Per Fahlberg

Applied course in Molecular biology

Botany 15 hec

Spring 2010

Department of Plant and Environmental Sciences University of Gothenburg

Examiner: Mats Ellerström

Department of Plant and Environmental Sciences

University of Gothenburg

Supervisors: Mats X Andersson and Mats Ellerström

Department of Plant and Environmental Sciences

University of Gothenburg

The search for the missing link

Investigation of the levels of arabidopsides in Arabidopsis thaliana

i

Abstract Plants have physical barriers and many different secondary metabolites as protection against

pathogens. A build-up of chloroplastic substances called arabidopsides has been found as a

response to wounding and during the hypersensitive response in Arabidopsis thaliana. It is

still not fully understood what functions arabidopsides may have in wounding and defence

responses, but they have been shown to inhibit the growth of both bacterial and fungal

pathogens in vitro as well as to be processed to more potent signalling molecules. We have

determined that high levels of arabidopsides A and B are detected shortly after freezing (5

min) in Col-0, but quickly decreased while arabidopsides E and G increased at the same rate.

Thus, freezing can be used as high throughput method to determine the accumulation of

arabidopsides after wounding. We confirmed earlier studies that have shown that there exists

a large variation in the levels of arabidopsides between the A. thaliana ecotypes Col-0 and

C24. Arabidopside levels were measured in F2 plants obtained from a cross between Col-0

and C24 plants and compared to the parental ecotypes Col-0 and C24. F2-plants with the

highest and lowest levels of arabidopsides, resembling the expression in Col-0 and C24

respectively, were used for mapping the gene responsible. During the course of this

investigation we have narrowed the area of interest down to a section on chromosome four,

containing approximately 648 genes, whereof 86 were predicted (using ChloroP 1.1) to code

proteins with a chloroplastic transfer peptide. Interestingly the lipoxygenase 2 gene earlier

reported to be responsible for the accumulation of arabidopsides is not present in the interval.

Thus we have identified a novel locus crucial for arabidopside accumulation.

Sammanfattning Växter har fysiska barriärer och många olika sekundära metaboliter som skydd mot patogener.

Ackumulering av kloroplastiska substanser kända som arabidopsider har visats som respons

på skada och under den hypersensitiva responsen hos Arabidopsis thaliana. Det är ännu oklart

vilka funktioner arabidopsider kan ha vid skada och försvarsresponser, men de har visats

kunna inhibera tillväxten hos både bakteriella och fungala patogener in vitro och de kan också

processeras till mer potenta signalmolekyler. Vi har konstaterat att höga nivåer av

arabidopsiderna A och B återfinns kort efter frysning (5 min) i Col-0, men snabbt minskar

medan arabidopsiderna E och G ökar i samma hastighet. Frysning kan alltså användas som en

metod med hög kapacitet för att bestämma ackumuleringen av arabidopsider efter skada. Vi

bekräftade tidigare studier som har visat att det finns en stor variation i mängden

arabidopsider mellan A. thaliana ekotyperna Col-0 och C24. Arabidopsidnivåerna mättes i F2

plantor från en korsning mellan Col-0 och C24 och jämfördes med parentala ekotyperna Col-0

och C24. F2-plantor med de högsta och lägsta nivåerna arabidopsider, liknande ekotyperna

Col-0 respektive C24, användes för mappning av den ansvariga genen. Under denna

undersökning har vi minskat området av intresse till en sektion på kromosom fyra,

innehållande approximalt 648 gener, varav 86 förutsades (med ChloroP 1.1) koda proteiner

med en kloroplastisk transferpeptid. Intressant nog så är lipoxygenase 2, genen som tidigare

sagts vara ansvarig för ackumulering av arabidopsider, inte med i detta intervall. Vi har alltså

identifierat ett nytt lokus som är avgörande för ackumulering av arabidopsider.

ii

Index Introduction ................................................................................................................................ 1 Materials and methods ............................................................................................................... 2

Plant material ...................................................................................................................................... 2

Lipid extraction of arabidopsides and Acyl-MGDG ............................................................................. 2

Instrumentation and software ............................................................................................................ 2

Gel electrophoresis ............................................................................................................................. 2

Results ........................................................................................................................................ 3 Ecotypes .............................................................................................................................................. 3

Age-series ............................................................................................................................................ 3

Time-series .......................................................................................................................................... 4

F2-plants .............................................................................................................................................. 5

Discussion .................................................................................................................................. 7

Summary .................................................................................................................................... 7 Acknowledgments ...................................................................................................................... 8 References .................................................................................................................................. 8

1

Introduction Plants, like all other living organisms, are constantly under pressure for survival and during

the course of evolution plants have acquired many specializations and adaptations to survive.

Competition with other plants, herbivores and pathogens that can cause physical damage and

potentially lead to death are examples of stressors that plants need to overcome.

Plants in contrast to humans do not have an acquired immune system, but they both have

an innate immune system. The innate immune system uses both preformed as well as

inducible defences and responses, such as physical barriers and many different secondary

metabolites for protection against pathogens [1].

Pathogens like bacteria and fungi must cause damage to the cell to infect and often use

different substances to facilitate infection. Plant cells can sense such intruding pathogens by

detecting the damages from infection (pieces of cell material) or recognition of non-self

molecules, known as pathogen associated molecular patterns (PAMPs; like parts of fungi

membrane or bacterial flagella) [2,3]. Plants have numerous receptors to sense PAMPs, and

their recognition leads to activation of defence mechanisms [2,4].

Recognition of pathogens generally leads to a build-up of different secondary metabolites

and hormones. The defence response in plant tissue may ultimately lead to the affected and

neighbouring cells undergoing the hypersensitive response (HR), a type of programmed cell-

death to stop further pathogenic infection and spreading [5].

Accumulation of substances named arabidopsides have been detected at very high levels

during HR in the model plant Arabidopsis thaliana. Different arabidopsides (named by

alphabetical letters in succession, currently A to G are known) are made when the oxylipins

12-oxo-phytodienoic acid (OPDA) and dinor-oxo-phytodienoic acid (dn-OPDA) are esterified

to the galactolipids mono- or digalactosyldiacylglycerol (MGDG/DGDG) in different

combinations [6].

Whereas the exact function of arabidopsides is not known, they inhibit the growth of

bacterial and fungal pathogens in vitro, indicating a protective role. When arabidopsides are

broken down there is an increase in free OPDA/dn-OPDA and jasmonic acid (JA), indicating

an additional function as signal deposit, prolonging the signalling [6,7].

There are many different ecotypes of A. thaliana, all accumulating different levels of

arabidopsides as response to wounding or pathogen recognition. Until recently it was not

known what genes are responsible for this accumulation. In 2009 it was reported that

lipoxygenase 2 (LOX2) has connection to the arabidopside response, since the levels of

arabidopsides increase after wounding in wild type (wt) A. thaliana, but not in LOX2 mutants

[8].

The aim of this work was to set up a system for high throughput analysis of arabidopsides

as well as to investigate any age dependency of arabidopside formation. Moreover a number

of A. thaliana ecotypes were rescreened to determine the natural variation in arabidopside

formation. For the two selected ecotypes, Col-0 and C24, there exists a large variation in the

accumulation of arabidopsides as well as a related molecule, acyl-MGDG. Acyl-MGDG was

described by Heinz more than forty years ago [9,10] and is made when an acyl group from a

galactolipid is transferred to a MGDG molecule.

F2 plants from a cross between Col-0 and C24 were used in this study to by genetic

mapping determine the genes responsible arabidopside accumulation.

2

Materials and methods

Plant material

The wt Col-0 and C24 plants were grown together with F2-plants in growth chamber under

short-day conditions (8 h light, 16 h darkness, approximately 200 µE light intensity, 22 °C

day temperature and 19 °C night temperature). Samples were cut-out using a cork borer (6

mm diameter) that collected leaf-discs with roughly the same weight (5 mg). Ten leaf-discs

were collected from each plant from randomly selected leafs, and placed in double-distilled

water (ddH2O) in six-well plates.

Lipid extraction of arabidopsides and Acyl-MGDG

Five discs were collected from each well, dried with paper-tissue, put in test-tubes and boiled

in 2-propanol for 5 min (105 °C). The 2-propanol were then dried out under nitrogen on

heating-block (all drying were done at 35 °C) and chloroform, methanol, ddH2O and

butylated hydroxytoluene (2 mL; 1:2:0.8 vol., 0.05% BHT) were added to the tubes, and left

in 10 °C for 40 min. The remaining five leaf-discs from each well were collected and treated

the same way as previously described, but were frozen in liquid nitrogen and thawed at room

temperature (RT) for one hour before boiling in 2-propanol. The tubes were then sonicated

(approximately 45 min, until the leaf discs were de-pigmented) where after phase separation

was induced by adding 500 µL potassium sulphate (0.38 M) and 500 µL chloroform with

BHT (0.05% BHT). After separation, the lower phase from each tube were transferred to new

tubes, 1 mL chloroform with BHT (0.05% BHT) were added to the old tubes, vortexed,

centrifuged (5 min; 2000 rpm) and the lower phases were then collected and pooled with

those previously collected. The pooled phases were then dried under nitrogen on heating-

block and resuspended in 100 µL chloroform. Silica columns (Discovery® DSC-Si SPE,

52654-U, 500 mg) were primed with 2.5 mL chloroform. The samples were then added to the

columns, washed with chloroform (2 x 2.5 mL), moved to new tubes and eluted with a

mixture of methanol and acetone (2 x 2.5 mL; 10% methanol in acetone) to collect the lipid

fraction. The fractions were then dried out under nitrogen on heating-block, resuspended in 35

µL acetonitrile, vortexed and transferred to vials for the high performance liquid

chromatography (HPLC).

Instrumentation and software

The HPLC (HP Series 1050) were used together with an Evaporative Light Scattering

Detector (ELSD; Sedex 45, connected by HP interface 35900E). The instruments were

controlled by the HP ChemStations for LC software, also used for data analysis. The HPLC

were run with gradients of acetonitrile:ddH2O (85:15; solvent A) and 2-propanol (solvent B)

in a pre-programmed method (100% solvent A from 0 to 5 min, 20% solvent A from 45 to 50

min, 100% solvent A from 55 to 70 min). The column used for HPLC were Prevail C18 (3

µm, 150 x 2.1 mm with guard column), with a flow at 0.2 mL/min. Detection with ELSD at

40 °C, 2.3 bar pressure and gain 7. Arabidopsides were detected by UV absorbance at 220

nm.

Gel electrophoresis

Agarose gels (SeaKem® LE Agarose) and TAE buffer (from 1L 50x stock: 242 tris base, 57.1

mL glacial acetic acid, 100 mL 0.5M EDTA pH 8.0) where used as running buffer for

electrophoresis.

3

Results

Ecotypes

The amounts of arabidopsides were measured in five weeks old plants of twelve ecotypes,

both before and after freeze-thawing, and were found to be lower than expected. The

expression of LOX2 has been shown to increase with age/maturation, and the low values were

believed to be connected to the age of the plants. The same plants were re-sampled two weeks

later and the amounts of arabidopsides were found to be higher (Figure 1). The C24 ecotype

had significantly lower amount of arabidopsides compared to the other ecotypes at seven

weeks of age, making C24 a good candidate to cross with Col-0 for genetic mapping. An age-

series experiment were started to see how the levels of arabidopsides changed with age in

Col-0 and C24.

Figure 1: The sum of arabidopsides E and G were higher in seven week old plants (7w) compared to the five week old plants (5w).The levels of arabidopsides A and B did not affect the results (not shown).

Age-series

Plants were collected once a week at four to seven weeks of age (discarded after sampling).

The amount of arabidopsides was higher in sampled plants that were somewhere between six

and seven weeks of age compared to younger plants (Figure 2). This gave a reference age at

seven weeks for sampling F2 plants, to ensure that the collected material would be easily

distinguishable in arabidopside levels (either high like Col-0 or low like C24). The amounts of

acyl-MGDG decreased with increased age in both ecotypes (Figure 3). A similar experiment

was conducted using the pathogenic bacteria Pseudomonas syringae (DC3000:AvrRpt2). The

results were similar to freeze-thawing, but the amounts of arabidopsides in Col-0 peaked

between four and five weeks instead of six and seven weeks of age (results not shown).

0

200

400

600

800

1000

1200

Col-0 Ws-0 C24 Ler Mt-0 Ts-U Cvi-0 Per-1 For-2 No-0 Co-4 Gr-1

nm

ol a

rab

ido

psi

de

s /

g FW

Arabidopsides E+G

Control 5w

Frozen 5w

Control 7w

Frozen 7w

4

Figure 2: Levels of arabidopsides (A-G means total sum of A, B, E and G) after freeze-thawing increased during maturation and peaked in plants between six and seven weeks of age.

Figure 3: The amounts of acyl-MGDG after freeze-thawing decreased in both ecotypes during maturation.

Time-series

When measuring the amount of arabidopsides one hour after freeze-thawing, there were

almost exclusively arabidopside E and G, and very little A and B. To see how the

accumulation of arabidopsides changes with time an experiment was done with different

thawing-times before extracting lipids. Arabidopsides A and B were high after five minutes of

thawing, whereas arabidopsides E and G were low. As the amounts of A and B decreased in

the following time-points, E and G increased at the same rate (Figure 4).

0

100

200

300

400

500

600

700

Col-0 C24

nm

ol a

rab

ido

psi

des

/ g

FW

Arabidopsides A-G after freeze-thawing

5w Control

5w Frozen

6w Control

6w Frozen

7w Control

7w Frozen

8w Control

8w Frozen

0

100

200

300

400

500

600

700

5 6 7 8

µg

acyl

-MG

DG

/ g

FW

Weeks

Acyl-MGDG

Col-0

C24

5

Figure 4: Levels of arabidopsides after freezing and 0 to 120 minutes thawing in Col-0. Arabidopsides A and B decreased while E and G increased at the same rate.

F2-plants

Sampling of F2-plants was done when they were seven weeks of age. Plants with the highest

(most like Col-0) and with the lowest (most like C24) amounts of arabidopsides were selected

for DNA mapping (see example of data in Figure 5). A direct connection between

lipoxygenase 2 (LOX2) and the levels of arabidopsides was suspected, but when using a

LOX2-marker (constructed by Anders Nilsson, unpublished results) with the F2-plants, the

results showed that LOX2 is not directly responsible for the accumulation of arabidopsides

(data not shown). Numerous markers were tested and the best match was on chromosome

four. Further mapping narrowed the area of interest down to between the markers GOT6 and

GOT9 (constructed by Anders Nilsson; Figure 6. Also see Appendix I, Table A1). This

section contain approximately 648 genes (data from TAIR), whereof 86 were predicted to

contain a chloroplast transfer peptide using ChloroP 1.1 [11]. All F2-plants resembled Col-0

more than C24 (in coloration, leaf thickness, hairs etc.) and no connection between visible

phenotype and levels of arabidopsides were found.

0

200

400

600

800

1000

1200

0 50 100 150

nm

ol a

rab

ido

psi

de

s /

g FW

min (thawed)

Arabidopsides A+B and E+G after freeze-thawing

Mean A+B

Mean E+G

6

Figure 5: Levels of arabidopsides (A-G means total sum of A, B, E and G) in 44 F2-plants and parentals (Col-0 and C24) after freeze-thawing, sorted on arabidopside levels.

Figure 6: Mapping of genes responsible for arabidopside accumulation on chromosome four. The area of interest were found to be between the GOT6 and GOT7 markers.

0

100

200

300

400

500

600

72

23

22

82

41

62

34

01

73

41

2C

24

27

41

18

31

13 3

37

29 5 6

35

38 2 8

21

33

19

39

20

42

25

30

36

15

26

14

43 4 1

10 9

11

Co

l-0

44

nm

ol a

rab

ido

psi

des

/ g

FW

Arabidopsides A-G

A-G

7

Discussion The genes responsible for arabidopside accumulation were found to be between the

markers GOT6 and GOT9 on chromosome four. Previous studies by Staal et al. (2006) have

identified areas on chromosome four responsible for resistance against the fungal pathogen

Leptosphaeria maculans [12]. Arabidopside accumulation have been thought to have

connection to the LOX2 gene on chromosome two, supported by the findings of Glauser et al.

(2009) [8], but the results from this study show that the LOX2 gene is not responsible for

arabidopside accumulation. No previous studies show any connection between genes on

chromosome four and arabidopside accumulation, thus we have identified a novel locus

crucial for arabidopside accumulation.

The result from the time-series experiment indicates that arabidopside A and B may be

converted to E and G within 30-60 minutes after freezing. Enzymatic transfer of OPDA to

arabidopside A molecules to form arabidopsode E were suggested by Andersson et al. (2006)

[6], and the fact that arabidopside E and G increases at approximately the same rate as A and

B decreases supports that theory. These results also show that freezing can be used as a high

throughput method for determining accumulation of arabidopsides after wounding.

The age-series clearly show that the age of sampled plants must be taken into consideration

before conducting experiments or comparing results from different studies. The initial results

from testing five weeks old plants from twelve different ecotypes showed little difference in

arabidopside levels between the different ecotypes; the difference between Col-0 and C24

were too low to be useful for mapping. All ecotypes except C24 had much higher

arabidopside levels two weeks later. This confirms earlier studies that have shown large

variations in the levels of arabidopsides between the Col-0 and C24 ecotypes.

Only a few F2-plants had lower amounts of arabidopsides than C24 controls. This result

indicates that that low levels of arabidopsides in C24 could be a recessive trait, and that there

are probably more than one gene responsible for arabidopside formation. The majority of the

plants with low levels of arabidopsides used for mapping were slightly higher in arabidopside

levels than the C24 control plants, explaining why some of the plants sorted as “low” in

arabidopsides were found to be heterozygote (Appendix I, Table A1).

The age of the plants in weeks is not the best way of determining the maturation level since

small differences in environment can give large changes in growth and maturation over time.

In this study all experiments were done on plants grown together with control plants used to

get the best results. This way the maturation of the F2-plants would be as similar to the

control plants as possible.

Summary The amounts of arabidopside A and B is high shortly after damage, but decreases quickly

as arabidopside E and G increases. The amount of arabidopsides is also dependent on the

age/maturation of the plants.

The gene/s responsible for the increased levels of arabidopsides after wounding are likely

positioned on chromosome four, somewhere between the GOT6 and GOT9 markers.

Interestingly the lipoxygenase 2 gene earlier reported to be responsible for the accumulation

of arabidopsides is not present in the interval. Thus we have identified a novel locus crucial

for arabidopside accumulation.

8

Acknowledgments I want to thank my supervisors Mats Ellerström and Mats X. Andersson for giving me the

opportunity to be a part of their research and critical reading of this report. Thanks also for the

expert technical expertise of Mats X. Andersson and Anders Nilsson. Thanks to Nathalie

Buhot and Oskar Johansson for their help and nice conversations.

References 1. Menezes H, Jared C (2002) Immunity in plants and animals: common ends through

different means using similar tools. Comp Biochem Physiol C Toxicol Pharmacol

132: 1-7.

2. Hammond-Kosack KE, Parker JE (2003) Deciphering plant-pathogen

communication: fresh perspectives for molecular resistance breeding. Curr Opin

Biotechnol 14: 177-193.

3. Andersson MX, Kourtchenko O, Dangl JL, Mackey D, Ellerstrom M (2006)

Phospholipase-dependent signalling during the AvrRpm1- and AvrRpt2-induced

disease resistance responses in Arabidopsis thaliana. Plant J 47: 947-959.

4. Nurnberger T, Brunner F, Kemmerling B, Piater L (2004) Innate immunity in plants

and animals: striking similarities and obvious differences. Immunol Rev 198:

249-266.

5. Al-Daoude A, de Torres Zabala M, Ko JH, Grant M (2005) RIN13 is a positive

regulator of the plant disease resistance protein RPM1. Plant Cell 17: 1016-1028.

6. Andersson MX, Hamberg M, Kourtchenko O, Brunnstrom A, McPhail KL, et al.

(2006) Oxylipin profiling of the hypersensitive response in Arabidopsis thaliana.

Formation of a novel oxo-phytodienoic acid-containing galactolipid, arabidopside

E. J Biol Chem 281: 31528-31537.

7. Kourtchenko O, Andersson MX, Hamberg M, Brunnstrom A, Gobel C, et al. (2007)

Oxo-phytodienoic acid-containing galactolipids in Arabidopsis: jasmonate

signaling dependence. Plant Physiol 145: 1658-1669.

8. Glauser G, Dubugnon L, Mousavi SA, Rudaz S, Wolfender JL, et al. (2009) Velocity

estimates for signal propagation leading to systemic jasmonic acid accumulation

in wounded Arabidopsis. J Biol Chem 284: 34506-34513.

9. Heinz E, Tulloch AP (1969) Reinvestigation of the structure of acyl galactosyl

diglyceride from spinach leaves. Hoppe Seylers Z Physiol Chem 350: 493-498.

10. Heinz E (1967) [On the enzymatic formation of acylgalactosyldiglyceride]. Biochim

Biophys Acta 144: 333-343.

11. Emanuelsson O, Nielsen H, Heijne Gv. ChloroP 1.1 Prediction Server. Retreived 15

Mar 2007, from http://www.cbs.dtu.dk/services/ChloroP/

12. Staal J, Kaliff M, Bohman S, Dixelius C (2006) Transgressive segregation reveals two

Arabidopsis TIR-NB-LRR resistance genes effective against Leptosphaeria

maculans, causal agent of blackleg disease. Plant J 46: 218-230.

Appendix I

Table A1: Genetic mapping with different markers. The area of interest is between the GOT6 and GOT7 markers.

Arabidopside A-G (nmol/g FW) 434 283 215 351 850 800 567 325 70 700 620 600 422 739 240

Sample 10 6B 19B 3B 19A 15A 25 7B 9B 7A 13A 11A 17 1B 7

Chrom 4 – High levels of arabidopsides

JV30/31* Col Het Het Het Col C24 Col C24 C24

MN4.2 Col Col Col Col Col Het Het Het C24 C24 Het Col Col Col

GOT3 Col Col Col

Het Het

Col Col Col

NGA8* Col

Het

Col Col

GOT4 Col Col Col Col Col Het Het Het Het C24 Het Col Col Col

GOT6 Col Col Col Col Col Col Col Col Col Col Col C24 Col Col

GOT9 Col Col Col Col Col Col Col Col Col Col Col Col Col Col

GOT7 Col Col C ol Col Col Col Col Col Col Col Col Col Het Col

GOT5 Col Col Col Col Col Col Col Col Col Col ? Col Het Het

GOT8 Col Col Col Col Col Col Col Col Col Col Col Col C24 Het

Arabidopside A-G (nmol/g FW) 23 22 36 24 49 17 33 60 49 29 120 26 22 130 100

Sample 14 2B 15 20 24 16B 26 5A 22 4B 17A 14B 11B 12A 2A

Chrom 4 – Low levels of arabidopsides

JV30/31* C24 Het Het C24 Col C24 C24 Het Het

MN4.2 C24 C24 C24 C24 Col Het Het Het C24 Het C24 Col Het Het

GOT3 C24 C24 C24 C24 Het Het Col C24

C24 Col NGA8* C24 C24 C24 C24 Col Het

GOT4 C24 C24 C24 C24 C24 Het Het Het C24 Het C24 Het Het Het

GOT6 C24 C24 C24 C24 C24 C24 C24 C24 Het Het C24 Het Het Het

GOT9 C24 C24 C24 C24 C24 C24 C24 C24 C24 C24 C24 Het ? Het

GOT7 C24 C24 C24 Het C24 C24 C24 Het C24 C24 Het Het Het Col

GOT5 C24 C24 C24 C24 C24 C24 Het

C24 Het Het Col

GOT8 C24 C24 C24 Het C24 C24 C24 Het Col C24 Het Het Het Col