Quantitative method for in vitro matrigel invasivenessmeasurement through image analysis software

Gabriel Gallo-Oller • Juan A. Rey •

Javier Dotor • Javier S. Castresana

Received: 16 December 2013 / Accepted: 20 June 2014 / Published online: 3 July 2014

� Springer Science+Business Media Dordrecht 2014

Abstract The determination of cell invasion by matrigel

assay is usually evaluated by counting cells able to pass

through a porous membrane and attach themselves to the

other side, or by an indirect quantification of eluted specific

cell staining dye by means of optical density measurement.

This paper describes a quantitative analytical imaging

approach for determining the invasiveness of tumor cells

using a simple method, based on images processing with

the public domain software, ImageJ. Images obtained by

direct capture are split into the red channel, and the gen-

erated image is used to measure the area that cells cover in

the picture. To overcome the several disadvantages that

classical cell invasion determinations present, we propose

this method because it generates more accurate and sensi-

tive determinations, and it could be a reasonable option for

improving the quality of the results. The cost-effective

alternative method proposed is based on this simple and

robust software that is worldwide affordable.

Keywords Image analysis � Invasion � Matrigel � ImageJ

Introduction

Cell invasion requires a cell to migrate through an extracel-

lular matrix or basement membrane extract and is a pivotal

process to achieving functions such as wound repair, cell

differentiation, angiogenesis, embryonic development, tumor

invasion and metastasis. Cell invasion involves important

mechanisms for a wide array of biological processes exhibited

by normal cells and by metastatic tumor cells. Tumor cell

invasion into the surrounding tissues and metastasis to distant

sites of the body constitute the most lethal effects of solid

tumors (glioblastoma, melanoma, prostate, breast, bladder,

liver and lung cancer among others) [1–4]. These two char-

acteristics, invasion and metastasis, are major causes of

treatment failure and death for cancer patients [5] and involve

complex interactions between cancer cells and their extra-

cellular environment. They include multistep processes

involving cell adhesion, motility, and enzyme-dependent

infiltration of extracellular matrix [1, 2, 6].

The in vitro quantitative assessment of tumor cell trans-

migration across artificial anatomic barriers is a useful and

widespread tool in the study of tumor cell invasion capacity

[7]. The most widely used method consists of a porous filter

used as a barrier to determine the ability of cells to penetrate

and migrate to the opposite side of the filter [6]. This widely

used ‘‘invasion assay’’ requires wells with membrane bot-

toms, called inserts, was first described in 1986 by Terranova

et al. [8] and then standardized by Albini et al. [9]. In this

model, a matrix of solidified basement membrane compo-

nents (such as matrigel) is required, allowing the evaluation

of adhesion, local degradation and/or locomotion into the

region of the matrix [5, 9].

G. Gallo-Oller � J. S. Castresana

Brain Tumor Biology Unit-CIFA, University of Navarra School

of Sciences, Pamplona, Spain

J. A. Rey

IdiPaz Research Unit, La Paz University Hospital, Madrid, Spain

J. Dotor (&)

Digna Biotech, Orense 85, Edificio Lexington, 28020 Madrid,

Spain

e-mail: jdotor@unav.es

J. S. Castresana (&)

Department of Biochemistry and Genetics, University of

Navarra School of Sciences, Irunlarrea 1, 31008 Pamplona,

Spain

e-mail: jscastresana@unav.es

123

Mol Biol Rep (2014) 41:6335–6341

DOI 10.1007/s11033-014-3556-0

The two main types of analysis in matrigel invasiveness

are direct quantification by counting transmigrated cells,

and indirect quantification of the eluted cell staining dye

and subsequent optical density (OD) measurement [6].

Several limitations have been found in the first approach.

Firstly, the quantitative measurement is carried out manu-

ally by visually counting the cells that have migrated

through the matrix, something which is subjective, time-

consuming and laborious due to the large, single filter area

and the multi-chambers and replicates to be evaluated.

Secondly, the sampling of a single filter often yields fluc-

tuating results due to the non-uniformed distribution and

overlapping of migrated cells, resulting in cell aggregates.

Thirdly, the percentage of migrated cells varies, depending

on the number of cells initially placed on the upper surface;

this makes comparison of published results a great chal-

lenge [6, 7, 10].

In spite of some drawbacks, cell counting is probably the

most sensitive and accurate method of quantifying cell

number. However, the difficulties in scoring total cell num-

ber by visual counting make it time-consuming and tedious

for investigators to determine a true percentage of migration

and/or invasion [11, 12]. In 1993 Muir et al. [6] proposed the

indirect test for assessing cell invasion, based on OD mea-

surements of the eluted cell staining dye. In some cases, and

based on our experience, this test might present lower sen-

sitivity. For example, the need to correct for high background

due to unspecific dyes binding to the matrigel, resulted in

slightly lower values [6]. Although increasing number of

replicates and the use of multichamber formats have been

included as partial solutions for solving the limitations of the

indirect approach, these strategies can be replaced by rapid

and semiautomatic methods for assessing cell number in

filter-based assays [6, 7].

Therefore, a third approach for improving cell invasion

determination might be a computer-assisted image analysis

system which could guarantee a nonsubjective and efficient

quantification. The use of software or computer devices

provides a more objective quantification, and generally

highly reproducible results [13]. The computerized image

analysis allows the transformation of a qualitative percep-

tion into quantitative parameters, improving the capability

of measurement over large sets of images, reducing inter

and intra observer variations, and increasing the robustness

of the technique [13, 14].

There are a good number of free software systems that

are often applied in biomedicine which do not require

advanced computer skills. One of them, ImageJ, is a Java-

based image-processing and analysis program, designed at

the National Institutes of Health. ImageJ is already avail-

able as freeware and is widely used in several scientific

fields [15]. The use of ImageJ and other image analysis

programs can be found elsewhere, including area

calculation, particle counting, shape and size analysis, dot

blot analysis, color analysis, and other applications [15–

21]. Using the ‘‘area method’’, described by Girish et al.

[22], the total area occupied by stained cells can be selected

using ImageJ’s thresholding tool.

ImageJ is a well-known and useful program for biology

image analysis [23], and several reports indicate their use

for invasion assay analysis, although no protocol descrip-

tion has been made [24–26]. The aim of this paper is to

describe the use of ImageJ software and its application for

assessing in vitro cell invasion. With this purpose in mind,

we have carried out experiments in order to show inva-

siveness by matrigel. We compared the indirect quantifi-

cation by elution of cell dye and further OD measurement,

versus the direct image analysis with ImageJ.

Materials and methods

Cell lines and cell culture

Two glioblastoma-derived cell lines were used in this work:

A172 and U87-MG, both provided by the American Type

Culture Collection (ATCC, Manassas, VA). A172 was grown

with RPMI/GlutaMAXTM medium, and U87-MG with Dul-

beccos modified Eagles medium/GlutaMAXTM plus nones-

sential aminoacids at 5 %. Both media were supplemented

with heat-inactivated 10 % fetal bovine serum, 1 % penicil-

lin/streptomycin and 0.1 % amphotericin B. All cell lines and

assays were grown and performed, respectively, at 37 �C in a

humidified atmosphere of 5 % CO2.

Matrigel invasion assay

Matrigel concentrations were optimized based on the

migration capability of each cell line and according to a

previous in vitro assay optimization process (data not

shown). Final concentrations of 2.7 or 1.6 lg/ll of Matrigel

(BD Matrigel TM Basement Membrane Matrix Growth

Factor Reduced, BD Biosciences, Bedford, MA) for A172 or

U87-MG cell lines, were respectively adjusted in the upper

face of every insert (24-Millicell Culture Plate Filter Inserts,

0.8 lm PET, Millipore, Billerica, MA, USA), and incubated

at 37 �C for 30 min until solidified. Cells (2.5 9 105) were

resuspended in their respective culture media supplemented

with 0.5 % FBS in the absence (control group) or presence

(treated group) of 100 lg/ml of P144, a TGF-b inhibitor

peptide [27, 28], and then added to the upper chamber. The

lower reservoirs contained a medium supplemented with

10 % FBS. After 24 h of incubation, cells on the upper sur-

face of the membranes were removed with cotton swabs, and

cells on the lower surface of the membranes were fixed and

stained with crystal violet (PANREAC, Castellar del Valles,

6336 Mol Biol Rep (2014) 41:6335–6341

123

Barcelona, Spain) at 0.5 %. To cover most of the filter area at

1009 magnification (109 plan objective and 109 ocular

lenses), 10–15 images of each filter were acquired with an

inverted light microscope NIKON ECLIPSE TS100 (NI-

KON, Melville, NY, USA) with Digital SGHT DS-LS

camera (NIKON, Melville, NY, USA). Then, the solubilized

dye was measured on a microplate reader Spectra MR

(DYNEX Technologies, Chantilly, VA, USA) at 560 nm.

Three independent experiments were performed for each

condition.

TGF-b inhibitors have been widely proposed and are

being developed as antitumoral therapies in patients with

cancer [29]. For this reason we decided to use an available

and well know TGF-b inhibitor peptide (P144) with

established antimetastatic activity [13]. In the context of

this work P144 is used as a cell invasiveness inhibitor to

check for the sensitivity of the analytical method.

Image analysis

This method was based on a similar previous procedure

developed for stained liver tissue quantification, and it was

first applied to invasiveness assays in this work. Image

analysis was performed using the public domain, free soft-

ware ImageJ (NIH, Bethesda, Maryland, USA; http://rsb.

info.nih.gov/ij/), a Java application which runs in most

operative systems. A modified macro for quantifying stained

liver tissue (http://rsbweb.nih.gov/ij/docs/examples/stained-

sections/index.html) was applied. The images in JPEG for-

mat were first split into red, green and blue channels using the

RGB stack command. Once the blue and green channels

were discarded, the images were performed through red

channel, since this channel showed less background levels.

Threshold was manually adjusted until only stained cells

were highlighted in red. Next, Set Measurements dialogue

Fig. 1 Image processing steps

Mol Biol Rep (2014) 41:6335–6341 6337

123

was checked for: ‘‘Area’’, ‘‘Area Fraction’’, ‘‘Limit to

Threshold’’ and ‘‘Display Label’’. The obtained data quan-

tify the area and the area percentage of the image covered by

stained cells. After adjustment of conditions for invasiveness

quantification in the first image, subsequent images require

minimal threshold tuning to compensate light and back-

ground variability among images (Fig. 1).

Statistical analysis

The data were expressed as the mean ± standard deviation

of triplicated wells per group from at least three independent

experiments. According the normal distribution, data were

analyzed by Mann–Whitney U test or t test. All the statistical

analysis as well as Pearsons correlation coefficient and the

Bland–Altman analysis were generated using GraphPad

Prism for Windows version 5.04. Differences were consid-

ered significant when the p-value was lower than 0.05.

Results and discussion

Invasiveness by colorimetric determination

Invasion assays were performed using 24-Millicell Culture

Plate Filter Inserts, 0.8 lm PET (Millipore, Billerica, MA,

USA) coated with matrigel. After 24 h, invading cells

attached to the lower surface of the membranes were

indirectly measured by elution of the dye. The number of

cells that migrated to the lower surface could not be

determined by counting, due to non-uniform distribution of

the migrated cells, some cell aggregation, and overlapping

of the tested cell lines.

In the indirect measurement, a reduction of invasiveness

of 35.45 % (p = 0.0305) and 27.22 % (p = 0.0016) for

A172 and U87-MG, respectively, compared to controls was

observed (Fig. 2). These observations are consistent with

results from previous work (data not shown), where P144

always produced a 25–30 % of invasiveness reduction in

these cell lines.

Image analysis by ImageJ

Direct measurement was assessed using a variation of the

protocol for quantifying stained liver tissue. Dyed cells

were highlighted with the threshold function and then the

red area was measured, corresponding to the cells that

migrated to the lower surface of the membrane. Membrane

was photographed to cover most of its extension, then the

cell dye was eluted with acetic acid 10 % for the indirect

measurement. Using the free software ImageJ as previously

described, the area corresponding to the migrated cells in a

specific scale for each picture of the insert’s membrane was

obtained (Fig. 1). Following the direct method with Ima-

geJ, the results showed a major and accurate decrease in

the percentage of invasiveness when comparing with the

results obtained by the indirect method (Fig. 2). In both

cell lines, the decrease was close to 70 % and showed

statistical differences for A172 (p = 0.0043) and U87-MG

(p = 0.0317).

In previous determinations, we noticed that the data

obtained by OD measurement did not always correspond

with the direct microscopic observations, and showed dif-

ferences undetected by the indirect measurement. To con-

firm this point, the invasiveness assay was performed at

increasing concentrations of P144 inhibitor peptide in cell

line A172, and the results were obtained by both, the

indirect and direct method. The sensitivity curve demon-

strated the improvement in the analysis of invasiveness

using ImageJ analysis with respect to the indirect method.

In all the tested concentrations, ImageJ determined higher

values of invasiveness than the indirect method (Fig. 3),

coinciding better with direct visual appreciations (Fig. 4),

even at low concentrations of P144.

Since the threshold adjustment in the analysis proposed

could be considered subjective, we performed an assay in

order to analyze the intra-experimenter and inter-experi-

menter variability. We randomly selected 15 images that

were analyzed twice by experimenter 1 and once by

experimenter 2, following the proposed protocol. The data

and analyses are summarized in Fig. 5. The Pearsons

Fig. 2 Invasiveness in U87-MG and A172 cells lines. Indirect

quantification by colorimetry (a) and direct measurement with ImageJ

(b) are represented. The data represents the percentage of invasive-

ness compared to the control (100 %) and the cell line treated with

P144. The percentage represents the reduction of invasiveness under

P144 treatment

6338 Mol Biol Rep (2014) 41:6335–6341

123

coefficient showed correlation between the two results

obtained by the same experimenter (p \ 0.0001), coincid-

ing with no statistical differences between the two media

obtained (p [ 0.05). The Bland–Altman analysis showed

the concordance between the two sets of observations (data

not shown). Respect to the inter-experimenter variation; the

data obtained by two different experimenters showed

similar results. Taken together, resulting data showed that

the method is not significantly affected by variations due to

intra and inter-experimenters analysis. Considering the

automation option offered by IMAGEJ, these external

variation factors could even be substantially diminished.

Probably the most widely used method for in vitro

migration/invasion analysis is the modified Boyden

Chamber, due to the plasticity of the assay incorporating

changes and variations that can be adapted to specific

requirements [10–12]. We already summarized the differ-

ent limitations of this technique [6, 7, 10]: from protocol

optimization for each cell line to the selection of a specific

measurement method according to the availability of the

resources. The data analysis obtained with this approach

can be processed in different ways. In this work, we

measured invasion using the basic protocol of the modified

Boyden chamber, coating the membrane with matrigel, and

carrying out two different analyses: an indirect colorimetric

assay and an image analysis using the ImageJ software.

We demonstrated that the new approach by ImageJ

introduces an improvement in the quality and sensitivity of

cell invasion data. In our first assays, we already consid-

ered that the dye elution protocol introduced some vari-

ability in our results, as the decrease in the percentage of

cell invasion did not correspond with visual observations

(Fig. 4). We selected an indirect assay based on OD

determination of a diluted cell staining dye because (1) we

could not perform direct counting of transmigrated cells

due cell aggregation, and (2) this OD method is the most

widely used determination method, due to the fact that

other indirect methods are more time-consuming and

require specific equipment [11].

As we expected, the data obtained with ImageJ always

showed remarkable differences between cell invasion val-

ues for control and P144-treated groups. The relative

invasion values dropped from 70–74 to 27–35 % in the

same inserts analyzed in the P144-treated cells. To confirm

this observation, we carried out a concentration–response

sensitive curve to discard possible effects due to the pep-

tide concentration used. Again, the analysis with ImageJ

showed a lower value of invasion in all cases, supporting

optimized method accuracy.

Despite the fact that our initial measurements were

based on direct observation (Fig. 4), the incorporation of

ImageJ analysis improved our results. When a direct

Fig. 3 Dose–response sensitivity curve in A172 cell line. The

comparison between the two different methods used is represented.

The determination by ImageJ always found more accurate data than

the OD 560 nm determination, in all P144 concentrations. The

logarithm scale of the P144 concentrations tested are graphed on the

X axis. The percentage represents the reduction of invasiveness

compared to the control. Each point represents the mean of at least

three independent experiments. The dotted line represents 100 % of

the control value

Fig. 4 Direct visual observation. Representative areas of transmigrated cells in the lower face of inserts in control (a) and P144-treated (b) cell

lines. Direct observation by optic microscopy evidences more remarkable differences than by OD measurement

Mol Biol Rep (2014) 41:6335–6341 6339

123

determination such as cell counting is not feasible, the use

of indirect methods is the most logical option, even though

this selection is conditioned by the resources available in

the laboratories.

The use of image analysis software for research is

constantly increasing due to the fact it improves the quality

of data analysis [13, 16–19, 21, 22, 30]. ImageJ is widely

used, and it is a relatively user-friendly free resource [23],

reasons that turn it into a good choice for direct determi-

nation of invasion on matrigel assays when other methods

or software are not available.

The versatility of modified Boyden chamber to condi-

tion variability, such as the addition of multiple growth

factors, and the availability of different measuring tech-

niques, makes this protocol the first choice for determining

effects on invasiveness in different study conditions [10,

11]. However, when limitations such as equipment, costs,

or inherent cell line characteristics -such as cell aggrega-

tion-appear, an evaluation of a new approach is required.

The use of ImageJ allows us to detect more remarkable

differences in comparison with the widely used indirect

method (OD determination of the eluted cell staining dye)

and it also avoids the need for more complex determina-

tions such as cell labelling with fluorescent dyes [11].

Furthermore, ImageJ is a public domain software that does

not require complicated manual adjustments, and the

optimization of the technique is compatible with standard

laboratory equipment and resources.

In conclusion, software tools for image analysis generally

require computer skills, and the commercial packages are

generally expensive or at least not available for free. Here we

have described an additional application for the ImageJ

software that increases accuracy and improved data mea-

surement for matrigel invasion analysis, and provides an

achievable approach when different technique complica-

tions arise, or when technical resources are limited. Even, the

ImageJ software gives the researchers the option of devising

an automatic protocol, allowing the analysis of a big set of

images. Although we have not included this automation

setting in the present work, the possibility is left open for

useful laboratory in-house tools developments.

Acknowledgments Authors are grateful to Laura Stokes for help

with editing the manuscript. G. Gallo-Oller was supported by a fel-

lowship from the Department of Education of the Government of

Navarra, Pamplona, Spain. This research was supported in part by

grants from the Department of Health of the Government of Navarra,

Caja Navarra (Project 13912), Fundacion Universitaria de Navarra,

Pamplona; and Fondo de Investigacion Sanitaria (PI-081849, to JSC;

and PI-101972, to JAR), Madrid.

Conflict of interest The authors declare no conflict of interest

related to this work.

References

1. Elliott RL, Blobe GC (2005) Role of transforming growth factor

Beta in human cancer. J Clin Oncol 23:2078–2093

2. Onishi M, Ichikawa T, Kurozumi K, Date I (2011) Angiogenesis

and invasion in glioma. Brain Tumor Pathol 28:13–24

3. Valster A, Tran NL, Nakada M, Berens ME, Chan AY, Symons

M (2005) Cell migration and invasion assays. Methods

37:208–215

Fig. 5 Intra and inter-

experimenter variations

analysis. Correlation analysis

and media comparison for intra-

experimenter (a and b,

respectively) and for inter-

experimenter (c and d,

respectively) analysis. The

p and R2 values are shown in the

correlation plots (a and d). The

media comparison did not show

significance for (b) the variation

intra-experimenter

(b) (p = 0.9315) neither inter-

.experimenter (d) (p = 0.6718)

6340 Mol Biol Rep (2014) 41:6335–6341

123

4. Koul HK, Pal M, Koul S (2013) Role of p38 MAP kinase signal

transduction in solid tumors. Genes Cancer 4:342–359

5. Sieuwerts AM, Klijn JG, Foekens JA (1997) Assessment of the

invasive potential of human gynecological tumor cell lines with

the in vitro Boyden chamber assay: influences of the ability of

cells to migrate through the filter membrane. Clin Exp Metastasis

15:53–62

6. Muir D, Sukhu L, Johnson J, Lahorra MA, Maria BL (1993)

Quantitative methods for scoring cell migration and invasion in

filter-based assays. Anal Biochem 215:104–109

7. Li YH, Zhu C (1999) A modified Boyden chamber assay for

tumor cell transendothelial migration in vitro. Clin Exp Metas-

tasis 17:423–429

8. Terranova VP, Hujanen ES, Loeb DM, Martin GR, Thornburg L,

Glushko V (1986) Use of a reconstituted basement membrane to

measure cell invasiveness and select for highly invasive tumor

cells. Proc Natl Acad Sci USA 83:465–469

9. Albini A, Iwamoto Y, Kleinman HK, Martin GR, Aaronson SA,

Kozlowski JM, McEwan RN (1987) A rapid in vitro assay for

quantitating the invasive potential of tumor cells. Cancer Res

47:3239–3245

10. Hulkower KI, Herber RL (2011) Cell migration and invasion

assays as tools for drug discovery. Pharmaceutics 3:107–124

11. Kirilova A, Lockwood G, Choi P, Bana N, Haider MA, Brock

KK, Eccles C, Dawson LA (2008) Three-dimensional motion of

liver tumors using cine-magnetic resonance imaging. Int J Radiat

Oncol Biol Phys 71:1189–1195

12. Cassola VF, Lima VJ, Kramer R, Khoury HJ (2010) FASH and

MASH: female and male adult human phantoms based on poly-

gon mesh surfaces: I. Development of the anatomy. Phys Med

Biol 55:133–162

13. Prasad K, Prabhu GK (2012) Image analysis tools for evaluation

of microscopic views of immunohistochemically stained speci-

men in medical research—a review. J Med Syst 36:2621–2631

14. Tourassi GD (1999) Journey toward computer-aided diagnosis:

role of image texture analysis. Radiology 213:317–320

15. Bearer EL (2003) Overview of image analysis, image importing,

and image processing using freeware. Curr Protoc Mol Biol

Chapter 14:Unit 14 15

16. Prodanov D, Heeroma J, Marani E (2006) Automatic mor-

phometry of synaptic boutons of cultured cells using granulo-

metric analysis of digital images. J Neurosci Methods

151:168–177

17. Carmona R, Macias D, Guadix JA, Portillo V, Perez-Pomares JM,

Munoz-Chapuli R (2007) A simple technique of image analysis

for specific nuclear immunolocalization of proteins. J Microsc

225:96–99

18. McAtee PA, Hallett IC, Johnston JW, Schaffer RJ (2009) A rapid

method of fruit cell isolation for cell size and shape measure-

ments. Plant Methods 5:5

19. Safadi RA, Musleh AS, Al-Khateeb TH, Hamasha AA (2010)

Analysis of immunohistochemical expression of k19 in oral

epithelial dysplasia and oral squamous cell carcinoma using color

deconvolution-image analysis method. Head Neck Pathol

4:282–289

20. Cai Z, Chattopadhyay N, Liu WJ, Chan C, Pignol JP, Reilly RM

(2011) Optimized digital counting colonies of clonogenic assays

using ImageJ software and customized macros: comparison with

manual counting. Int J Radiat Biol 87:1135–1146

21. Kim US, Kim SJ, Baek SH, Kim HK, Sohn YH (2011) Quanti-

tative analysis of optic disc color. Korean J Ophthalmol

25:174–177

22. Girish V, Vijayalakshmi A (2004) Affordable image analysis

using NIH Image/ImageJ. Indian J Cancer 41:47

23. Collins TJ (2007) ImageJ for microscopy. Biotechniques

43:25–30

24. Hood JL, Pan H, Lanza GM, Wickline SA (2009) Paracrine

induction of endothelium by tumor exosomes. Lab Invest

89:1317–1328

25. Kitzing TM, Wang Y, Pertz O, Copeland JW, Grosse R (2010)

Formin-like 2 drives amoeboid invasive cell motility downstream

of RhoC. Oncogene 29:2441–2448

26. Kam Y, Karperien A, Weidow B, Estrada L, Anderson AR,

Quaranta V (2009) Nest expansion assay: a cancer systems

biology approach to in vitro invasion measurements. BMC Res

Notes 2:130

27. Ezquerro IJ, Lasarte JJ, Dotor J, Castilla-Cortazar I, Bustos M,

Penuelas I, Blanco G, Rodriguez C, Lechuga Mdel C, Greenwel

P, Rojkind M, Prieto J et al (2003) A synthetic peptide from

transforming growth factor beta type III receptor inhibits liver

fibrogenesis in rats with carbon tetrachloride liver injury. Cyto-

kine 22:12–20

28. Llopiz D, Dotor J, Casares N, Bezunartea J, Diaz-Valdes N, Ruiz

M, Aranda F, Berraondo P, Prieto J, Lasarte JJ, Borras-Cuesta F,

Sarobe P (2009) Peptide inhibitors of transforming growth factor-

beta enhance the efficacy of antitumor immunotherapy. Int J

Cancer 125:2614–2623

29. Akhurst RJ, Hata A (2012) Targeting the TGFbeta signalling

pathway in disease. Nat Rev Drug Discov 11:790–811

30. Maximova OA, Taffs RE, Pomeroy KL, Piccardo P, Asher DM

(2006) Computerized morphometric analysis of pathological

prion protein deposition in scrapie-infected hamster brain. J His-

tochem Cytochem 54:97–107

Mol Biol Rep (2014) 41:6335–6341 6341

123