620

Tropical Biomedicine 28(3): 620–629 (2011)

Optimal protein extraction methods from diverse sample

types for protein profiling by using Two-Dimensional

Electrophoresis (2DE)

Tan, A.A.1, Azman, S.N.1, Abdul Rani, N.R.2, Kua, B.C.2, Sasidharan, S.1, Kiew, L.V.3, Othman, N.1,Noordin, R.1 and Chen, Y.4*

1Institute for Research in Molecular Medicine (INFORMM), Universiti Sains Malaysia2Fisheries Research institute (FRI), Batu Maung, Pulau Pinang3Department of Pharmacology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia4Oral Cancer Research and Coordinating Centre (OCRCC) and Dental Research & Training Unit,Faculty of Dentistry, University of Malaya, 50603 Kuala Lumpur, Malaysia*Corresponding author email: chenyeng@um.edu.myReceived 4 May 2011; received in revised form 2 June 2011; accepted 12 June 2011

Abstract. There is a great diversity of protein samples types and origins, therefore theoptimal procedure for each sample type must be determined empirically. In order to obtain areproducible and complete sample presentation which view as many proteins as possible onthe desired 2DE gel, it is critical to perform additional sample preparation steps to improvethe quality of the final results, yet without selectively losing the proteins. To address this, wedeveloped a general method that is suitable for diverse sample types based on phenol-chloroform extraction method (represented by TRI reagent). This method was found to yieldgood results when used to analyze human breast cancer cell line (MCF-7), Vibrio cholerae,Cryptocaryon irritans cyst and liver abscess fat tissue. These types represent cell line,bacteria, parasite cyst and pus respectively. For each type of samples, several attempts weremade to methodically compare protein isolation methods using TRI-reagent Kit, EasyBlue Kit,PRO-PREPTM Protein Extraction Solution and lysis buffer. The most useful protocol allows theextraction and separation of a wide diversity of protein samples that is reproducible amongrepeated experiments. Our results demonstrated that the modified TRI-reagent Kit had thehighest protein yield as well as the greatest number of total proteins spots count for all typeof samples. Distinctive differences in spot patterns were also observed in the 2DE gel ofdifferent extraction methods used for each type of sample.

INTRODUCTION

2DE Gel-based proteomics analysis stillremains the most effective way to resolvecomplex protein mixtures. In a 2DE system,proteins are separated by two distinctproperties according to their net charge inthe first dimension and molecular mass inthe second dimension. One of the greateststrengths of 2DE is the ability to resolveproteins that have undergone some form ofpost-translational modification. For example,the phosphorylated and glycosylated form ofa protein can be resolved from the non-phosphorylated and non-glycosylated form

by the 2DE method (Schulenberg et al., 2003;Zhou et al., 2007). In short, the 2DE method isable to detect different isoforms of proteinsthat arise from alternative mRNA splicingand proteolytic processing. Through thisapproach, real total protein expression wouldbe qualitatively and quantitatively compared(Yan et al., 1999).

In this study, four different kits ormethods, including the phenol-chloroformextraction method represented by TRI-reagent and EasyBlue, PRO-PREPTM ProteinExtraction Solution and lysis buffer wereused and applied. Using a diverse collectionof protein samples represented by the

621

human breast cancer cell line (MCF-7), liverabscess fat tissue/pus, Vibrio cholerae, andCryptocaryon irritans cyst, each type of2DE gel sample was compared in terms oftheir protein yield and spot pattern for eachof the above test kits. Although many studieshad reported on the technique used in theextraction of mammalian cell line (Wang et

al., 2004; Hardouin et al., 2006), none of themhave included the more complicated samples,such as pus and parasite cyst.

The presence of highly concentratedlipid, cholesterol, fibrin, extracellular fluid,mineral ions and actual necrotic tissuescomplicates proteomic analysis of liverpus samples extracted from liver abscesspatients (Wells, 2009). Furthermore, C.

irritans, a ciliated protozoan parasite whichinvades eyes, gills and skin of host fish andcauses white spots disease in marine fish,(Brown, 1951) is always a challenge when itcomes to protein extraction and separation.According to Herwig (1978), the parasitebecomes encysted during the reproductivestage and resists chemical treatment(Matthews et al., 1993; Kesintepe, 1995).Besides, Vibrio cholerae, a motile, gramnegative bacterium causing cholera diseaseamong humans is another interesting sampletype to study. The bacteria possesses anouter membrane with strong polyanioniclipopolysaccharide and glycerophospholipidselectrostatically linked by divalent cationsthat forms an effective barrier againstantibiotics, detergents and chemicals(Nikaido & Vaara, 1985; Sukupolvi & Vaara,1989; Vaara, 1992). Despite the recentadvances in proteomic research, the analysisof V. cholerae and C. irritans cyst proteomestill remains a considerable challenge dueto their thick and chemical resistive wall.

To our knowledge, no study to date hasaddressed the proteome map for parasite cystand pus samples due to difficulties in theprotein extracting procedure. Althoughvarious methods are available for proteinextraction, most of the methods requirefurther optimization to extract proteins fromsamples. Therefore, our aim of this study isto develop and standardize a method toextract the total protein from different types

of samples effectively by generating highly-resolved 2DE protein profiles.

MATERIALS AND METHODS

Sample Preparation

Cell culturing

MCF-7 cells were cultured in Dulbecco’smodified Eagle’s medium (DMEM)containing 10% Fetal Bovine Serum(Invitrogen) and maintained in a 5% CO2

atmosphere at 37ºC. Cells were allowed togrow until 80-90% confluent and weresubsequently harvested by using trypsindigestion. Approximately 5x105 of cells werethen spin down in a 1.5mL microcentrifugetube by centrifugation (4ºC, 10 minutes, 220g)before subjected to each extraction methods.

Harvesting of Cryptocaryon irritans cyst

The C. irritans population was maintainedusing the sea bass, Lates calcarifer, as a hostin a tank containing sea water kept at 27±1ºC.Glass slides were placed on the bottom ofthe tank for 15 hours to collect adhering C.

irritans cysts. The collected cysts were thencleaned with sterile sea water and transferredto a 1.5mL microcentrifuge tube andcentrifuged at 2000g for 5 minutes at 4ºC toremove the sea water. The cysts were storedat -80ºC until further processing.

Harvesting of Vibrio cholerae

Vibrio cholerae strain el tor was obtainedfrom Microbiology & Parasitology Depart-ment, School of Medical Science, UniversitiSains Malaysia. The bacteria were grownin Luria Bertani (LB) broth at 37ºC in ashaking incubator for 24 hours. Precedingeach protein extraction method, cells wereharvested from 10mL of culture broth bycentrifugation (4ºC, 15 minutes, 1000g).

Pus Sample Collection

Pus samples from liver abscess patients werecollected from the Hospital Universiti SainsMalaysia (HUSM), Kelantan, Malaysia.Approximately 1-3mL of liver pus wasaseptically aspirated from the liver abscessunder ultrasound guidance from each patient.

622

This study was conducted in accordance withthe requirements of USM Human EthicsCommittee.

Protein Extraction

TRI-reagent Kit

Protein extraction using TRI-Reagent kit(Molecular Research Center, Cincinnati, US)was done as recommended by the supplier.Briefly, 1mL of TRI-reagent and chloroformwas added sequentially per 5x105 of cells /0.05g of pus and centrifuged at 12 000g at4ºC for 15 minutes to separate into DNA, RNAand protein layers. Instead of following themanufacturer’s protocol, we modified theprotein precipitation method where theprotein was precipitated by pre-chilledacetone at ice for 15 minutes.

EasyBlue Kit

Protein Extraction using EasyBLue Kit(Talron Biotech) was done as recommendedby the supplier. Briefly, 1mL of EasyBluereagent and chloroform was addedsequentially per 5x105 of cells/ 0.05g ofpus and centrifuged at 12 000g at 4ºC for 15minutes to separate into DNA, RNA andprotein layers. Isopropanol was then addedto precipitate the protein.

PRO-PREPTM Protein Extraction Solution

Protein Extraction using PRO-PREPTM

Protein Extraction Solution (iNtRONBIOTECHNOLOGY) was performed asrecommended by the supplier. Briefly, 5x105

of cells or 0.05g of pus tissue was suspendedin 40µl or 1mL of PRO-PREPTM solutionrespectively and incubated in -20ºC for 30minutes. The resulting cell lysate was thencentrifuged at 14 000g for 5 minutes at 4ºC.The supernatant was collected and subjectedto downstream processing.

Lysis Buffer

Harvested cells, bacteria, pus or cysts weredisrupted with a cocktail of 8M urea (Bio-Rad), 2M thiourea (Bio-Rad), 4% (w/v) CHAPS(GE HealthCare) , 20mM DTT, 2% Pharmalyte(GE healthcare) and protease inhibitorcocktail (GE HealthCare) and incubated for30 minutes in 4ºC. The resulting cell lysatewas then centrifuged at 14 000g for 15

minutes at 4ºC. The supernatant was collectedand subjected to downstream processing.

Two-dimensional Electrophoresis

The extracted proteins was rehydrated in150µl of rehydration buffer (8M urea, 2Mthiourea, 4% CHAPS, 0.5% pharmalyte (GEHealthcare), 20mM DTT) overnight in 13cmprecast immobilized dry strips pH 4-7 (GEHealthcare Bio-Sciences, Uppsala, Sweden).The strips were then subjected to isoelectricfocusing (IEF) as previously described (Chenet al., 2008). After the IEF, the strips wereequilibrated and subjected to seconddimensional separation at 16ºC using the 8-18% gradient polyacrylamide gel in thepresence of sodium dodecyl sulfate (SDS-PAGE). All samples were analyzed induplicate.

Silver staining

The 2D-PAGE gels were visualized by silverstaining as described by Heukeshoven &Dernick (1985).

Image analysis

The LabScan image scanner (Version 5) wasused to capture and store images of 2DE gels.The ImageMaster™ 2D Platinum Software(Version 5) was used to evaluate the proteinprofiles and present information obtainedfrom the 2DE gels. Average number ofproteins spot (n) including the unresolvedpeptides in each sample was tabulated.

RESULTS

Analysis of 2DE protein profiles

When these four types of samples weresubjected to 2DE and silver staining underthe resolving conditions adopted in thepresent study, most of the high abundanceproteins were detected. When the 2DEexperiments were performed on humanbreast cancer cell line (MCF-7) using fourdifferent kits or methods, comparable proteinprofiles or patterns were obtained for most ofthe resolved proteins (Figure 1). However,profiles of cell line treated with TRI-Reagentkit had shown distinctive protein spots withhighest amount of protein spots (n=175)

623

shown as compared to other kits or methodsused (Figure 1, Panel A). Panels B, C and D ofFigure 1 demonstrate typical 2DE proteinprofiles of cell lines treated with EasyBLueKit, PRO-PREPTM and lysis bufferrespectively.

The average number of protein spotsappeared fewer in the lysis buffer method(n=155), EasyBLue Kit (n=125) and finallyPRO-PREPTM (n=71) (Table 1). Although acomparable protein content was obtained byusing the lysis buffer method, the buffer was

Table 1. Average number of protein spots visualized on 2DE protein profiles of MCF-7 cell line, Vibrio cholera,

Cryptocaryon irritans cyst and pus, which samples extracted using four different kits or methods. Allsamples were analyzed in duplicate

Type of samplesAverage number of protein spots visualized on the 2DE protein profiles, (n)

TRI-reagent EasyBlue reagent ProPrep solution Lysis Buffer

MCF-7 cell line 175 125 71 155Vibrio cholerae 595 ns ns nsCryptocaryon irritans cyst 435 ns ns nsPus 306 ns ns ns

# n.s the protein spots was not detected accurately due to the heavy streaking on the 2DE gel

Figure 1. Representative 2DE protein profiles of the MCF-7 lysate extracted using (A) TRI-reagent, (B) EasyBlue reagent, (C) ProPrepTM solution and (D) lysis buffer. MCF-7 lysateextracted samples were treated with four different methods before subjected to 2DE andsilver stained. Protein spots were compared and analysed using ImageMaster™ 2D PlatinumSoftware Version 5

624

unable to eliminate the contaminatedsubstance effectively thus causing slightstreaking on the gel (Figure 1, Panel D). Ofthese four methods tested, ProPrep solutiongave the lowest protein content extractedwith major lines of streaking (Figure 1, PanelC). Even though the EasyBlue reagentcontains the same key ingredients (phenoland guanidine thiocyanate) as TRI-reagent,less protein was recovered and minorstreaking observed in samples treated withTRI-reagent (Figure 1, Panel B).

Similar analysis was performed on C.

irritans cyst, V. cholerae and the liverabscess fat tissue/pus using four differentextraction kits or methods. The results alsorevealed the highest expression of proteinspots in those samples treated using TRI-Reagent kit. The average number of proteinspots visualised and detected in C. irritans

cyst (Figure 2), V. cholerae (Figure 3) andthe liver abscess fat tissue/pus (Figure 4) washigh at n= 435, 595 and 306, respectively.Again, of all kits or methods used, only thosesamples treated with TRI-reagent kitachieved the best separation and producedhighly resolved gel. This had been shown in

Figure 1, 2, 3 and 4, where minimum streakingwith reproducible results only observed inthose samples treated with TRI-reagent kit.The other kits and methods were unable toproduce usable images due to heavystreaking or low number of proteins spotscounted (data not shown).

DISCUSSION

The key to obtain adequate results is in theway samples are treated. Ideally, the patternof the 2DE gel should reflect the proteincomposition without any losses andmodifications. The samples must not becontaminated with proteins and peptides notbelonging to the sample. Furthermore, toomuch salt in the sample disturbs IEF andleads to streaking patterns. For example,amphoteric compounds in samples willbuffer the gradient excessively in the areasof theirs pIs, which results in vertical narrowareas without protein spots. Therefore, thechemicals used must be of low salt contentand of high purity. Finally, the time taken forthe treatment must be kept to a minimum to

Figure 2. Representative 2DE protein profiles of the C. irritans cyst extractedusing TRI- reagent. TRI-reagent treated C. irritans cyst samples were subjectedto 2DE and silver stained. Protein spots were compared and analysed usingImageMaster™ 2D Platinum Software Version 5

625

Figure 3. Representative 2DE protein profiles of the V. cholerae extractedusing TRI-reagent. TRI-reagent treated V. cholerae samples were subjectedto 2DE and silver stained. Protein spots were compared and analysedusing ImageMaster™ 2D Platinum Software Version 5

Figure 4. Representative 2DE protein profiles of the pus from liver abscessextracted using TRI-reagent. TRI-reagent treated pus samples weresubjected to 2DE and silver stained. Protein spots were compared andanalysed using ImageMaster™ 2D Platinum Software Version 5

626

reduce the possibility of protein losses andmodification.

We attempted to standardize proteinpreparation method for good reproducible2DE protein profiles from diverse sampletypes, which was represented by humanbreast cancer cell line (MCF-7), V. cholerae,C. irritans cyst and liver abscess fat tissuein this study. We found that the most effectiveprotein extraction methods for the 2DEanalysis would be the phenol-chloroformextraction method by TRI-reagent kit. The kitallowed the visualization of maximumproteins spots while at the same time,produced highly resolved gels.

By using 2DE and image analysis, anymethods or kits used generally demonstratedgood expression of proteins for MCF7,mammalian cell lines. The samplepreparation of lysates from cultured cells isrelatively easy to prepare and stable. Unlikeother sample types, cell lines do not have aprotective mechanical barrier, thus theplasma membrane could be easily broken byany treatment to release the protein content.Among the four extraction methods tested inthis study, TRI-reagent gave the best resultswith most protein spots visualized, minimumstreaking and a highly resolved image.Although experiments performed using lysisbuffer and EasyBLue extraction had shownalmost comparable results as TRI-reagent,the protein expression had markedly lessspots and more streaking lines. ProPrep kitwas tested but found unsuitable for extractingprotein from MCF7.

By evaluating the protein profiles of V.

cholerae strain El Tor, it was found that thesamples treated with TRI reagent showedvisualization of most protein spots usingsilver staining. Also, in comparison to aprevious proteome map of V. cholerae strainEl Tor reported by Coelho et al. (2004), TRI-reagent kit was able to recover 56 more spotsthan reported. It was found that TRI-reagentis the most effective way to extract proteinsfrom this bacteria cell without involvingmechanical disruption (eg. Ultrasonication)to break the cell wall. Other kits or methodspresented significant challenges inproducing a presentable gel for V. cholerae

sample preparation; hence, they are not

suitable for protein extraction withoutmechanical disruption.

Our group is the first to report the 2DEprotein profile of parasite cyst of C. irritans.The TRI-reagent kit was able to extractprotein effectively despite the tough cysticwall of the parasite. On the other hand, muchstreaking was observed for those samplestreated with lysis buffer, EasyBlue andProPrep kit. These kits created a majorinterference during the IEF. It was deducedthat due to its chemical properties, the TRI-reagent not only lyses most of the cyst wall,furthermore it effectively removes salts andions that can interfere with IEF in order toproduce a good 2DE gel.

This study is also the first to isolateproteins from complicated and challengingpus samples using TRI reagent kit beforesubjecting to 2DE separation. In general,proteins extracted by most extractionmethods affected the IEF and subsequentlyproduced heavy streaking lines on the gel dueto the interference of substances such as salt,lipid and ions in the pus. However, TRI reagenttreated samples still rendered the best amongthe methods compared although minorstreaking was observed. In short, withregards to 2DE profiling studies, TRI-reagentis still the best for extraction of protein fromthick and chemically resistant cell wallsamples coupled with the presence ofinterfering compounds such as DNA, RNA,carbohydrate, proteolytic enzymes andoxidative enzymes.

TRI-reagent mainly contains phenol,chloroform and a chaotropic denaturingsolution (guanidine thiocyanate). Guanidinethiocyanate and phenol are strongdenaturalization and lysis reagents that cansimultaneously separate RNA, DNA andproteins from biological samples throughcentrifugation (Wu, 1995; Bracete et al.,1999). The protein samples separate intothe upper aqueous phase (RNA) and alower organic phase (DNA & protein)(Chomczynski & Sacchi, 1987). The DNA oflower organic phase is further precipitatedusing ethanol, whilst precipitation of proteinoccurs in acetone. Chloroform together withacetone will facilitate the dissolution of lipid(Merrill & Fleisher, 1932). For this study, the

627

TRI reagent protein precipitation method wasmodified by using ice cold acetone insteadof acetone kept in room temperature. This isin agreement to the study conducted byAskonas (1951) and Merrill & Fleisher (1932)that the precipitation of protein achieved thebest result by minimizing the degradation ofprotein at low temperature (0-5ºC) alone. Inshort, TRI-reagent has the ability to separateout RNA and DNA which effectivelyminimized the contaminant present andreduce the streaking on gel. Besides,guanidine thiocyanate denatures protease bypreventing further protein degradation andsubsequently enhances the protein recoveryof samples. All these factors make TRIreagent an excellent kit to extract proteinspecifically when used with cold acetone.

Although EasyBlue also containsguanidine thiocyanate and phenol, theamount of protein extracted was still lesserthan that treated with TRI-reagent. Itdemonstrated more streaking andinterference. We attribute this to theisopropanol protein precipitation used inEasyBlue protocol. Comparable results withTRI-reagent treatment was obtained when theprotocol was modified by replacingisopropanol precipitation solution withacetone. By acetone precipitation, we agreedthat the protein concentration could beretrieved with an increase of 10-20% (Wu,1995) and with reduced streaking on gel(result not shown).

ProPrep protein extraction solution is thefastest performing kit which requires lessthan 30 minutes to complete the extractionand contains 5 different kinds of proteaseinhibitors. However, unsatisfying results wereobtained when compared to TRI-reagent,which lacks protease inhibitors. Due to theshort incubation time of the samples usingthe ProPrep reagent, incomplete lysis mighthave caused production of low protein spots.The failure to remove interfering agentsfrom the samples also contributed to majorstreaking lines in the gel.

Conventional lysis buffer was used as acontrol in this study. This buffer extractedproteins from MCF7 cell lines rather

efficiently. However, the streaking on those2DE gels indicated that lysis buffer fails toeliminate the interfering substanceeffectively. A further step involving sampleclean-up is required to eliminate thestreaking lines. Another setback of the lysisbuffer is its inefficiency to extract proteinfrom challenging samples such as cyst andbacteria. Mechanical disruptions may berequired to facilitate the extraction of proteinfrom these samples.

Though TRI-reagent effectively extractedmost of the protein in a sample, the runningprocess is tedious and requiresapproximately 2-3 hours to complete.Furthermore, materials used in the kits (i.e.phenol and chloroform) are hazardous and afume hood facility is required. Nonetheless,it is still the best method to extract proteinfrom various challenging and difficultsamples. Its compatibility with MS-MS(Xanthopoulou et al., 2010) makes it a mostideal protein extraction method.

Appropriate sample preparation iscrucial for the outcome of 2D electrophoresis(Fountoulakis, 2001). The choice of methodused for sample preparation is the mostcritical step in any proteomics strategy aseach step influences protein yield, biologicalactivity and the structural integrity of thetarget protein. Among the four differentmethods tested, TRI-reagent delivered thehighest protein recovery and resolution for2DE proteome map as compared to others.Through the analysis of these 2DE proteinprofiles of diverse sample types involvinghuman breast cancer cell line (MCF-7), V.

cholerae, C. irritans cyst and liver abscessfat tissue which represented most cell lines,bacteria, cyst and pus, TRI reagent was foundto be an ideal method to extract proteins fromthese challenging samples. Furthermore, itrequired little optimization for each sampletype and is suitable for small laboratorieswith limited resources. It should be notedthat we were only able to optimize the proteinextraction method for the above samples.Further studies should be done on otherchallenging samples to confirm theeffectiveness of TRI reagent.

628

Acknowledgements. This study was fundedby a Research University grant fromUniversiti Sains Malaysia (Grant No 1001/CIPPM/811126) and the USM FellowshipAward.

REFERENCES

Askonas, B.A. (1951). The use of organicsolvents at low temperature for theseparation of enzymes. Application toaqueous rabbit muscle extract.Biochemical Journal 48: 42-48.

Bracete, A.M., Mertz, L.M. & Fox, D.K. (1999).Isolation of total RNA from smallquantities of tissue and cells. Focus 21:38-39.

Brown, E.M. (1951). A new parasiticprotozoan, the causal organism of a whitespot disease in marine fish Cryptocaryon

irritans gen. and sp. n. (Agenda andabstr. Sci. Meet., Zool. Soc. London, 1950).Proceedings of the Zoological Society of

London 11: 1-2.Chen, Y., Lim, B.K., Peh, S.C., Puterishafinaz,

A.R. & Onn, H. (2008). Profiling of serumand tissue high abundance acute-phaseproteins of patients with epithelial andgerm line ovarian carcinoma. Proteome

Science 6: 20.Chomczynski, P. & Sacchi, N. (1987). Single-

step method of RNA isolation by acidguanidinium thiocyanate-phenol-chloroform extraction. Analytical

Biochemistry 162: 156-159.Coelho, A., Santos, E.O., Faria, M.L.H.,

Carvalho, D.P., Soares, M.R., Kruger,W.M.A. & Bisch, P.M. (2004). A proteomereference map for Vibrio cholerae El Tor.Proteomics 4: 1491-1504.

Fountoulakis, M. (2001). Amino Acids, pp.363. 21th Ed. Springer-Verlag, Austria.

Hardouin, J., Canelle, L., Vlieghe, C., Lasserre,J., Caron, M. & Joubert-Caron, R. (2006).Proteomic Analysis of the MCF7 BreastCancer Cell Line. Cancer Genomics and

Proteomics 3: 355-368.Herwig, N. (1978). Notes on the treatment of

Cryptocaryon. Drum and Croaker 18: 6-12.

Heukeshoven, J. & Dernick, R. (1985).Simplified method for silver staining ofproteins in polyacrylamide gels and themechanism of silver staining. Electro-

phoresis 6:103-112.Kesintepe, M. (1995). Light and electron

microscopic observation on the ciliateCryptocaryon irritans throughout thelife cycle, PhD Thesis, University ofGeorgia, Athens, Georgia.

Matthews, B.F., Matthews, R.A. & Burgess, P.J.(1993). Cryptocaryon irritans

(Ichthyophthiriidae): the ultrastructure ofthe somatic cortex throughout the lifecycle. Journal of Fish Diseases 16: 339-349.

Merrill, M.H. & Fleisher, M.S. (1932). Factorsinvolved in the use of organic solventsas precipitating and drying agents ofimmune sera. The Journal of General

Physiology 16: 243-256.Nikaido, H. & Vaara, M. (1985). Molecular

basis of bacterial outer membranepermeability. Microbiological Reviews

49: 1-32.Schulenberg, B., Aggeler, R., Beechem, J.M.,

Capaldi, R.A. & Patton, W.F. (2003).Analysis of Steady-state ProteinPhosphorylation in Mitochondria using anovel fluorescent phosphosensor dye.The Journal of Biological Chemistry

278: 27251-27255.Sukupolvi, S. & Vaara, M. (1989). Salmonella

typhimurium and Escherichia colimutants with increased outer membranepermeability to hydrophobic compounds.Biochimica et Biophysica Acta 988: 377-387.

Vaara, M. (1992). Agents that increase thepermeability of the outer membrane.Microbiological Reviews 56: 395-411.

Wang, D., Jensen, R.H., Williams, K.E &Pallavicini, M.G. (2004). Differentialprotein expression in MCF7 breastcancer cells transfected with ErbB2,neomycin resistance and luciferase plusyellow fluorescent protein. Proteomics

4: 2175-2183.Wells, H.G. (2009). Chemical Pathology, pp.

269-271. Large Type Edition, BiblioLife,Charleston, South Carolina.

629

Wu, L.C. (1995). Isolation and long-termstorage of proteins from tissues and cellsusing TRIzol reagent. Focus 17: 98-100.

Xanthopoulou, A.G., Anagnostopoulos, D.,Vougas, K., Anagnostopoulos, A.K.,Alexandridou, A., Spyrou, G., Siafaka-Kapadai, A. & Tsangaris, G.T. (2010). Atwo-dimensional proteomic profile ofTetrahymena thermophila whole celllysate. In Vivo 24: 443-456.

Yan, J.X., Sanchez, J.C., Tonella, L., Williams,K.L. & Hochstrasser, D.F. (1999). Studiesof quantitative analysis of proteinexpression in Saccharomyces cerevisiae.

Electrophoresis 20: 738-742.Zhou, H., Liu, Y., Chui, J., Guo, K., Shun, Q., Lu,

W., Jin, H., Wei, L. & Yang, P. (2007).Investigation on glycosylation patternsof proteins from human liver cancercell lines based on the multiplexedproteomics technology. Archives of

Biochemistry and Biophysics 459: 70-78.