Micron 41 (2010) 247–256

Effect of AEE788 and/or Celecoxib on colon cancer cell morphology usingadvanced microscopic techniques

P. Venkatesan, Soumen Das, M. Muthu Rama Krishnan, Chandan Chakraborty,Koel Chaudhury, Mahitosh Mandal *

School of Medical Science and Technology, Indian Institute of Technology, Kharagpur 721302, West Bengal, India

A R T I C L E I N F O

Article history:

Received 1 September 2009

Received in revised form 19 October 2009

Accepted 20 October 2009

Keywords:

Metastasis

Apoptosis

AEE788

Celecoxib

Morphology

Colon cancer

Confocal fluorescence

Cytoskeleton

A B S T R A C T

Analysis of changes in cancer cell morphology and cytoskeletal element induced by external stimuli is

focus of current cancer chemotherapeutic studies. Cancer cell cytoskeleton is complex network of

interwoven protein fibers composed of microtubules, microfilaments and intermediate filaments. These

interwoven protein fibers are responsible for maintaining cell morphology, movement, adhesion and

transmembrane signal transmission. In this study, morphological and cytoskeletal changes induced by

AEE788 and/or Celecoxib on colon cancer cell HCT 15 were analyzed using advanced microscopic

techniques. Cell proliferation assay was used for determining IC50 of AEE788 and/or Celecoxib on HCT 15.

Confocal microscopic analysis of AEE788 and/or Celecoxib treated HCT 15 was performed using

Rhodamine-Phalloidin (actin stain) and Hoechst 33342 (nuclear stain). Atomic force (AFM) and scanning

electron microscopic (SEM) studies were also performed to analyze cell morphology and cell wall

extension (filopodia and lamellipodia). In addition, quantitative analysis of morphological parameters

was studied using cellular image processing technique. This is the first report that combination of

AEE788 and Celecoxib additively increase growth inhibition and cell death on human colon cancer cell

HCT 15 as estimated by cell proliferation assay. Morphological analysis of AEE788 or Celecoxib treated

HCT 15 cell for 24 h have not revealed significant change in morphology under phase contrast

microscopy. But, slight morphological changes were observed in combination (AEE788 + Celecoxib)

treated HCT 15 for 24 h. In contrast, high resolution confocal laser fluorescence and atomic force

microscopic studies have revealed cell shrinkage, disorganized actin filament and, loss of filopodia and

lamellipodia. These changes were more prominent in combination of AEE788 and Celecoxib treated HCT

15 than either drug alone. These results may suggest antiproliferative and antimetastatic activity of

AEE788 and/or Celecoxib. Quantitative analysis of morphological parameters using cellular image

processing technique have shown decrease in mean area, perimeter, compactness and eccentricity of

combination drug treated cells than either drug alone. These results further support the confocal and

AFM study. Scanning electron microscopic study of AEE788 and/or Celecoxib treated HCT 15 has also

shown morphological changes and loss of filopodia and lamellipodia. In conclusion, this investigation of

morphological and cytoskeletal changes using advanced microscopic techniques present a significant

foundation for evaluating anticancer activity of a drug and form a new strategy for evaluating effect of

AEE788 and/or Celecoxib on colon cancer.

� 2009 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Micron

journa l homepage: www.e lsev ier .com/ locate /micron

1. Introduction

Based on World Health Organization (WHO) report, coloncancer is the second leading source of cancer-related deathsworldwide and accounts for 677,000 deaths per year. Mostcommonly used chemotherapeutic agents for treating colon cancerare Fluorouracil (5-FU) and leucovorin. However, drug resistance

* Corresponding author. Tel.: +91 3222 283578; fax: +91 3222 282221.

E-mail addresses: vnktsnvnk@iitkgp.ac.in (P. Venkatesan),

mahitosh@smst.iitkgp.ernet.in (M. Mandal).

0968-4328/$ – see front matter � 2009 Elsevier Ltd. All rights reserved.

doi:10.1016/j.micron.2009.10.008

and activation of multiple signaling pathways in cancer leads torecurrence of the disease which necessitates critical need foradditional targeting agent to produce additive or synergistic effect(Yokoi et al., 2005). Globally, several research groups are workingtowards effective treatment of colon cancer (Arico et al., 2002; Linet al., 2005; Sakoguchi-Okada et al., 2007; Sasaki et al., 2007).Among different types of anticancer agents, AEE788 exhibits targetspecificity and antiproliferative activity on different cancer celllines (Yu et al., 2007). AEE788 is a 7H-pyrrolopyrimidine thatinhibits multiple tyrosine kinases which mainly include vascularendothelial growth factor receptor (VEGFR), epidermal growthfactor receptor (EGFR) (Kuwai et al., 2008; Rebhun et al., 2006;

P. Venkatesan et al. / Micron 41 (2010) 247–256248

Younes et al., 2005; Yu et al., 2007). Further, AEE788 has alsoshown to potentiate inhibitory activity of CPT-1 (Camptozar) oncolon cancer cell line (Sasaki et al., 2007; Yokoi et al., 2005).Celecoxib, a selective cyclooxygenase-2 (COX-2) inhibitor, hasbeen extensively investigated for the treatment of colon cancereither alone or in combination with other anticancer agents(Auman et al., 2008; Gaisera et al., 2008; Hsiao et al., 2007; Maieret al., 2004; Marcel et al., 2008; Sakoguchi-Okada, 2007). As a neo-adjuvant Celecoxib has also been reported to potentiate effect ofother chemotherapeutic agents (Liu et al., 2008).

Study on osteosarcoma has revealed a close relationshipbetween morphology and function of the cell (Docheva et al.,2008). Some cancer cell possesses cell wall extension (lamellipo-dia) projecting towards other cells. Number, length and orienta-tions of the lamellipodia are critical factors in characterization of acancer cell. It has already been reported that 70% of human lungcarcinoma cell are bipolar, while the rest contains one or morelamellipodium (Oberleithner et al., 1993; You et al., 2000). Ingeneral, size of the lamellipodium can be up to 100 mm long incancer cell. In motile cells, filopodia is a fine protrusion thatemanate from lamellipodium (thin sheets of actin filaments) withapproximate diameter of 100–300 nm, height less than 50 nm andextends up to 5 nm (Doak et al., 2008). Both the filopodia andlamellipodia are highly dynamic, rapidly assembled and disas-sembled structure which has important roles in motility, guidanceand adhesion of cells. Further, actin filaments present in thecytoskeleton (F-actin and myosin) have also been reported toparticipate in motility, invasion and other cellular signalingprocess of cancer cells (Docheva et al., 2008; Zhou et al., 2008).

High resolution advanced microscopic techniques such asconfocal laser fluorescence microscopy, atomic force microscopyand scanning electron microscopy are capable of providingmorphological features and cytoskeletal information of a cell(Doak et al., 2008). While analyzing the confocal images,researchers typically assess changes in cellular structures and/orin intra-cellular molecule expression. Nevertheless, the observa-tion is solely qualitative and often leads to considerable variation.To improve the reliability of confocal microscopic analysis byminimizing the subjectivity, it is important to analyze morpho-logical features from the image in a quantitative way using digitalimage processing technique. This technique may be used forquantitative analysis of morphological features of a cancer cell(Esgiar et al., 1998, 2002; Gil et al., 2002). AFM, a scanning probemicroscopy, has been proved to be a powerful tool in cell biologydue to its high resolution capabilities. In addition, the morpho-logical analysis of cancer cell using high resolution SEM would alsoprovide clear insight regarding filopodia and lamellipodia (Lindaet al., 1976).

Anticancer agents mediated morphological changes in cancercell may potentially be valuable for evaluating anticancer activityof a drug (Bour-Dill et al., 2000). This morphological analysis alongwith molecular analysis can be used for cancer diagnosis andtherapeutics (Zhou et al., 2008). To the best of our knowledge, noresearch work has been reported to examine anticancer activity ofa drug using advanced microscopic techniques. This has promptedus to utilize the advanced microscopic techniques such as confocallaser fluorescence microscopy, AFM and SEM to assess morpho-logical and cytoskeletal changes in cancer cell.

2. Materials and methods

2.1. Cell line and culture condition

Colon cancer cell line HCT 15 was obtained from NationalCentre for Cell Science, Pune India. AEE788 and Celecoxib weregenerously provided by Novartis Pharma and Aarthi Drug Ltd.,

India, respectively. HCT 15 was grown in RPMI 1640 mediumsupplemented with 10% fetal bovine serum (FBS). Adherentmonolayer HCT 15 culture was maintained on plastic cultureflask and incubated at 37 8C in 5% CO2. The culture was maintainedno longer than 12 weeks after recovery from frozen stocks.

2.2. Cell proliferation assay of AEE788 and/or Celecoxib

AEE788 and Celecoxib were dissolved in DMSO to prepare stocksolutions. Then AEE788 (0–20 mM) and Celecoxib (0–200 mM)stocks were diluted to appropriate concentrations. Cell proliferationassay was performed using standard procedure as describedpreviously (Kim et al., 2005; Lee et al., 2008; Wanga et al., 2008).HCT 15 cells were cultured at a density of 2.5 � 103 cells/well in a 96-well plate for 24 h and treated with either drug (AEE788 (0–20 mM)/Celecoxib (0–200 mM) alone or in combination for 72 h. After 72 h oftreatment, 100 ml of MTT solution (1.0 mg/ml in RPMI) was added toeach well and the plate was incubated for another 4 h at 37 8C. Thepurple formazan crystal was dissolved in 100 ml DMSO per well. Theformazan dye was measured spectrophotometrically using micro-plate reader (570 nm Biorad Microplate reader 5804R). All the assayswere performed in triplicate and percentage of cell viability relativeto the untreated control cells (% control) defined as: [(Abs570 treatedcells)/(Abs570 control cells)]� 100, where absorbance is abbre-viated to Abs. In this study, IC50 of the respective drugs were used totreat colon cancer HCT 15 for 24 h and further analyzed usingadvanced microscopic techniques.

2.3. Confocal laser fluorescence microscopy

Confocal laser fluorescence microscopic study was performedas mentioned in previous studies (Hsiao et al., 2007; Lin et al.,2005). In brief, the HCT 15 cell was cultured using RPMI 1640medium supplemented with 10% FBS on a sterile glass slide for 2days at 37 8C, 5% CO2. Cells were then treated with IC50 of AEE788,Celecoxib, and combination of AEE788 and Celecoxib for 24 h. Aftertreatment, the cells were washed three times with PBS (pH 7.4)followed by incubation in 3.7% formaldehyde in PBS for 10 min.The samples were further washed with PBS and permeabilizedusing 0.1% Triton X-100 for 5 min. Then non-specific binding siteswere blocked using PBS containing 10% FBS and 0.3% Triton X-100for 1 h and incubated with Rhodamine-Phalloidin for 30 min atroom temperature. Following PBS wash and the cells were stainedwith 5 mg ml�1 Hoechst 33342 for 30 min. Fluorescence imagesfrom the stained constructs were obtained using confocal laserscanning microscope at 40� (CLSM, Olympus FV 1000 attachedwith inverted microscope IX 81, Japan) equipped with Argon(488 nm) and HeNe (534 nm) lasers. Two-dimensional multi-channel image processing was done using FV 1000 Advancesoftware version 4.1 (Olympus, Japan).

2.4. Morphological analysis by cellular image processing

In order to segment out cytoplasm and nucleus of the controland drug treated HCT 15 following image processing techniqueshas been used. It was observed that these images have a fairamount of ambient noise. Therefore, to reduce this noise medianfiltering was applied on the images. Color deconvolution is anortho-normal representation which is used for enhancingcontrast between cytoplasm and nuclei of HCT 15 cell. Further-more, the cytoplasm and nuclei are extracted using Otsu method(Jeong et al., 2005; Shahbazi et al., 2008). Region and boundarybased shape features have been considered to analyze shape of thecell. Region based shape features viz., area, compactness andboundary based features viz., perimeter, eccentricity and FD wereextracted and evaluated for these segmented cytoplasm and

Fig. 1. Antiproliferative effect of AEE788 and/or Celecoxib. The inhibitory effects of

AEE788 and/or Celecoxib were measured using a MTT assay.

P. Venkatesan et al. / Micron 41 (2010) 247–256 249

nuclei to discriminate between control and drug treated coloncancer cell HCT 15.

2.4.1. Feature extraction

A binary image is mathematically represented by

f ðx; yÞ ¼ 1; if ðx; yÞs the objectf ðx; yÞ ¼ 0; Otherwise

Area: The area was obtained by counting number of pixelspresent in binary image of the cytoplasm or nucleus.

Area ¼X

x

Xy

f ðx; yÞ

where f(x,y) is binary image of cytoplasm or nucleus.

f ðx; yÞ ¼ 1; ifðx; yÞs the objectf ðx; yÞ ¼ 0; Otherwise

Perimeter: Perimeter of the cytoplasm and nucleus wasobtained by counting number of boundary pixels present in thecytoplasm or nucleus.

Perimeter ¼X

x

Xy

f ðx; yÞ

where f(x,y) is extracted boundary of the cytoplasm or nucleus

f ðx; yÞ ¼ 1; ifðx; yÞs the objectf ðx; yÞ ¼ 0; Otherwise

Eccentricity: Two features viz., eccentricity and compactnesshave been extracted for our study. Eccentricity was calculated bythe following equation

Eccentricity ¼ ða2 � b2Þ

1=2

a

where a is major axis and b is minor axis.Compactness: Compactness is proportional to the area of each

cytoplasm and nucleus divided by the square of perimeter, whichis mathematically represented as

Compactness ¼ ðPerimeterÞ2

area

In addition, fractal dimension (FD) is one of the methods toperform texture analysis. Colon cancer cell boundary can beviewed as three dimensional object having intensity variation ascovering surface on the two-dimensional spatial plane. Any surfaceA in Euclidean n-space is self-similar if A is the union of Nr distinct(non-overlapping) copies of itself scaled up or down by a factor of r.Mathematically FD is defined by

FD ¼ log Nr

log Nð1=rÞ

2.5. Atomic force microscopy

About 1 � 105/ml of HCT 15 cells were seeded on a cover slip(coated with 0.1% lysine) using RPMI 1640 medium supplementedwith 10% FBS and incubated at 37 8C with 5% CO2 for 24 h. Themedium was changed 1 day after seeding to clear any dead cells onthe cover slip. Then the cover slips with cells were treated with IC50

of either drug (AEE788/Celecoxib) alone or combination for 24 h.After the treatment, all cover slips were removed from incubatorand processed for AFM. In brief, formaldehyde fixed cells werewashed in PBS and dehydrated in series of alcohol gradient (50%,70%, 90%, 95% and 100%) each for 10 min. After dehydration,samples were dipped in HMDS (1,1,1,3,3,3-hexamethyl disilazane)and air dried. It was then mounted and thorough examination ofsurface was performed in tapping mode using atomic forcemicroscope (Veeco CPII, USA) (Li et al., 2008). The tapping modesettings were as follows: 0.5 Hz scan rate with a resolution of256 � 256 data points per scan. AV-shaped silicon nitride

cantilever (MMP-11123, Veeco Instruments Inc., USA) havingspring constant 40 N/m, length 115-135 mm and radii of curvatureless than 10 nm were used.

2.6. Scanning electron microscopy

HCT 15 cells were treated with the IC50 of AEE788 and/orCelecoxib for SEM analysis. Cells were washed three times in 0.1 Mcacodylate buffer (pH 7.4) and post fixed using ice-cold 1% OsO4 for1 h. Cells were then dehydrated with grades (50%, 70%, 95% and100%) of ethanol. Next, the samples were placed in HMDS(1,1,1,3,3,3-hexamethyl disilazane) for 5 min to overcome dryingeffect. Samples were then air dried at room temperature andmounted on stub. Next, they were placed in vacuum chamber ofSEM gold coating apparatus and gold was coated at 2.5 KV, 20–25 mA for 120 s. The morphogram of the HCT 15 were thenobserved using a JEOL JSM-5800 Scanning Microscope using 20 kVacceleration voltages (Hodges, 1972; Glaser et al., 1977).

2.7. Statistical analysis

All the statistical analysis was performed by graphpad prism5software. Data were presented using mean � SD.

3. Results and discussion

3.1. Cell proliferation assay of AEE788 and/or Celecoxib

Cell proliferation assay was used to evaluate the effect ofAEE788 and/or Celecoxib on the proliferation of colon cancer cellHCT 15. Dose-dependent growth inhibitory effects of Celecoxib (A),AEE788 (B-dashed line) and combination (B-thin line) are shown(Fig. 1). Either drug alone or in combination has inhibited theproliferation of HCT 15 in a dose-dependent manner. The growth ofHCT 15 cells in media containing 2% FBS was inhibited by Celecoxibat an IC50 of 102.71 mM (Fig. 1(A)). Further, HCT 15 cells exhibitedAEE788 mediated growth inhibition at an IC50 of 7.509 mM(Fig. 1(B)-dashed line), which was further reduced by the additionof 50 mM Celecoxib (reducing IC50 of AEE788 from 7.509 mM to1.164 mM). AEE788 and/or Celecoxib decreased the HCT 15 cellproliferation by 42%, 38% and 48% respectively.

These results signify antiproliferative activity of AEE788 and/orCelecoxib on HCT 15 cell line. It is inferred from the study that

Fig. 2. Phase contrast microscopic image (Leica 20x) of AEE788 and/or Celecoxib treated HCT 15 for 24 h.

P. Venkatesan et al. / Micron 41 (2010) 247–256250

Celecoxib potentiates the effect of AEE788 towards the coloncancer cell line HCT 15.

3.2. Morphological analysis of HCT 15

Morphology of AEE788 and/or Celecoxib treated HCT 15 wasobserved under phase contrast microscope (Leica 20x). The resultsindicated that no significant morphological changes in individualdrug treated groups with respect to control (Fig. 2). Thecombination treated HCT 15 group has shown slight morphologicalchanges which include retraction of cellular processes and cellshrinkage. In contrast, the control HCT 15 cells have displayed wellspread with a flattened morphology.

But, disadvantages associated with the phase contrast micro-scopy are faint images, inaccurate detection and diffraction limit.Standard optical microscopy used for cell surface visualization haslimitation in resolution (more than 200 nm) due to the diffraction

Fig. 3. Confocal laser fluorescent microscopic image (40�) of Control HCT 15 (single and

with arrow.

limit which control the fine microscopic structure associated withcancer cell/biological samples. Hence, it is difficult to observeanticancer agent mediated fine morphological changes in cancercell using phase contrast microscopy (Ling et al., 2003; Song et al.,2005).

3.3. Confocal laser fluorescence microscopic analysis

In general, nano-microscopic techniques such as confocal andAFM are not being used competently for analyzing cells/tissues.Morphological features and submicron structures of cell obtainedfrom the confocal and AFM image provide advantages overtraditional light microscopic techniques. Examples of such imagesinclude, but are not restricted to cell morphology, cytoskeletalelements, organelles and volume. Consequently, some studies haveutilized the both confocal microscopy and AFM to examinestructure–function relationships in biological systems (Li et al.,

cluster) cell. In the confocal image, healthy filopodia and lamellipodia are marked

P. Venkatesan et al. / Micron 41 (2010) 247–256 251

2008; Scott et al., 1997; Zhou et al., 2008). Although the confocalmicroscopy and AFM are prevailing instruments in analysis ofcells/tissues, more information with finer details may be achievedif combined with various others techniques such as SEM andcellular image processing techniques.

In this study, we have attempted to obtain clear view ofmorphology and actin organization of AEE788 and/or Celecoxibtreated HCT 15 cell by confocal laser fluorescence microscopy. Thecontrol group has displayed elongated shape cells and dense

Fig. 4. Confocal laser fluorescent microscopic image of AEE788 and/or Celecoxib treated c

marked with arrow.

network of actin forming organized parallel filamentous structurein cytoplasm (Fig. 3). After 24 h of treatment with AEE788 (Fig. 4),the cells were observed to be round in shape and, sparse andirregular with no striation in actin filaments organization.However, after 24 h of Celecoxib treatment (Fig. 4), insignificanteffect was observed on cell morphology and actin filamentsorganization. The combination treated cells have appeared roundmorphology and more irregular actin filament organization(Fig. 4).

ell HCT 15 cell. In the confocal image (40�), truncated filopodia and lamellipodia are

Fig. 5. (a) Confocal laser fluorescence image (40�) of control HCT 15, (b and c) segmented cytoplasm and nuclei of (a); (d) confocal laser fluorescence image of

AEE788 + Celecoxib treated HCT 15; (e and f) segmented cytoplasm and nuclei of (d).

Fig. 6. Bar graph showing (a) cytoplasmic shape features of control and

AEE788 + Celecoxib treated HCT 15 (b) nuclei shape features of control and

AEE788 + Celecoxib treated HCT 15 (n = 25 cells from control and each treatment

group were analyzed).

P. Venkatesan et al. / Micron 41 (2010) 247–256252

We further investigated significance of filopodia and lamellipo-dia in HCT 15 cell. Either AEE788 alone or in combination withCelecoxib for 24 h have resulted in loss of filopodia and lamellipodia(Fig. 4) which may signify antimetastatic activity the drugs. On thecontrary, Celecoxib treated HCT 15 displayed partially intactfilopodia and lamellipodia on the cell surface. Both the disorganizedactin and terminated filopodia and lamellipodia in HCT 15 suggestAEE788 and/or Celecoxib target cytoskeletal elements which canfurther be investigated for colon cancer therapy. Previous studieshave reported antimetastatic activity of AEE788 and Celecoxib oncolon cancer cell (Roh et al., 2004; Yokoi et al., 2005).

3.4. Morphological analysis by image processing

Fig. 5(a) and (d) shows original confocal images of control andAEE788 + Celecoxib treated HCT 15 cells respectively. Fig. 5(b), (c),(e) and (f) is showing segmented cytoplasm and nucleusrespectively. Fig. 6(a) and (b) shows bar diagram for basic shapefeatures of cytoplasm and nucleus respectively. These bardiagrams (mean of area, perimeter, compactness and FD) suggestthat there has been a significant difference between the controland drug treated HCT 15 of both cytoplasm and nucleus (n = 25cells from control and each treatment groups were analyzed).

Our study indicates that there has been a decreasing tendency ofcytoplasmic area, perimeter, compactness and eccentricity of drug-treated cells as compared to control. But, cytoplasmic FD values ofdrug-treated cells were shown to increase as compared with controlHCT 15. In addition, the eccentricity of cytoplasm has shownsignificant difference between control and drug treated cells. On thecontrary, the eccentricity of nucleus has not shown significantdifference between control and drug treated cells. These results mayindicate cytotoxicity of the drugs which result in cell shrinkage andmembrane damage. The quantitative analysis of morphologicalchanges further supports confocal laser microscopic analysis ofAEE788 and/or Celecoxib treated colon cancer cell HCT 15.

3.5. AFM analysis

AFM analysis of cancer cell provides valuable information oncytoskeleton and morphology. In this study, we have analyzed the

morphological features of AEE788 and/or Celecoxib treated HCT 15using AFM. The control HCT 15 cell has revealed numerousfilopodia and lamellipodia at periphery of the cell (Fig. 7).

In addition, the control cell has displayed smooth and elongatedin shape with dense filopodia and lamellipodia at terminal part.These ultrastructural parts may signify invasive property of theproliferating cancer cell. This finding has also been supported byour confocal images where numerous filopodia and lamellipodiaare present in the terminal part of the HCT 15 cells (Fig. 3).

The AEE788 treated HCT 15 cell has displayed rough andshrunken cell morphology with truncated lamellipodia at terminal

Fig. 7. Tapping mode AFM image of Control HCT 15, (A and B) 2D image, (C) phase, (D) 3D image of a selected area, respectively. In the figure healthy filopodia and lamellipodia

are marked with arrow.

P. Venkatesan et al. / Micron 41 (2010) 247–256 253

part of the cell (Fig. 8). However, in Celecoxib treatment, less cellshrinkage and reduction in lamellipodia structure were observedas compared to the AEE788 treated cells. It is worth to note that,there are some intact lamellipodia at both body and terminal partof the Celecoxib treated HCT 15 cell. This observation indicatesthat less cytotoxic potency of Celecoxib (Fig. 9) than AEE788 onHCT 15 cell.

Fig. 8. Tapping mode AFM image of AEE788 treated HCT 15. (A and D) are 2D image, (B a

images. The truncated filopodia and lamellipodia are marked with arrow.

AFM image of combination treated HCT 15 have shown roughand shrunken cell morphology with complete loss of both filopodiaand lamellipodia (Fig. 10). This suggests that the combinationtreatment produced additive effect on cell morphology andcytoskeleton. To the best of our knowledge this is the first studyconducted using AEE788 and/or Celecoxib on colon cancer cell HCT15 morphology using AFM. These AFM image substantiate the

nd E) are phase image and (C and F) are 3D image respectively. (D–F) Selected area

Fig. 9. Tapping mode AFM image of Celecoxib treated HCT 15. (A, D and G) 2D image, (B, E and H) phase and (C, F and I) 3D image respectively. The truncated filopodia and

lamellipodia are marked with arrow.

P. Venkatesan et al. / Micron 41 (2010) 247–256254

confocal study of the effect of these two anticancer agents on coloncancer cell line.

3.6. SEM analysis

In the study of cancer therapy, high resolution SEM is a vitaltool for analysis of surface and morphological features of cancer

Fig. 10. Tapping mode AFM image of AEE788 + Celecoxib treated HCT 15. (A) 2D image, (B

arrow.

cell. Though morphological characteristics of normal and cancercells have been well studied using SEM, only few studies havebeen used SEM to find out anticancer agent mediatedmorphological changes in cancer cell (Jacobs et al., 1976).The study was performed to analyze the morphology of AEE788and/or Celecoxib treated colon cancer cell HCT 15. From theobtained SEM image (Fig. 11), the control cells have appeared

) phase, and (C) 3D image. The truncated filopodia and lamellipodia are marked with

Fig. 11. Scanning electron photomicrograph of the surface of colon cancer cell HCT 15 before and after treatment with AEE788 and/or Celecoxib for 24 h. In control HCT 15, the

healthy filopodia, and lamellipodia are marked with short arrow. In AEE788 and/or Celecoxib treated HCT 15, the truncated lamellipodia and filopodia are marked with long

arrow.

P. Venkatesan et al. / Micron 41 (2010) 247–256 255

flat, smooth and large number of slender and filamentous lateralcell wall extension. This observation suggests a highly dynamiccell surface which may reflect an inherent property of motileHCT 15 cells.

The drug treated HCT 15 cell has exhibited thick morphology,small ruffles, irregular retraction of cytoplasm from substratumand less number of cytoplasmic extensions. The cytoplasmicextension has been greatly reduced in AEE788 treated HCT 15 thanCelecoxib. In case of combination treatment, the HCT 15 has showncomplete loss of filopodia and lamellipodia. These results wellsupport the antiproliferative and anti-invasive activity of AEE788and/or Celecoxib on HCT 15.

4. Conclusion

The results suggest that combination treatment inhibits cellgrowth and cell viability more efficiently. The aberration in HCT 15morphology and actin filament organization mediated by AEE788and/or Celecoxib may be attributed to growth inhibitory and anti-invasive activity of the drug. Further, it is believed that thequantitative analysis of cell morphology can be recommended toevaluate the efficacy of anticancer agents along with otherestablished techniques with higher priority. In conclusion, it isthe first report that evaluates effect of AEE788 and/or Celecoxib onhuman colon cancer cell. Analysis of morphological features andalong with intracellular signaling molecule can provide anadditional platform for evaluating anticancer activity of a drugHowever, precise mechanism behind the inhibition of AEE788 andCelecoxib treated colon cancer cell HCT 15 needs furtherinvestigation.

Acknowledgements

We are grateful to Novartis Pharma and Aarthi Drug Ltd., forgenerously providing AEE788 and Celecoxib respectively. Further,we also like to thank Mr. Debashis Gayen for skilled help inconfocal laser fluorescence microscopy. This work was supportedby funds from the School of Medical Science and Technology,Indian Institute of Technology, Kharagpur, India. One of the authorMr. Soumen Das sincerely acknowledges the Council of Scientificand Industrial Research (CSIR) for providing fellowship 9/81(674)/08 – EMR - I, dated 26-03-08).

References

Arico, S., Pattingre, S., Bauvy, C., Gane, P., Barbat, A., Codogno, P., Ogier-Denis, E.,2002. Celecoxib induces apoptosis by inhibiting 3-phosphoinositide dependentprotein kinase-1 activity in the human colon cancer HT-29 Cell line. J. Biol.Chem. 277 (31), 27613–27621.

Auman, J.T., Church, R., Lee, S., Watson, M.A., Fleshman, J.W., Mcleod, H.L., 2008.Celecoxib pre-treatment in human colorectal adenocarcinoma patients is asso-ciated with gene expression alterations suggestive of diminished cellular pro-liferation. Eur. J. Cancer 44, 1754–1760.

Bour-Dill, C., Gramain, M., Merlin, J., Marchal, S., Guillemin, F., 2000. Determinationof intracellular organelles implicated in daunorubicin cytoplasmic sequestra-tion in multidrug-resistant MCF-7 cells using fluorescence microscopy imageanalysis. Cytometry 39, 16–25.

Doak, S.H., Rogers, D., Jones, B., Francis, L., Conlan, R.S., Wright, C., 2008. High-resolution imaging using a novel atomic force microscope and confocal laserscanning microscope hybrid instrument: essential sample preparation aspects.Histochem. Cell Biol. 130, 909–916.

Docheva, D., Padula, D., Popov, C., Mutschler, W., Clausen-Schaumann, H., Schieker,M., 2008. Researching into the cellular shape, volume and elasticity of mesench-ymal stem cells, osteoblasts and osteosarcoma cells by atomic force micro-scopy. J. Cell. Mol. Med. 12 (2), 537–552.

Esgiar, A.N., Naguib, R.N.G., Sharif, B.S., Bennett, M.K., Murray, A., 1998. Microscopicimage analysis for quantitative measurement and feature identification ofnormal and cancerous colonic mucosa. IEEE Trans. Inform. Technol. B 2, 197–203.

Esgiar, A.N., Naguib, R.N.G., Sharif, B.S., Bennett, M.K., Murray, A., 2002. Fractalanalysis in the detection of colonic cancer images. IEEE Trans. Inform. Technol. B6, 54–58.

Gaisera, T., Becker, M.R., Habel, A., Reussc, D.E., Ehemanne, V., Ramid, A., Siegelin,M.D., 2008. TRAIL-mediated apoptosis in malignant glioma cells is augmentedby Celecoxib through proteasomal degradation of surviving. Neurosci. Lett. 442(2), 109–113.

Gil, J., Wu, H., Wang, B.Y., 2002. Image analysis and morphometry in the diagnosis ofbreast cancer. Microsc. Res. Tech. 59, 109–118.

Glaser, R., Muma, V., Farrugia, R., Munger, B., 1977. Scanning electron microscopy ofthe surfaces of hamster embryo cells transformed by herpes simplex virus.Cancer Res. 37, 4420–4422.

Hodges, G.M., 1972. A scanning electron microscope study of the surface feature ofmammalian cells in vitro. J. Cell Sci. 2, 233–247.

Hsiao, P., Chang, C., Liu, H., Tsai, C., Chiu, T.H., Chao, J., 2007. Activation of p38mitogen-activated protein kinase by celecoxib oppositely regulates survivinand gamma-H2AX in human colorectal cancer cells. Toxicol. Appl. Pharmacol.222, 97–104.

Jacobs, J.B., Arai, M., Cohen, S.M., Friedell, G.H., 1976. Early lesions in experimentalbladder cancer: scanning electron microscopy of cell surface markers. CancerRes. 36, 2512–2517.

Jeong, H., Kim, T., Hwang, H., Choi, H., Park, H., Choi, H., 2005. Comparison ofthresholding methods for breast tumor cell segmentation. IEEE 392–395.

Kim, S., Schiff, B.A., Yigitbasi, O.G., Doan, D., Jasser, S.A., Bekele, B.N., Mandal, M.,Myers, J.N., 2005. Targeted molecular therapy of anaplastic thyroid carcinomawith AEE788. Cancer Ther. 4 (4), 632–640.

Kuwai, T., Nakamura, T., Sasaki, T., Kitadai, Y., Kim, J., Langley, R.R., Fan, F., Wang, X.,Do, K., Kim, S., Fidler, I.J., 2008. Targeting the EGFR, VEGFR, and PDGFR on coloncancer cells and stromal cells is required for therapy. Clin. Exp. Metastasis 25,477–489.

P. Venkatesan et al. / Micron 41 (2010) 247–256256

Lee, H., Lee, K., Park, T.G., 2008. Hyaluronic acid paclitaxel conjugate micelles:synthesis, characterization, and antitumor activity. Bioconjug. Chem. 19 (6),1319–1325.

Li, Q.S., Lee, G.Y.H., Ong, C.N., Lim, C.T., 2008. AFM indentation study of breast cancercells. Biochem. Biophys. Res. Commun. 374, 609–613.

Linda, E., Malick, L., Robert, L., 1976. Scanning electron microscopy of in vitrochemically transformed mouse embryo cells scanning electron microscopy of invitro chemically transformed mouse embryo cells. J. Cell Biol. 68, 654–664.

Ling, Y., Liebes, L., Zou, Y., Perez-Soler, R., 2003. Reactive oxygen species generationand mitochondrial dysfunction in the apoptotic response to bortezomib, a novelproteasome inhibitor, in human H460 non-small cell lung cancer cells. J. Biol.Chem. 278 (36), 33714–33723.

Lin, J., Hsiao, P., Chiu, T.H., Chao, J., 2005. Combination of cyclooxygenase-2inhibitors and oxaliplatin increases the growth inhibition and death in humancolon cancer cells. Biochem. Pharmacol. 70, 658–667.

Liu, B., Shi, Z., Feng, J., Tao, H., 2008. Celecoxib, a cyclooxygenase-2 inhibitor, inducesapoptosis in human osteosarcoma cell line MG-63 via down-regulation of PI3K/Akt. Cell Biol. Int. 32, 494–501.

Maier, T.J., Schilling, K., Schmidt, R., Geisslinger, G., Grosch, S., 2004. Cyclooxygen-ase-2 (COX-2)-dependent and independent anticarcinogenic effects of cele-coxib in human colon carcinoma cells. Biochem. Pharmacol. 67, 1469–1478.

Marcel, N.A., Bijman, C.A., Hermelink, M.P.A., Berkel, V., Laan, A.C., Janmaat, M.L.,Peters, G.J., Boven, E., 2008. Interaction between celecoxib and docetaxel orcisplatin in human cell lines of ovarian cancer and colon cancer is independentof COX-2 expression levels. Biochem. Pharmacol. 75, 427–437.

Oberleithner, H., Giebisch, G., Geibel, J., 1993. Imaging the lamellipodium ofmigrating epithelial cells in vivo by atomic force microscopy. Pflugers Arch.425, 506–510.

Rebhun, R.B., Langley, R.R., Yokoi, K., Fan, D., Gershenwaldy, J.E., Fidler, I.J., 2006.Targeting receptor tyrosine kinase on lymphatic endothelial cells for thetherapy of colon cancer lymph node metastasis. Neoplasia 8 (9), 747–757.

Roh, J., Sung, M., Park, S., Heo, D., Lee, D.W., Kim, K.H., 2004. Celecoxib can preventtumor growth and distant metastasis in postoperative setting. Cancer Res. 64,3230–3235.

Sakoguchi-Okada, N., Takahashi-Yanaga, F., Fukada, K., Shiraishi, F., Taba, Y., Miwa,Y., Morimoto, S., Iida, M., Sasaguri, T., 2007. Celecoxib inhibits the expression ofsurvivin via the suppression of promoter activity in human colon cancer cells.Biochem. Pharmacol. 73, 1318–1329.

Sasaki, T., Kitadai, Y., Nakamura, T., Kim, J., Tsan, R.Z., Kuwai, T., Langley, R.R., Fan, D.,Kim, S., Fidler, I.J., 2007. Inhibition of epidermal growth factor receptor andvascular endothelial growth factor receptor phosphorylation on tumor-asso-ciated endothelial cells leads to treatment of orthotopic human colon cancer innude mice. Neoplasia 9 (12), 1066–1077.

Scott, S., Schaus, R., Eric, R., 1997. Henderson cell viability and probe-cell membraneinteractions of XR1 glial cells imaged by atomic force microscopy. Biophys. J. 73,1205–1214.

Shahbazi, M., Sattarib, M., Ghazi, M., 2008. Automatic polyp detection from CTcolonography using mathematical morphology. Int. Arch. PhotogrammetryRemote Sens. Spat. Inform. Sci. 37 (B5), 823–828.

Song, G., Mao, Y.B., Cai, Q.F., Yao, L.M., Ouyang, G.L., Bao, S.D., 2005. Curcumininduces human HT-29 colon adenocarcinoma cell apoptosis by activating p53and regulating apoptosis-related protein expression. Braz. J. Med. Biol. Res. 38(12), 1791–1798.

Wanga, L., Li, Q., Zua, Y., Fua, Y., Chena, L., Lva, H., Yaoa, L., Jianga, S., 2008. Anti-proliferative and pro-apoptotic effect of CPT13, a novel camptothecin analog, onhuman colon cancer HCT8 cell line. Chem. Biol. Interact. 176, 165–172.

Yokoi, M.K., Thaker, P.H., Yazici, S., Rebhun, R.R., Nam, D., He, J., Kim, S., Abbruzzese,J.L., Hamilton, S.R., Fidler, I.J., 2005. Dual inhibition of epidermal growth factorreceptor and vascular endothelial growth factor receptor phosphorylation byAEE788 reduces growth and metastasis of human colon carcinoma in anorthotopic nude mouse. Cancer Res. 65, 3716–3725.

You, H.X., Lau, J.M., Zhang, S., Yu, L., 2000. Atomic force microscopy imaging of livingcells: a preliminary study of the disruptive elect of the cantilever tip on cellmorphology. Ultramicroscopy 82, 297–305.

Yu, H., Friday, B.B., Lai, J., McCollum, A., Atadja, P., Roberts, L.R., Adjei, A.A., 2007.Abrogation of MAPK and Akt signaling byAEE788 synergistically potentiateshistone deacetylase inhibitor-induced apoptosis through reactive oxygen spe-cies generation. Clin. Cancer Res. 13 (4), 1140–1148.

Younes, M.N., Yigitbasi, O.G., Park, Y.W., Kim, S., Jasser, S.A., Hawthorne, V.S., Yazici,Y.D., Mandal, M., Bekele, B.N., Bucana, C.D., Fidler, I.J., Myers, J.N., 2005. Anti-vascular therapy of human follicular thyroid cancer experimental bone metas-tasis by blockade of epidermal growth factor receptor and vascular growthfactor receptor phosphorylation. Cancer Res. 65, 4716–4727.

Zhou, D., Jiang, X., Xu, R., Cai, Y., Hu, X., Xu, G., Zou, Y., Zeng, Y., 2008. Assessing thecytoskeletal system and its elements in C6 glioma cells and astrocytes by atomicforce microscopy. Cell. Mol. Neurobiol. 28, 895–905.