MeV)
University of Notre
57
Radioactive Materials Handling System
March 1, 2006
University of Notre Dame
58
Root Exudate Experiment
March 1, 2006
University of Notre Dame
59
High resolution 2D imagers
5 cm x 5 cm x 1.5 cm
2mm x 2mm pixels (0.1 mm gap)
20 cm x 30 cm field of view
March 1, 2006
University of Notre Dame
60