Lipidomics based analysis in magnaporthe oryzae

oryzae and Rice Blast Disease 1.3 Important Signaling Pathways of M.. oryzae Research 1.6 Proposed Model for Turgor Pressure Production 1.8 Aims of the Project... oryzae; 3 employed the

LIPIDOMICS-BASED ANALYSIS IN MAGNAPORTHE ORYZAE

1.1 Challenges in Food Supply

1.2 Life Cycle of M oryzae and Rice Blast Disease

1.3 Important Signaling Pathways of M oryzae

1.4 Lipids and Their Metabolism

1.5 Recent Advances of M oryzae Research

1.6 Proposed Model for Turgor Pressure Production

1.8 Aims of the Project

CHAPTER 2: MATERIALS AND METHODS

2.6 Lipid Profiling and Quantification by HPLC/MS

2.7 Analysis of Phosphatidylinositol Phosphates

2.8 Inhibitor Design by Bioinformatic Approaches

2.8.1 Homologous Modeling of Tps1’s Structure

2.8.2 Screening of MLSMR and Docking Results Analysis 2.8.3 Lead Optimization

2.8.4 Molecular Dynamics Simulation by Gromacs

CHAPTER 3: GENERAL LIPIDOMIC ANALYSIS

3.1 Lipid Body Staining

3.3 Semi-quantitative Analysis of the Lipidome

CHAPTER 4: BETA OXIDATION AND PATHOGENICITY

4.1 Phospholipids and TAGs in Mutants and WT

4.2 Conclusion

CHAPTER 5: IN SILICO INHIBITOR DESIGN

5.1 Structure Modeling by Modeller

5.2 Screening of MLSMR for Inhibitors

5.3 Lead Optimization

5.4 Molecular Dynamics by Gromacs

5.5 Conclusion

CHAPTER 6: DISCUSSION

6.1 Lipid Body Staining

6.2 Profiling and Quantification of the Lipidome

6.3 Validation of the Mechanism for Turgor Pressure Production 6.4 Beta-oxidation: Mitochondria vs Peroxisomes

6.5 Lead 25 for Blast Disease Control

CONCLUSION

REFERENCES

APPENDICES

SUMMARY

Magnaporthe oryzae (M oryzae) is the causal agent of the rice blast disease

Triacylglycerides (TAGs) were one of the major sources used to generate

turgor pressure as a means for M oryzae to penetrate into host’s leaf Lipids

therefore play a very important role in the pathogenesis However, there is up

to date no lipidomics study of M oryzae available As Part I of project, our research for the first time analyzed the lipidome of M oryzae and quantified

the lipid species across different time points along the pathological cycle The lipidomics study as a platform was further used to analyze two beta oxidation pathway mutants and proposed possible explanation for their nonpathogenicity Our data had also shown interesting information and was suggestive of a possible mechanism for turgor production

Previous studies already discovered that trehalose synthase (Tps1) was not only responsible for the production of trehalose and utilization of nitrogen source, but also the regulation of several NADPH-dependent transcriptional corepressors, namely Nmr1, Nmr2, and Nmr3, which can each bind NADP Therefore, as for Part II of the project, the structure of Tps1 was modeled for

screening of possible inhibitors in silico against a database of 400k compounds,

and molecular dynamics studies were also done for some of the best hits Advice was then given for future inhibitor design in the context of rice blast control

To summarize, this project had: 1) profiled the lipidome of M oryzae; 2) identified key lipid species for turgor generation of M oryzae; 3) employed

the lipidomics approach as the platform to study some nonpathogenic mutants; 4) proposed a possible mechanism for turgor production; 5) screened chemical

databases for possible inhibitors of a key enzyme (Tps1) involved in pathogenesis

List of Figures

Page

Figure 1 The pathogenesis cycle of M oryzae 13

Figure 2 A proposed model for turgor generation in

Figure 4 The staining the lipid droplets of M oryzae 50-53

Figure 5 The elution profiles of different lipid classes

& tabulation of lipid species identified

56-59

Figure 6 Quantification of phospholipids 62-64

Figure 8 Quantification of TAGs and DAGs 70

Figure 9 Quantification of PI-3,4-P2 and PI-4,5-P2 72

Figure 10 TAG analysis on WT, ΔEch1 and ΔFox2 77

Figure 11 Proposed function of trehalose metabolism 82

Figure 13 Structures of Tps1, 1gz5 and 2wtx aligned 85

Figure 14 The validation of Vina’s performance 88

Figure 15 The binding confirmation of the best 3

compounds when docked to Tps1

89

Figure 16 The structure, binding confirmation and

computed LogP of Lead 25

91

Supplementary

figure 1

TAG:DAG:PLs ratios over different time

points of the pathogenesis of M oryzae

DHAP dihydroxyactone phosphate

DMSO dimethyl sulfoxide

C albicans Candida albicans

PI-3-P phosphatidylinositol 3-phosphate

PI-3,4-P2 phosphatidylinositol 3,4-bisphosphate

PI-3,4,5-P3 phosphatidylinositol 3,4,5-trisphosphate

PI-4,5-P2 phosphatidylinositol 4,5-bisphosphate

S ferax Saprolegniaferax

U maydis Ustilago maydis

% w/v percentage weight by volume

% v/v percentage volume by volume

CHAPTER 1: INTRODUCTION 1.1 Challenges in Food Supply

It might be fair to say that the world today has never been able to solve the problem of the security of food supply, despite of the advancement of technology Possible causes would be climate change, increased demand for biofuels, structural shifts in food and agricultural systems, trans-boundary movement of disease and widespread land degradation (Thompson, Cohen et

al 2012, Thornton 2012) Taking China as an example, the rapid urbanization has a big and long term impact on both the public health (Van de Poel, O'donnell et al 2012)and its own environment (Zhu 2012), while the loss of farming lands would be one of the direct consequences One more example is the conversion of corn to ethanol in the United States since 2005,which was considered a major cause of global food price increases during that time (Albino, Bertrand et al 2012)

Meanwhile, crops diseases could further contribute to the insecurity of

food supply One example could be seen in the case of Phytophthora infestans (P infestans), the Oomycete agent of potato late blight and the primary cause

of the great Irish famine of the nineteenth century The disease caused significant yield losses in Ireland due to the wetness of the climate while there was a large proportion of the population who almost totally depended on potato as their food supply (Large 1940); eventually 1.5 million out of 8 million population died of starvation and another 1.5 million emigrated, when about a quarter of the emigrants died in transit (Klinkowski 1970) A recent case however would be the wheat stem rust, as it could infect 90% of the wheat strains and cause a global threat to the wheat production and threaten the food supply of billions (Singh, Hodson et al 2011)

How would the issue of food shortage be effectively handled? To answer the question, many ideas and suggestions could be given in relation to policy making and resource management; but as for scientists, research on the crops and the possible pathogens in the context of the crop diseases management should be apriority

1.2 Life Cycle of M oryzae and Rice Blast Disease

Magnaporthe oryzae (M oryzae) represents a serious threat to global rice

production due to its role as a causal agent of the rice blast disease, the most severe disease affecting cultivated rice, causing 10%-30% loss of rice production (Talbot 2003)

M oryzae reproduces both sexually (Saleh, Xu et al 2012) and asexually

Asexual spores are involved in rice blast disease; sexual reproduction, on the other hand, gives rise to perithecium, the fruiting body that carries numerous eight sporedasci under proper conditions (Valent, Farrall et al 1991)

Figure 1 (taken from Wilson and Talbot 2009) illustrates the life cycle of

M oryzae Rice blast infection is initiated when conidia attach to the

hydrophobic surface of a rice leaf by producing an adhesive material at the apex of the conidium (Hamer, Howard et al 1988) The conidia could germinate even only in the presence of water and form germ tubes within 2 hours (Talbot 2003) The germ tubes continue to extend for about 15-30 µm before swelling at their tips The tips would become flattened against the leaf surface, and differentiated into specialized dome-shaped structures called the appressoria (Bourett and Howard 1990) Some possible conditions for appressorium

differentiation are leaf surface topography and the absence of exogenous nutrients (Dean 1997) When the surface is hydrophilic, some chemicals could

be used to still induce appressorium differentiation, taking cutin and lipid monomers for example (Gilbert, Johnson et al 1996) The cell wall of an appressorium is rich in chitin, and a layer of melanin is deposited underneath the cell wall of the appressorium (Bourett and Howard, 1990) The melanin

layer enables M oryzae to withstand the physical force produced by the

appressorium when penetrating the plant cuticle (de Jong, McCormack et al 1997)

A conidium contains three nuclei in all and one in each cell During appressorium differentiation, nuclear division takes place in the germ tube: about 4 to 6 h after inoculation, one nucleus migrates to the germ tube, and this nucleus then differentiates into two daughter nuclei by a single mitotic division (Veneault-Fourrey, Barooah et al 2006) One of the two daughter nuclei would then move into the appressorium while the other returns to the conidium After 12 to 15 h, the three nuclei in the conidium begin to degrade and only the nucleus in the appressorium remains (Veneault-Fourrey, Barooah

et al 2006) The nutrient storage is also mobilized from the original conidium into the appressorium as soon as mitotic division takes place After all these, a specialized septum would be developed and separate the appressorium and the collapsing conidium Autophagic cell death of the conidium would then follow and the appressorium is left intact on the plant leaf (Veneault-Fourrey, Barooah et al 2006, Kershaw and Talbot 2009)

After the leaf cuticle is ruptured, M oryzae starts to form a penetration

peg that swells into a primary infection hyphae; the infection hyphae in turn differentiates into a series of branched and bulbous invasive hyphae (Talbot 2003) After the initial epidermal cell is colonized, invasive hyphae begins to move into adjacent cells and infect host plant tissue (Valent, Farrall et al 1991,

Talbot, Ebbole et al 1993) After 4 days, disease lesions appear on the surface

of the leaf and M oryzae releases very large numbers of conidia into the

atmosphere and restarts the whole process of pathogenesis (Wilson and Talbot 2009)

There are 2 major symptoms on rice plants by the infection of M oryzae,

namely a leaf spot disease with large ellipsoid lesions on the surface of rice leaves (Talbot 1995) and also neck blast panicle blast symptoms in older rice plants (Wilson and Talbot 2009) The key to its pathogenesis, as what has been mentioned earlier on, is the formation of the appressorium, a specialized cell that serves to facilitate the process of invasion and penetration into plant tissues during infection by generating substantial amount of turgor pressure and physical force Glycerol was reported as the most abundant solute in the appressoria (Talbot 2003) and therefore represents the main osmolite responsible for generating the high turgor pressure (Thines, Weber et al 2000) The high turgor pressure, which could be up to 8 mPa (Howard, Ferrari et al 1991), enables the penetration peg to mechanically penetrate through the leaf surface of the host and thus allows the delivery of the fungal materials into the host cell (Howard, Ferrari et al 1991) As appressorium formation does not involve an exogenous supply of nutrients, the source of glycerol should be found inherently within the conidia, which consists of substantial amounts of lipids, glycogen and other storage products (Talbot 2003)

1.3 Important Signaling Pathways of M oryzae

Three signaling pathways have been shown to play an important role in

plant infection by M oryzae: the cyclic 3', 5' adenosine monophosphate

(cAMP) signaling pathway (Lee and Dean 1993, Xu and Hamer 1996, Umemura, Ogawa et al 2000), the Pmk1 mitogen-activated protein kinase

(MAPK) signaling pathway (Xu and Hamer 1996, Bruno, Tenjo et al 2004) and the Mps1 MAPK signaling pathway (Xu, Staiger et al 1998)

M oryzae uses the cAMP signaling pathway for surface recognition and

triggering of appressorium formation (Lee and Dean 1993, Umemura, Ogawa

et al 2000) Appressorium formation was restored in the presence of exogenous cAMP for a Δmac1 mutant (Choi and Dean 1997) However, there are evidences that cAMP signaling pathway has broad and divergent impact on growth and pathogenesis (Adachi and Hamer 1998)

In M Oryzae, there are three MAPK pathways characterized: the Pmk1

MAPK pathway which is involved in appressorium formation and penetration, the Mps1 MAPK pathway which is involved in conidiation and penetration, and the Osm1 MAPK pathway required for osmoregulation; however, only the first two genes are affecting the pathogenicity (Xu and Hamer 1996, Xu, Staiger et al 1998, Dixon, Xu et al 1999) It was also proposed that PMK1 acts downstream of a cAMP signal for appressorium formation since pmk1 mutants develop abnormal germ tubes in the presence of cAMP on hydrophilic

surfaces (Xu and Hamer 1996) Apart from M Oryzae, PMK1 homologues in appressorium-forming fungi, including M oryzae, Colletotrichum lagenarium (C lagenarium), Cochliobolus heterostrophus (C heterostrophus), and Pyrenophorateres, were found to be essential for appressorium formation and

plant infection (Lev, Sharon et al 1999, Takano, Kikuchi et al 2000,

Ruiz-Roldán, Maier et al 2001) For other fungi like Fusarium oxysporum (F oxysporum) and Botryti scinerea (B cinerea), PMK1 homologues are still

required for their pathogenicity (Zheng, Campbell et al 2000, Di Pietro, García‐Maceira et al 2004) It might be fair to say that for many fungi the PMK1 pathway is conserved for appressorium formation and other plant infection processes However, although PMK1 homologues have been identified in several fungi, it is still an on-going research on the activation of PMK1 and its

downstream effectors, and one example could be the work on MST12 (Park, Xue et al 2002) Meanwhile, distortion of the MPS1 MAPK pathway resulted

in the failure of the remodeling of the appressorium wall and the consequential inability of appressoria to penetrate plant cell surfaces (Xu, Staiger et al 1998)

MPS1 homolog in the maize pathogen C heterostrophus was also shown to be involved in melanin synthesis (Eliahu, Igbaria et al 2007) In the case of C lagenarium, knocking out MPS1 would even prevent the fungus from forming

appressoria (Kojima, Kikuchi et al 2002)

1.4 Lipids and Their Metabolism

Lipids are key components of cell membranes and actively involved in a range of biological functions: i.e energy storage, as precursors for hormone synthesis and also signaling (Sul and Wang 1998, Bozza, Melo et al 2007) Lipid biosynthesis occurs in the cytoplasm and utilizes acetyl-CoA as the precursor through a process known as lipogenesis (Kersten 2001) Because of the shared source of acetyl-CoA, lipid metabolism is tightly connected to carbohydrate metabolism It is even believed that proper lipid metabolism is a key molecular integrator of energy homeostasis, membrane structure and dynamics, and signaling (Wenk 2010)

Phospholipids are major components of cellular membranes that participate in a range of cellular processes Their structures exhibit a high degree of het- erogeneity: saturated fatty acyl groups are predominated in the sn-1 position, whereas unsaturated fatty acids are commonly found at the sn-2 position Fatty acid remodeling is believed to have contributed to the

incorporation of unsaturated fatty acids at the sn-2 position Our understanding

of the phospholipids’ roles and functions has been broadened by recent discoveries as well In the following sessions, they are briefly discussed based

on the classes

1.4.1.1 Phosphatidic acids (PAs)

PA is the simplest diacyl-glycerophospholipid and occurs only in small amounts (often less than a few mol%) in biological membranes but yet is crucial for cell survival, because it is a key intermediate in the biosynthesis of phospholipids and is involved in many signaling events

The synthesis is by two major de novo biosynthetic pathways that utilize either glycerol 3-phosphate (G-3-P) or dihydroxyacetone phosphate (DHAP)

as precursors (Carman and Zeimetz 1996) In the case of G-3-P, it is acylated

by G-3-P acyl- transferase at the sn-1 position to produce lysophosphatidic acid (LPA) DHAP is acylated at the sn-1 position by DHAP acyl- transferase

to produce 1-acyl-DHAP, which is reduced by 1-acyl-DHAP reductase to form LPA At this stage, LPA is further acylated by LPA acyltransferase in the sn-2 position to yield PA Phospholipids, including phosphatidylserine (PS), phosphatidylethanolamine (PE), and phosphatidylcholine (PC), are synthesized from PA through the cytidine diphosphate diacylglycerol pathway Different organelles (i.e., the ER, mitochondria, and in addition peroxisomes

of mammalian cells, lipid particles of yeast and chloroplasts of plants) could

be involved in the biosynthesis PA, and the redundancy of the biosynthetic systems could be that different pools of PA may serve as precursors for the synthesis of complex phospholipids and/or triacylglycerols (TAG) in different organelles, though the regulatory mechanisms for channeling PA to form the various species of acylglycerolipids under specific physiological conditions are still under research (Athenstaedt and Daum 1999)

PA and LPA are known to be involved in membrane fission and fusion

events (Weigert, Silletta et al 1999), and a conversion of LPA into PA may induce negative spontaneous monolayer curvature and membrane bending (Kooijman, Chupin et al 2003) In the light of that, PA has been found to facilitate the coalescence of contacting LDs in forming "supersized" lipid droplets in yeast (Fei, Shui et al 2011) Meanwhile, an important determinant

of the biological functions of PA is its anionic headgroup (Kooijman and Burger 2009), which is related to many PA-protein interactions It was also found to be involved in mTOR signal transduction and protein synthesis, being also a direct link between mTOR and mitogens (Fang, Vilella-Bach et al 2001) PA on the endoplasmic reticulum directly binds to the soluble transcriptional repressor Opi1p and makes it inactive outside the nucleus; upon the rapid consumption of PA by the addition of the lipid precursor inositol, Opi1p is released from the endoplasmic reticulum for its nuclear translocation and repression of target genes (Loewen, Gaspar et al 2004); such interaction between PA and Opi1p is known to be pH dependent and linked membrane biogenesis with nutrient availability as well (Young, Shin et al 2010)

1.4.1.2 PCs

PC is a major component of the cellular membrane, and its role could not

be replaced by PE (Wu, Ye et al 2010)

As for the synthesis, both de novo biosynthesis (the Kennedy Pathway)

and remodeling processes (the Lands cycle) take place: the saturated fatty acids found frequently at the sn-1 position of PC are believed to be derived from de novo biosynthesis, whereas the unsaturated fatty acids, usually found

at the sn-2 position in PC, are esterified mainly through the remodeling process (MacDonald and Sprecher 1991)

PC could be converted into PA, phosphocholine and DAG, which are in turn functioning as signaling molecules One example is seen in the case of the hydrolysis of PC by PC-PLC: the resulting phosphocholine and DAG in the

case of being stimulated by cytokines, growth factors, mitogens, and calcium ions have been implicated in intracellular signal transduction involved in the regulation of cell metabolism, growth, differentiation and induction of apoptosis (Szumilo and Rahden-Staron 2008) PC-specific phospholipase C has inhibitory effects on the pathways responsible for constitutive epithelial ovarian cancer cell stimulation and cell proliferation (Spadaro, Ramoni et al 2008), and that the elevated phosphocholine pool detected in epithelial ovarian cancer cells primarily results from upregulation/activation of ChoK and PC-phospholipase C involved in PC byosinthesis and degradation, respectively (Iorio, Ricci et al 2010)

1.4.1.3 PEs

In yeast, PE is synthesized by multiple pathways located in different subcellular compartments (i.e., ER, Golgi and mitochondria) whose PE products are functionally different，and one example is seen in the case of LysoPE that supports growth and replaces the mitochondrial pool of PE much more efficiently than and independently of PE derived from the Kennedy pathway (Riekhof and Voelker 2006)

The mitochondrial inner membrane contains two non-bilayer‐forming phospholipids, PE and cardiolipin, which affect the stability of respiratory chain supercomplexes differently (Böttinger, Horvath et al 2012) Mitochondrial PE deficiency impairs formation and/or membrane integration

of respiratory supercomplexes (Tasseva, Bai et al 2013) Recently, cardiolipin and mitochondrial PE are required to maintain tubular mitochondrial morphology and have overlapping functions in mitochondrial fusion (Joshi, Thompson et al 2012) PE has other important roles besides being a membrane component, and very often it is associated with other proteins Psd2,

a PS decarboxylase, is responsible for the synthesis of vacuolar membrane PE; the loss of the enzyme causes a specific reduction of the vacuolar membrane

PE but not the total PE levels and the subsequent loss of normal activity of a vacuolar ATP-binding cassette transporter protein called Ycf1: the mutant yeast strain is sensitive to cadmium (Gulshan, Shahi et al 2010) PE is involved in autophagy: the formation of Apg8-phosphatidylethanolamine has

an essential role in membrane dynamics during autophagy (Ichimura, Kirisako

et al 2000) PE plays an important role in the regulation of APP proteolysis and thus Aβ generation too (Nesic, Guix et al 2012)

1.4.1.4 PSs

Being a metabolically related metabolite of PE, PS is present in membranes of all eukaryotic and prokaryotic cells Depending on the type of organisms, different pathways would be used for synthesizing PS: as what is reviewed, in prokaryotes and the yeast, all PS is synthesized by a PS synthase that uses CDP-diacylglycerol and L-serine, while in mammalian cells a calcium-dependent base-exchange reactions in which the polar head-group (choline or ethanolamine) of a pre-existing phospholipid (PC or PE, respectively) is exchanged for L-serine is used; one more thing to note of is that mammalian mitochondria do not make PS whereas bacteria do (Vance and Tasseva 2013)

PS is not symmetrically distributed across the two leaflets of the membrane bilayer and studies on the erythrocyte membrane indicate that more than 96% of PS resides on the inner leaflet of the bilayer (Zachowski 1993) However, during the blood-clotting cascade, the trans-bilayer asymmetry of

PS in the plasma membrane of activated platelets is markedly altered so that

PS becomes exposed on the cell surface and the clotting factors would be recruited to the surface of platelets (Williamson, Bevers et al 1995, Majumder, Quinn-Allen et al 2008)

Due to the anionic nature of PS, the positively-charged motifs of some key signaling proteins, such as the tyrosine kinase Src, as well as the Ras and

Rho family of GTPases will bind to PS for their membrane targeting and activation (Sigal, Zhou et al 1994, Finkielstein, Overduin et al 2006, Lemmon 2008, Yeung, Heit et al 2009) The catalytic activity of several key signaling proteins such as such as synaptotagmin, dynamin-1 (Yeung, Heit et

al 2009) through the interaction between the C2 domains and PS PS could also interact with proteins containing PH domains, such as 3-phosphoinositide- dependent kinase-1 (Lucas and Cho 2011) and Akt (Huang, Akbar et al 2011)

1.4.1.5 Phosphatidylinositols (PIs) & Phosphoinositides (PIPs)

PI can be phosphorylated to form phosphatidylinositol phosphate (PI-4-P

or PIP), phosphatidylinositol bisphosphate (PIP2) and phosphatidylinositol trisphosphate (PIP3)

Sec14p is a major yeast PI transfer protein (PITP), regulating an essential interface between lipid metabolism and protein transport from Golgi membranes to the cell surface (Xie, Fang et al 1998) and even involved in the execution of developmentally regulated polarized membrane trafficking pathway (Vincent, Chua et al 2005)

PI could be metabolized into various inositol phosphates like mono- to polyphosphorylated inositols (i.e., PIP1, PIP2, PIP3…) The various PIPs play crucial roles in diverse cellular functions, such as cell growth, apoptosis, cell migration, endocytosis, and cell differentiation PI(4,5)P(2) is enriched in HIV-1, the depletion of which causes reduced HIV-1 budding (Chan, Uchil et

al 2008) The budding yeast Saccharomyces cerevisiae uses PI(3,5)P2 as its

chromatin architecture-modulating agent (Han and Emr 2011) At the same time, it is known that association with PI(3,4,5)P3 at the membrane facilitates phosphorylation and activation of Akt by PDK1 (Lawlor and Alessi 2001): subsequently, a host of other proteins could be phosphorylated and affect many cellular events Furthermore, it was demonstrated in hematopoietic cells

that functionally distinct PI(3,4,5)P3 pools exist (Bohnacker, Marone et al 2009) Recently, a specific interaction between a RhoGAP domain of Rgd1p and phosphoinositides was discovered, which was used by phosphoinositides

to specifically stimulate the RhoGAP activity of Rgd1p on Rho4p (Schlame and Otten 1991, Odaert, Prouzet-Mauleon et al 2011)

Synthesis of CL takes place at the inner mitochondrial membrane by the transfer of the phosphatidyl group of cytidinediphosphate-diacylglycerol (CDP-DAG) to PG, catalyzed by CL synthase (CLS); this conversion is so effective that only contain trace amounts of PG is found in mitochondrial membranes (Daum 1985) Tam41 is also required for such synthesis (Kutik, Rissler et al 2008) Mitochondrial fusion was found to do with the loss of CL and mitochondrial PE, which further leads to reduced levels of small and large isoforms of the fusion protein Mgm1p (Joshi, Thompson et al 2012) In yeast, lacking both CL and PE is synthetically lethal (Gohil, Thompson et al 2005) Recently, the yeast protein Taz1p was shown to function as a transacylase, catalyzing the reacylation of monolysocardiolipin to mature cardiolipin (Xu, Malhotra et al 2006).We now identified the protein encoded by reading frame YGR110W as a mitochondrial phospholipase, which deacylates de novo

synthesized CL (Beranek, Rechberger et al 2009)

Ceramide 1-phosphate (Cer-1-P) is the product of ceramide kinase and its product, and mediates calcium ionophore- and interleukin-1 β -induced arachidonic acid release, due to the activation of a species of PLA2 (Pettus, Bielawska et al 2003) Ceramide can also be broken down by ceramidases to sphingosine, which in turn is phosphorylated by sphingosine kinases to generate sphingosine 1- phosphate (S1P) (Spiegel and Milstien 2003) Activation of the S1P1 receptors stimulates downstream signals important for cell locomotion and lymphocyte recirculation and tissue homing (Hobson, Rosenfeldt et al 2001, ROSENFELDT, HOBSON et al 2001, Matloubian, Lo

et al 2004) Sphingomyelin (SM) is an abundant constituent of cellular membranes in a wide range of organisms Its high packing density and affinity for sterols help provide a rigid barrier to the extracellular environment and play a role in the formation of lipid rafts, specialised areas in cellular membranes with important functions in signal transduction and membrane trafficking (Simons and Toomre 2000, Holthuis, Pomorski et al 2001) SM synthesis is mediated by a SM synthase, which transfers the phosphorylcholine moiety from PC onto the primary hydroxyl of ceramide, thus generating SM and diacylglycerol (DAG) (Ullman and Radin 1974, Voelker and Kennedy

1982, Marggraf and Kanfer 1984)

Above, phospholipids are playing broad structural and signaling roles, and

to identify the identities and study the functions would be of fundamental and tremendous significance for any lipidomics study

Neutral lipids normally include TAGs (as well as the closely related DAGs and MAGs), steryl esters (SEs) and wax esters (WEs) They lack charged

groups and cannot integrate into bilayer membranes in substantial amounts The neutral lipids stored in lipid droplets are often used as nutrients and energy storage after enzymatic hydrolyzation; further more, the hydrolyzation products like sterols, DAG and fatty acids can also serve as building blocks for membrane formation, synthesis of steroid hormones and even energy purposes Some of the examples are briefly discussed here

1.4.2.1 DAGs

DAG has a glycerol backbone attached with two acyl chains and is often a

product of TAG hydrolysis However, de novo synthesis of DAG from glucose

hydrolysis via dihydroxyacetone phosphate and glycerol 3-phosphate is possible (Rossi, Grzeskowiak et al 1991) Hydrolysis of PI(4,5)P2 by phospholipase C (PLC) yields DAG as well Meanwhile, DAG can be formed through the breaking down of phosphatidylcholine by a phospholipase C-mediated mechanism (Besterman, Duronio

et al 1986) In mice, some acyl CoA:monoacylglycerol acyltransferase (MGAT) activity is found in its acyl-CoA:diacylglycerol acyltransferase 1 (DGAT1) as well, and this DGAT1 exhibits additional acyltransferase activities like wax monoester and wax diester synthases, and acyl CoA:retinol acyltransferase (ARAT) DAG works together with calcium to activate protein kinase C, which goes on to phosphorylate other molecules, leading to altered cellular activity DAG is also involved in the fusion and fission of membranes (Bankaitis 2002, Baron and Malhotra 2002) In diabetes, DAG plays a role in the glucose-induced activation of glomerular PKC and the subsequent glomerular hypertrophy (Craven, Davidson et al 1990) Further, DAG could be converted to PA and exert impacts through that

1.4.2.2 TAGs

TAG is the most common lipid-based energy reserve in nature The main pathway for synthesis of TAG is believed to involve three sequential acyl-transfers from acyl-CoA to a glycerol backbone (Bell and Coleman 1980,

Griffiths, Stobart et al 1988) For many years, DAGAT, which catalyzes the third acyl transfer reaction, was thought to be the only enzyme specifically involved in TAG synthesis It acts by diverting diacylglycerol (DAG) from membrane lipid synthesis into TAG (Griffiths, Stobart et al 1988) TAG synthesis can however occur in the absence of acyl-CoA: for example, the phospholipid : diacylglycerol acyltransferase (PDAT) in the yeast (Dahlqvist, Stahl et al 2000) Lipases from various organisms have been found to hydrolyze TAGs (Sobek and Gorisch 1988, Athenstaedt and Daum 2003, Gupta, Gupta et al 2004, Zimmermann, Strauss et al 2004)

The sequestration of neutral lipid like TAG in droplets provides a depot of stored energy that can be accessed in a regulated fashion according to metabolic need, and the stored TAG can also be used as substrate for synthesis

of other important cellular molecules, such as membrane phospholipids and eicosanoids For example, breaking down of TAG gives rise to glycerol and

free fatty acids (FFAs) In M oryzae, fatty acid beta oxidation has been shown

to play a role in appressorium physiology (Wang, Soanes et al 2007), and glycerol on the other hand contributes to the turgor pressure production, as what was mentioned earlier Transfer of storage carbohydrate and lipid

reserves to the appressorium occurs under the control of the PMK1 MAPK

pathway, and the subsequent appressorium turgor generation seems to be dependent on the breakdown of lipid and glycogen under the control of the

CPKA/SUM1-encoded PKA (Thines, Weber et al 2000) Because of the rapid

breaking down of TAGs and the development of membranous structures (i.e.,

germ tubes and penetration hyphae) during the pathogenesis of M oryzae,

fatty acid beta oxidation and subsequent activation of the glyoxylate cycle and gluconeogenesis could be also important players in the pathogenesis (Thines, Weber et al 2000, Weber, Wakley et al 2001) Past research showed that the

glyoxylate cycle was significant during plant infection by M oryzae and mutants lacking the ICL1 gene encoding isocitratelyase show a delay in

generation of disease symptoms (Koga, Yamauchi et al 1998) Meanwhile, the glyoxylate cycle has been shown to be required for pathogenicity of other phytopathogenic and human pathogenic fungi (Lorenz and Fink 2001, Idnurm and Howlett 2002, Solomon, Lee et al 2004), due to the shared need to

develop initially within a glucose-deficient environment But the fact thatΔicl1

mutants were still able to cause rice blast symptoms, however, was suggestive

of alternative metabolic pathways which are likely to be involved in the maturation of appressorium and development of penetration hyphae (Wang, Jenkinson et al 2005, Bhambra, Wang et al 2006, Ramos-Pamplona and Naqvi 2006)

SEs constitute a storage pool of sterols when their cellular amount exceeds the immediate requirement ， and the synthesis of SEs can be accomplished by two different enzymatic mechanisms (Oelkers and Sturley 2004) Firstly, a lecithin : cholesterol acyltransferase (LCAT) transfers a fatty acid directly from a glycerophospholipid to the hydroxyl group in position C3

of the acceptor sterol, being similar to that described for TAG synthesis in prokaryotes, yeast and plants The other route of SE formation is acyl-CoA dependent and catalyzed by an acyl-CoA: cholesterol acyltransferase (ACAT) Being spatially separated, the synthesis of SE by LCAT happens extracellularly, but ACAT esterifies sterols inside the cell The sterol moieties

of SEs are important membrane components and modulate the physical properties of a bilayer Furthermore, sterols serve as precursors for the synthesis of steroid hormones and bile acids in higher eukaryotes Thus, in all types of cell, sterol homeostasis is important

Enormous amounts of wax esters are produced and accumulated by some strains of Acinetobacter from n-alkanes or long-chain alkanols under nitrogen-limited conditions (Ishige, Tani et al 2002) For several eukaryotes,

wax esters have been found to have diverse and important biological functions like protecting living cells from desiccation, ultraviolet light and pathogens; they can be a typical energy store of plants Hydrolysis of WEs liberates fatty acids but can also generate long-chain alcohols for further metabolism (Athenstaedt and Daum 2006)

Sphingolipids are complex lipids with a long-chain amino alcohol, commonly trans-1,3-dihydroxy-2-amino-4-octadecene, sphingosine as the core structure Common examples of sphingolipids are ceramide, sphingomyelin, cerebroside, globoside and ganglioside As what has been reviewed, this group

of lipids is generally present in the outer leaflet of mammalian cell membranes; they are also found in the cell membranes of some bacterial and fungal groups, being possibly genus/species-specific (Olsen and Jantzen 2001) One example

is found in the case of P brasiliensis, the major glycosphingolipids were

identified as β -glucopyranosylceramides (Cer-Glc) having (4E, 8E)- 9-methyl-4,8-sphingadienine as long chain base in combination with either N-2’-hydroxyoctadecanoate or N-2’-hydroxy-(E)-3’-octadecenoate (Toledo, Levery et al 1999) As a step further into fungal sphingolipid metabolism, it is found out that fungi maintain two separate pools of ceramides to be used for the synthesis of different sphingolipids Ceramide backbones with C16 or C18 fatty acids linked to a 4,8-diene-9-methyl-sphingobase are exclusively precursors for GlcCer synthesis, whereas ceramide backbones with very long chain C24 and C26 fatty acids bound to phytosphinganine are restricted to the synthesis of the inositol-containing phosphosphingolipids (Toledo, Levery et

al 1999, Sakaki, Zahringer et al 2001, Toledo, Levery et al 2001, Toledo, Levery et al 2001)

Possible roles played by sphingolipids include cell signaling, heat-stress response, Ca2+ homeostasis, and transport of GPI-anchored proteins (Krönke

1999, Liu, Kleine et al 1999, Schneiter 1999, Huwiler, Kolter et al 2000) Being the central sorting station along the secretory pathway as well as the site

of synthesis of the sphingolipids, the Golgi apparatus’ capability of acting as a distillation apparatus for sphingolipids and cholesterol has been reviewed (Holthuis, Pomorski et al 2001) GlcCer of fungal origin can act as elicitors that trigger a plant defense response involving the transcription of specific genes and the structural features required have been identified in studies with

different molecular species of GlcCer from the phytopathogen M oryzae

(Koga, Yamauchi et al 1998, Umemura, Ogawa et al 2000)

Therefore, it is believed that lipid metabolism would be a wide range, large scale and signaling rich event, which also interconnects with the metabolism of other nutrients, for example, that of glycogen Lipids would be

of primary importance in understanding the development and pathogenesis of

a fungus lik M oryzae

1.5 Recent Advances of M oryzae Research

Autophagy has been an interesting topic for M oryzae research recently

A number of genes involved in non-selective autophagy were shown to be vital for virulence (Kershaw and Talbot 2009) Glycogen autophagy was also

proven to be closely related to the asexual differentiation of M oryzae such as

conidiation (Deng, Ramos-Pamplona et al 2009, Deng and Naqvi 2010)

In terms of interaction with the host, a septin ring was found to organize

an extensive toroidal F-actin network at the base of the infection cell surrounding the appressorium pore, making direct phosphoinositide linkage

with the membrane with the help of ERM proteins and acting as a diffusion barrier to ensure localization of proteins (for example the Rvs167 I-BAR protein, and the WASP/WAVE complex) that are involved in membrane curvature at the tip of the emerging penetration peg and F-actin polymerization (Dagdas, Yoshino et al 2012)

M oryzae was discovered to express Slp1 and overcome the first line of

immunity defense initiated by CEBiP (Mentlak, Kombrink et al 2012) Recognizing this and also to increase the disease resistance in rice, the chimeric receptor between the chitin elicitor receptor CEBiP and the receptor-like protein kinase Pi-d2 was adopted and the resistance was significantly improved (Kouzai, Kaku et al 2013)

Studies on the genomics (Dean 2005, Xue, Yang et al 2012) and proteomics (Kim, Kim et al 2004, Kim, Yu et al 2004, Bhadauria, Wang et al

2010, Wang, Wu et al 2011) of M oryzae are still ongoing A recent profiling

of the transcriptome during appressorium development also confirmed the important role played by autophagy, lipid metabolism and melanin biosynthesis in appressorium differentiation, with Pmk1 MAP kinase as a key global regulator of appressorium-associated gene expression (Soanes, Chakrabarti et al 2012)

However, because lipidomic studies using mass spectrometry (MS) are

relatively new technologies, there has been no lipidomic study on M oryzae

available up to date

1.6 Proposed Model for Turgor Pressure Production

Many plant fungi develop specialized structures called appressoria for their pathogenesis (Mendgen, Hahn et al 1996) and it was first discovered that

the appressoria of M oryzae uses glycerol to generate pressure that ruptures

plant cuticles (de Jong, McCormack et al 1997) However, as more research

was done, now it it known that there are three possible sources for M oryzae

to produce turgor pressure: glycogen, trehalose and lipids, which mainly are the TAGs; nonetheless, the exact mechanism for turgor pressure production is still unclear (Wang, Jenkinson et al 2005) Other than enable a fungus to penetrate into the host cells, turgor pressure is also involved in the hyphal growth (Gervais, Abadie et al 1999) The measurement of turgor pressure could be done using either pressure probe or micromanipulation (Wang, Hukin

et al 2006)

Glycogen rosettes are found to accumulate within appressoria during their development (Lepore, Gualtieri et al 2011); but at the onset of turgor generation, glycogen quickly disappears in the appressoria during melanization and turgor generation (Thines, Weber et al 2000, Lepore, Gualtieri et al 2011) Besides, glycogen hydrolysis is also found to have occurred before turgor generation, and interestingly, the enzymes by which glycerol could be synthesized from storage carbohydrates, namely nicotinamide adenine dinucleotide phosphate (NADH)-dependent glycerol-3-phosphate dehydrogenase and nicotinamide adenine dinucleotide phosphate (NADPH)-dependent glycerol dehydrogenase, have been proved to

be present in appressoria but not induced during turgor generation (Thines,

Weber et al 2000) On the other hand, in M oryzae, the breakdown of TAGs

is found to be the main source for glycerol production, along with free fatty acids (Thines, Weber et al 2000, Weber, Wakley et al 2001) It has been also

shown that peroxisomal fatty acid β-oxidation in M oryzae is required for its

pathogenicity (Ramos-Pamplona and Naqvi 2006) But puzzling as it may

seem to be, in the context of breaking down the glycogen storage, both GPH1 and SGA1 are found to be needed for conididation but not pathogenicity (Deng,

Ramos-Pamplona et al 2009, Deng and Naqvi 2010) Now the question is: while TAGs seem to play a major role in producing glycerol and therefore turgor pressure, what roles does glycogen play? To make things more confusing in a way, trehalose synthase Tps1 was proven a must for the full virulence of the fungus without which the turgor pressure is affected (Foster, Jenkinson et al 2003, Wang, Jenkinson et al 2005) while trehalases are not really needed for the turgor pressure generation (Foster, Jenkinson et al 2003)

On the other hand, detailed studies have identified a role for Tps1 in regulating the expression of several virulence related genes and explained the mechanism

of such regulation (Wilson, Jenkinson et al 2007, Wilson, Gibson et al 2010) However, there is still one question remaining: does trehalose have anything to

do with turgor production?

The level of trehalose is found to have increased during the first 2 h as the conidium germinates, but soon after that trehalose is almost completed degraded; the trehalase activity level remains high during the whole period of germination (Foster, Jenkinson et al 2003) All these would naturally lead us

to a question: could glycogen be (partially) converted to TAGs? If that is true, trehalose could have been synthesized and broken down very rapidly during the first 2 h of germination to ensure a healthy flow of inorganic phosphate

Therefore, a hypothetical model was proposed for turgor pressure

production (Figure 2) Based on the model, glycogen would be robustly

broken down for TAG synthesis In order that TAGs could be synthesized after the breakdown of glycogen, it is suggested that possibly the intermediates

of glycolysis like glyceraldehyde-3-phosphate (G3P) / dihydroxyacetone phosphate (DHAP) would be produced along the glycolysis pathway, which

could be used for that purpose (Coleman and Lee 2004) FFAs may be also

synthesized in a de novo manner at the same time Due to possibly a big

amount of FFAs produced, some could have been exported out of the conidia, which was indicated indirectly in a previous research work (EBATA, YAMAMOTO et al 1998), while some others would be used for energy production and / or synthesis of other molecules, i.e TAGs

As for trehalose metabolism, the regulatory role in glycolysis and gluconeogenesis was known (Thevelein and Hohmann 1995, Foster, Jenkinson

et al 2003), especially when a huge amount of glycogen must be metabolized

in a short time found in the case of M oryzae Having said that, mutants of M

oryzae lacking NTH1, a trehalase, are reduced in virulence, and the reason

appears to be due to a decreased ability to perform invasive growth within rice tissues (Foster, Jenkinson et al 2003) The continuous requirement for the turgor pressure and/or energy even during intra-cellular invasive growth is suspected, and NTH1 therefore could be a long acting enzyme that continuously breaks down trehalose to meet such a need This suggests a

possibility of continual synthesis of trehalose by M oryzae after the initial

deplete during germination A similar observation was found in the oomycete

Saprolegnia ferax (S ferax), whose rate of hyphal growth through solid

medium decreased as the turgor was reduced, and this observation suggested that turgor provides the force for invasive hyphal growth (Money 1995) In contrast, TRE1, which is responsible for encoding the main trehalase activity

during M oryzae spore germination, appeared to be completely dispensable

for pathogenesis This shows the insignificant role played by trehalose breakdown during the early stage of turgor pressure generation Therefore, it is proposed that trehalose and especially the synthesis of trehalose is to ensure a successful breakdown of glycogen and synthesis of TAG, providing the materials and energy needed for pathogenesis

Nowadays, to reduce costs and speed up the R&D process, computational methods able to predict the binding affinity of small molecules to specific biological targets are very often used to come out new hit compounds (Parenti and Rastelli 2012) Computer-aided drug design, especially when optimizing the leads, depends heavily on the accurate prediction of binding affinities between the ligands and protein receptor Docking has been widely used for screening chemical databases and lead selection (i.e virtual screening), due to

the speed and easy to use (Leach, Shoichet et al 2006, Moitessier, Englebienne et al 2009) Though generally acceptable at predicting correct binding modes and therefore useful for virtual screening, scoring functions still could not identify the crystallographically correct binding mode from other suggested poses during a docking study or accurately predict binding affinities (Ferrara, Gohlke et al 2004, Stjernschantz, Marelius et al 2006, Warren, Andrews et al 2006, Cheng, Li et al 2009, Cross, Thompson et al 2009) A very important reason is that during docking, the protein is mostly kept rigid and only the ligand would be allowed to be flexible In addressing the issue, a few current docking programs allow for some protein flexibility by making the side chains of the active-site residues flexible or by docking a ligand to an ensemble of protein structures (ClauBen, Buning et al 2001, Sherman, Day et al 2006, Bottegoni, Kufareva et al 2008, Totrov and Abagyan 2008) However, with all these considerations, binding confirmations could be still affected much by factors like induced-fit and solvation in the active site Besides, the observation on the need of intrinsic disorders for the normal function of some proteins (Dunker, Lawson et al 2001, Dunker, Brown et al 2002) would make the topic all the more complicated

At the same time, it has been shown by a few studies that performing molecular dynamics (MD) and free energy calculations starting from docked poses can greatly increase the accuracy of binding affinity predictions (Stjernschantz, Marelius et al 2006, Andér, Luzhkov et al 2008, Carlsson, Boukharta et al 2008, Stjernschantz and Oostenbrink 2010) Considered as an experimental bridge between the structures and macroscopic kinetic and thermodynamic data, MD simulation could be used to obtain information about dynamic and structural properties, some of which are mean atomic fluctuations and displacements, enzyme ligand binding efficiencies, free-energies, or even domain movements (Haberl, Othersen et al 2009) One

thing however to point out during MD simulation is the importance to use accurate starting structures, which has been reported in a number of studies (Stjernschantz, Marelius et al 2006, Andér, Luzhkov et al 2008, Carlsson, Boukharta et al 2008, Nervall, Hanspers et al 2008)

Because of the potential economical benefits, the control of rice blast

disease would remain a hot topic for research on M oryzae However, the first

step is to identify a protein target, the distortion of whose function would stop/alleviate rice blast disease Subsequently, compounds found in databases like the US National Institutes of Health (NIH) Molecular Libraries Small Molecule Repository (MLSMR) could be screened for potential inhibitors, and further molecular dynamics simulation study was employed to illustration possible mechanisms of their action It would be of great interest for future research to experimentally determine the efficiencies of such inhibitors and hopefully offer druggable candidates for rice blast disease control

1.8 Aims of the Project

In view of the crucial role played by the lipids in the pathology of M oryzae, a comprehensive knowledge of the lipidome would form a basis for

hypothesis generation In addition, as genome sequences are becoming increasingly available for different biological species, such a resource of lipidomic knowledge would add a complementary layer of information for system biology approaches (e.g annotation of biochemical pathways, evolution of enzymatic functions, etc.) Therefore, a lipidomic study to identify and quantify the lipid species would be a necessary first step before the functional studies Discoveries of the changes of the lipid species and their proposed functions are expected to widen our understanding of the pivotal role

of the lipid metabolism in the pathological processes; such understanding

could also be of help for identifying key enzymes involved in the pathogenesis and even guiding the corresponding inhibitor design for rice blast disease control As for as this is concerned, it is our best interest to also validate the

turgor production model we proposed in Section 1.6

Therefore, using liquid chromatography-mass spectrometry (LC-MS) technology as a technology platform, it is our desire to identify and quantify as

many lipid species as possible in the lipidome of M oryzae Polar

(phospholipids and ceramides) and non-polar lipids (TAGs and DAGs) are included in this study Since the latter class of lipids and fatty acyls might be functionally linked to turgor pressure production, two beta oxidation pathway

mutants, namely Δech1 (Hiltunen, Mursula et al 2003) and Δfox2

(Ramos-Pamplona and Naqvi, unpublished), are to be analyzed and compared with the wild type (WT) strain Information from such a study could be suggestive of possible functions played by mitochondria and peroxisomes in turgor pressure production A more detailed look into the TAG metabolism of the WT strain was also expected to provide valuable insights on the mechanism of turgor production

Finally, with the information from the lipidomic studies, we plan to explore the possibility of designing possible inhibitor(s) to interfere the lipid

metabolism of M oryzae and contributing to the rice blast disease control

through bioinformatic approach

Overall, for the first time, our research is aiming to profile and analyze the

lipidome and its changes of M oryzae across different time points as a

conidium eventually develops into an appressorium, validate a possible model for turgor pressure production, and last of all to design inhibitor(s) against the lipid metabolism of the fungus as well

CHAPTER 2: MATERIALS AND METHODS

The research work on the lipidomics of M oryzae involved multiple steps

of growing the different fungal strains, harvesting the conidia, extracting and analyzing the lipids, etc Though they were described in much detail in the following sections, a flow chart was also provided for better comprehension (Figure 3)

2.1 Growth Medium

M oryzae was allowed to grow on prune agar medium, whose formula is

as below: for every liter of medium, 1 g of yeast extract, 2.5 g of galactose, 2.5

g of sucrose, 20 g of agar and 40 ml of prune juice were added, and the pH value was adjusted to be 6.0 using 10 M NaOH solution

M oryzae was allowed to grow under light condition and at room

temperature for about 10 days before conidia were harvested using miracloth (CalBiochem, San Diego,CA)

2.2 The Fungal Strain, Growth Condition and Appressorium Formation

The WT strain B157 and the two knockout mutants (ΔEch1 and ΔFox2) were allowed to grow on prune agar (PA) plates for about 10 days before the conidia were applied to the glass cover slips for developing appressoria at either 13 time points ( 0 h, 2 h, 4 h, 6 h, 8 h, 10 h, 12 h, 14 h, 16 h, 18 h, 20 h,

22 h and 24 h) and 8 copies for each time point, or 4 time points (0 h, 2h, 6h and 24h) and 4 copies for each time point For each sample, about 70k conidia were used, following the procedures applied in previous work (Ramos-Pamplona and Naqvi 2006, Sun, Suresh et al 2006)

2.3 Lipid Body/Droplet Staining by BODIPY

Lipid droplets in conidia, gemtubes and appressoria were visualized by staining with BODIPY®493/503 (D-3922, Invitrogen) Conidia were harvested from 10-day PA plates cultured under light, followed by two washes and resuspended in distilled water to a concentration of 1×105 spores/ml

Conidia were then inoculated onto hydrophobic glass cover slips in a moist chamber at room temperature Samples were observed at intervals for appressorium formation and lipid mobilization, by mounting directly in fresh BODIPY® (493/503) solution with a final concentration of 10 µg/ml (stock, 1 mg/ml in ethanol) for 15 min

In order to visualize the nuclear dynamics and mitosis in vivo, an

hH1-RFP fusion protein (to label the nuclear histone H1) was expressed in

wild type Magnaporthe B157 using the Neurospora crassa (N crassa) CCG1 promoter

2.4 Live Cell Fluorescence Microscopy

Live cell fluorescence microscopy was performed using a Zeiss Axiovert 200M microscope (Plan Apochromat 100X, 1.4NA objective) equipped with

an UltraView RS-3 spinning disk confocal system (PerkinElmer Inc., USA).The system also included a CSU21 confocal optical scanner, 12-bit digital cooled Hamamatsu Orca-ER camera (OPELCO, Sterling, VA, USA), and a 491 nm 100mW and a 561nm 50mW laser illumination controlled by MetaMorphPremier Software (Ver 7.7.5, Universal Imaging, USA) The maximum projection was obtained from z-stacks consisted of 0.5 µm-spaced planes using the MetaMorph built-in module Lastly, all the images were analyzed by ImageJ (National Institutes of Health, Bethesda, MD)

2.5 Lipid Extraction

The total lipids were extracted using a similar approach described earlier

on (Shui, Guan et al 2010), though with certain modification Briefly, samples were homogenized using a PRO 200 homogenizer (PRO Scientific Inc.,