Rice is the staple food source of nutrients for both humans andlivestock throughout the world. The major proteins in rice seedsare the storage proteins, glutelins and prolamins. Their mRNAs arelocalized at different subdomains of the endoplasmic reticulum(ER). Glutelin mRNAs are targeted to the cisternal ER, whileprolamin mRNAs are targeted to the protein body ER (PB-ER)membranes that delimit intracisternal prolamin inclusions. Theglup4 mutant, which causes glutelin mRNAs to be transported tothe PB-ER instead of their normal destination on the cisternal ER,lacks the small GTPase, Rab5a. In order to find the underlyingcause of this error in mRNA targeting, these studies wereundertaken to identify an associated effector protein of Rab5a thatspecifies its role in RNA targeting. We used Escherichia coli toexpress GST-tagged-Rab5a positive and negative constructs. Rab5a-Q70L, the GTP fixed (active) form, was used as a positive controlwhile Rab5a-EM960, which has mutation in the effector region(G45D), was used as a negative control. The purified GST-Rab5aproteins were attached to Glutathione-linked Sepharose resin tocreate an affinity chromatography column. These columns wereused to “fish out” the Rab5a associated effector proteins in ricedeveloping seed extracts. The eluted proteins from the positive andnegative Rab5a columns were directly compared by SDSpolyacrylamide gel electrophoresis. Initial studies show nosignificant differences in the polypeptides eluted from thesecolumns.

• Q70L, EM960 and Vector columns were both successfullyprepared.

• To prevent non-specific binding, the rice seed extract wascrudely purified by passing through the GST-tag column. Sincethe eluted extract also contained GST-tag protein, futureexperiment may require decreased DTT concentration in theCSB buffer.

• The use of SDS and Urea elution buffer led to the elution ofGST-Rab5as as well, thus, we have to reconsider elution bufferconditions.

• Comparison of the elution samples by SDS-PAGE of positiveRab5a column, Q70L, with the negative column, EM960, wefound specific bands in Q70L elute fraction that does notappear in the negative control, but protein content is too low.

Rice is a major staple food source and as a source of dietaryprotein to both human and livestock (1, 2). The major storageproteins in rice (Oryza Sativa) are prolamine and glutelin, which aresynthesized on the endoplasmic reticulum (ER). In rice endosperm,prolamin protein makes up 20-30% of total rice storage protein (1,2). Prolamin mRNA is targeted to the PB-ER where it is translatedand then it forms intracisternal inclusions granules within the ERlumen, a structure called PB-I (3, 4, 5). On the other hand, glutelinmakes up 60-80% of the protein content in endosperm (6, 2, 7).Glutelin mRNAs are targeted to the cisternal ER. After they aresynthesized as 57kDa glutelin precursor (proglutelin), they aretransported to protein storage vacuoles (PSV) where they arecleaved into 30-36kDa acidic and 19-22kDa basic subunits, thusforming the PB-II (2, 3, 6, 8).

The glup4 mutant was induced by chemical mutagenesis using N-Methyl-N-Nitrosourea (MNU) (7, 9). This mutant is characterized byhigh proglutelin accumulation and the mis-localization of glutelinmRNA to the PB-ER rather than to the cisternal ER (4, 7, 8). Frommap-based cloning results, it was determined that this Glup4 geneencodes a small GTPase Rab5a, a protein involved in membranevesicular transport (3, 7, 8). The Rab family functions as regulators ofdistinct steps in membrane trafficking by recruiting specific effectorproteins onto membranes in their active GTP fixed form (10).Through their effectors, Rab GTPases regulate vesicle formation,actin- and tubulin-dependent vesicle movement, and membranefusion (10). Rab proteins are catalyzed by a GDP/GTP exchangefactor (GEF) (10). In the Drosophila egg, VPS22, 25, and 36 are partof an ESCRT-II complex that is associated with vesicle formation andbicoid mRNA localization (11). Therefore, rice Rab5a and VPS9 (GEFfor Rab5a) may have the ability to participate in the mRNA targetingmechanism.

In order to learn more of the Rab5a associated trafficking processinvolved, we set out to find Rab5a-binding proteins from rice seedextract. In doing so, we expressed two Rab5a variants in E. coli. TheRab5a-Q70L mutant is the constitutively active form and EM960, aglup4 mutant line, contains a mutation (G45D) in its effector region,thus preventing its role in specifying membrane vesicular transportand targeting. We prepared affinity chromatography columns toprobe for the unknown effector protein which is essential formembrane-vesicule associated RNA targeting (9). Results from SDS-PAGE of elution fractions showed specific bands in the positivecontrol. However, given the low protein content, we have toconsider optimizing our experimental conditions.

•Optimization of Rab5a affinity chromatography column. -Check buffer condition.

•Tandem mass spec analysis of candidate effector protein bands if specific bands are presented in Q70L elute fraction.

•Alternative methods: try pull-down assay or co-IP.

1. Shewry, Peter R., and Arthur S. Tatham. "The Prolamin Storage Proteins of Cereal Seeds: Structure and Evolution."Biochemistry Journal 267 (1990): 1-12. Print.

2. Kawakatsu, Taiji, Masayuki P. Yamamoto, Sakikio Hirose, Masahiro Yano, and Fumio Takaiwa. "Characterization of a NewRice Glutelin Gene GluD-1 Expressed in the Starchy Endosperm." Journal of Experimental Biology 59.15 (2008): 4233-245.Print.

3. Crofts, Andrew J., Naoko Crofts, Julian P. Whitelegge, and Thomas W. Okita. "Isolation and IdentiWcation Ofcytoskeleton-associated Prolamine MRNA Binding Proteins Fromdeveloping Rice Seeds." Planta 231 (2010): 1261-276. Print.

4. Satoh-Cruz, M., M. Fukuda, M. Ogawa, and T. Kumamaru. "Glup4 Gene Encodes Small GTPase, Rab5a in Rice." RiceGenetics Newsletter 25: 48-49. Print.

5. Ogawa, Masahiro, Toshihiro Kumamaru, Hikaru Satoh, Nobuo Iwata, Takeshi Omura, Zenzburo Kasai, and KunisukeTanaka. "Purification of Protein Body-I of Rice Seed and Its Polypeptide Composition." Plant and Cell Physiology 28.8(1987): 1517-527. Print.

6. Takaiwa, Fumio, Shoshi Kikuchi, and Kiyoharu Oono. "A Rice Glutelin Gene Family- A Major Type of Glutelin MRNAs CanBe Divided into Two Classes." Mol Gen Genet 208 (1987): 15-22. Print.

7. Doroshenk, Kelly A., et. al. "Characterization of the rice glup4 mutant suggests a role for the small GTPase Rab5 in thebiosynthesis of carbon and nitrogen storage reserves in developing endosperm". 2010. Unpublished.

8. Fukuda, Masako., et. al. " The small GTPase Rab5a is essential for intracellular transport of proglutelin from Golgiapparatus and endosomal membrane organization in developing rice endosperm". 2010. Unpublished.

9. Hagiwara, M., K. Kobayashi, T. Tadokoro, and Y. Yamamoto. "Rab5 Affinity Chromatography without Nonhydrolyzable GTPAnalogue." Z Naturforsch C 64.3-4 (2009): 303-06. Print.

10. Grosshans, Bianka L., Darinel Ortiz, and Peter Novick. "Rabs and Their Effectors: Achieving Specificity in MembraneTraffic." PNAS 103.32 (2006): 11821-1827. Print.

11. Irion, Uwe, and Daniel St Johnston. "Bicoid RNA Localization Requires Specific Binding of an Endosomal Sorting Complex."Nature 445 (2007): 554-58. Print.

This work was supported by the National Science Foundation REU program under grant DBI-0605016.

Purification of Rab5a Effector Responsible for Glutelin mRNA Location via Affinity Chromatography

Alice La1, Mio Sato2, Thomas W. Okita2.1CalState Fullerton; 2Institute of Biological Chemistry, Washington State Univ.

2010 REU in Interdisciplinary Plant Genomics

Grew E. coli cells containingconstructs in LB media untilOD600 is around 0.6. Add IPTGto final concentration of 0.1-1mM to express GST-fusedRab5a proteins.

Centrifuge to collect E.colipellet and resuspend in lysisbuffer, followed by celldisruption by micofluidizer.Then centrifuge again tocollect the supernatantcontaining the GST-fusedRab5s protein. Filter thesupernatant through 0.2μmpore filter.

Incubate supernatant with Glutathione Sepharoseresin for 2 hours to create mutant Rab5a affinity columns.

Check the starting material, resin, and flow through for the column condition by SDS PAGE.

De-husk and homogenize 7gof developing seeds in (3mLper 1g) CSB buffer percolumn. Then centrifuge at100g for 1min to removelarge starch grain. Thencentrifuge supernatant at3000 rpm for 10 min to getpellet. Resuspend the pelletin CSB buffer containing 1%Triton X-100.

Spin at 100Kg for 60 min.

Incubate the Rab5a column with 1mM GTP. Then was wash with NS buffer (w/o GTP). Mix resin with rice extract and incubate for 2hours at 4°C on rotating mixer.

Collect flow through sample.

Wash with buffer A until A280 is near zero.

Wash with buffer B until A280 is near zero.

Add 0.5 mL Elute buffer (A/B) and collect elutant as

sample. (Elute sample)

Check starting material, flow through, wash A, wash B, and elute sample fractions via SDS-PAGE.

To determine effector protein, cut out specific band in Q70L elute sample and for tandem mass spectrometry analysis.

supernatant

Pellet: Resuspendthe pellet in CSBbuffer containing1% Triton X-100,200mM NaCl. ThenSpin at 100Kg for60min.

Then dilute with an equal volume of CSB to

dilute detergent concentration.

Summary of BuffersLysis buffer 1xPBS, 5mM MgCl2 and 5mM 2-ME, 1mM PMSFCSB buffer5 mM Hepes-KOH pH7.510 mM MgOAc, 2mM EGTA1 mM PMSF, 1mM DTT5 ug/mL Leupeptin (5mg/mL)1 ug/mL PepstatinA (1mg/mL)Nucleotide Stabilization buffer (NS buf.)20 mM HEPES-NaOH pH7.5100 mM NaCl, 5 mM MgCl2, 1 mM DTT, 1 mM GTPWash buffer A20 mM HEPES-NaOH pH.7.5100 mM NaCl, 5 mM MgCl21mM DTTWash buffer B (High Salt)20 mM HEPES-NaOH pH.7.5250 mM NaCl, 5 mM MgCl21mM DTTElution buffer A0.2% SDS, 4M UreaElution buffer B20 mM HEPES-NaOH pH.7.52 M NaCl, 20 mM EDTA, 1mM DTT

end

star

t

Summary of Events

Fig. A1 Q70L Column Fig. A2 EM960 Column

Fig. A3 GST-tag column

1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8

1 2 3 4 5 6 7 7-2 8

Legend1: T=02: T=33: ppt.4: sup.5: After filtration6: FT 7: Wash (125 mL)7-2: Wash (125mL excess)

8: Resin

Column Preparation: The purified GST-Rab5a mutant proteins were incubated with

Glutathione Sepharose. To check for the binding efficiency of the mutant Rab5a to the

column, samples were collected at various points during the Rab5a mutant purification

and column preparation process as indicated in the legend. Fig. A1 is the positive Rab5a-

Q70L construct column. Fig. A2 is the negative construct column, EM960. Fig. A3 is the

GST-tag column used to crudely remove nonspecific binding in seed extract sample. The

50kDa band corresponds to GST-Rab5a mutant and the 26kDa band is the GST-tag.

1 2 3 4 5 6

1 2 3 4 5 6

Fig. B1 Q70L column

Fig. B2 EM960 column

Legend1: Starting Material 2: FT3: CSB buffer wash (10mL)4: Wash buffer A (40 mL)5: Wash buffer B (50 mL)6: Resin after Elute

Loading Columns with Rice Seed Extract: For all columns, lane 1

corresponds to the amount of starting material after simple purification

through GST-tag column to remove nonspecific binding proteins. The

amount of protein in fraction collected after different washes, (lanes 5-

6) as indicated in the legend, were also analyzed via SDS PAGE on a

15% polyacrylamide gel. The resin (lane 6) was also checked for any

bound proteins that was not eluted from the column to the elution

fraction.

GST Tag

Rab5a

72

55

43

34

26

72

55

43

34

26

72

55

43

34

26

Q70L EM960

Comparing Positive and Negative Columns. The elution

samples were analyzed via SDS PAGE on a 12% gel. Upper gels

(CBB and silver staining) were SDS/Urea elute sample. Lower gel

(silver staining) was high salt buffer elute sample. Equal protein

amount loaded between Q70L and EM960s sample based on

A280 value. The blue arrows indicate the bands that appear in the

positive control, Q70L column, but not in the negative control,

EM960 column. These bands were not detected in the CBB

staining gel indicating low protein content.

Q70L EM960

Q70L EM960

Rab5a

GST

Tag