A mesenchymal stromal cell line resistant to paclitaxel that spontaneously differentiates into...

A mesenchymal stromal cell line resistant to paclitaxelthat spontaneously differentiates into osteoblast-like cells

Augusto Pessina & Francesca Sisto & Valentina Coccè & Loredana Cavicchini &Emilio Ciusani & Laura Gribaldo & Arianna Bonomi

Received: 14 August 2010 /Accepted: 9 December 2010 /Published online: 29 December 2010# Springer Science+Business Media B.V. 2010

Abstract The mesenchymal stromal cell line SR-4987 has been established in our laboratory from thebone marrow of BDF/1 mice. Recent information onmesenchymal stem cells biology and the need to dealwith well-characterized cell lines suggest to criticallyconsider the existent data on this cell line by updatingthem with new investigations on growth parameters,in vitro plasticity, and drug sensitivity to anti-cancer,anti-inflammatory, and a histone deacetylase inhibitor.SR-4987 cells show a population doubling time of24.5±5.4 h, a plating efficiency of 2.87±1.19%, andunder stimulation maintain only in part their multi-potency by differentiating towards chondro-osteogeniclineages but not into adipogenic. Surprisingly, thesemesenchymal stromal cells differentiate spontaneouslyinto osteoblast-like cells and this is significantlystimulated by valproic acid. SR-4987 cells show a

dramatic resistance to paclitaxel (PTX) with aresistance index of 39.6 times (evaluated versusMOLT-4 leukemia) and of 68.2 (versus HT-29colorectal carcinoma). SR-4987 resistance is reversedby verapamil and correlates with high expression of P-glycoprotein that is down-modulated by PTX. Takentogether, our results indicated that SR-4987 line is avery interesting cell model useful to investigate bothdrug sensitivity resistance and physiopathologicalaspects related to mesenchymal cell function.

Keywords Drugsensitivity . Mesenchymalstemcells .

Osteogenesis . Paclitaxel resistance . SR-4987

Introduction

The presence of fibroblast-like cells, termed colony-forming unit—fibroblast (CFU-F) in the bone mar-row, described since many years by Friedenstein et al.(1966), appeared as cells capable of limited self-renewing but able to differentiate into variousconnective tissue lineages. When these cells havebeen better studied and characterized, they resultedconstituted by a homogeneous cell population capableof extensive self-renewal and with a wide range ofdifferentiation potential (e.g., bone, cartilage, tendon,adipocytes, muscle, hepatocytes) if stimulated withappropriate factors and were named mesenchymalstromal cells (MSCs) (Haynesworth et al. 1992). Therecent progress in using MSCs in regenerative

Cell Biol Toxicol (2011) 27:169–180DOI 10.1007/s10565-010-9179-x

A. Pessina (*) : F. Sisto :V. Coccè : L. Cavicchini :A. BonomiDepartment of Public Health–Microbiology–Virology,University of Milan,Via Pascal 36,20133 Milan, Italye-mail: [email protected]

E. CiusaniFondazione IRCCS Istituto Neurologico Besta,Milan, Italy

L. GribaldoIHCP, Joint Research Centre,Ispra, VA, Italy

medicine confirms the high plasticity degree of thesecells, arising also questions on their biological activitythat need to be more investigated also by using invitro experimental models. A murine MSC line (SR-4987) originated and established in our laboratoryfrom adherent cells of a long-term bone marrowculture (Pessina et al. 1992) has a fibroblast like-morphology, is very susceptible to the clonogeniceffect of FGFs, and, very singularly, together withtypical characters of stromal cells (as the productionof M-CSF), expresses also B cell markers. Moreover,it is tumorigenic and able to produce sarcomas insyngeneic mice (Pessina et al. 1997). This line hasbeen used in porous carriers to construct a three-dimensional hematopoietic culture system (Takagiet al. 1999; Takagi 2005) and for gene transfectionstudies (Opavsky et al. 2001).

The recent knowledge acquired on the mesenchy-mal stem cell biology, together with the need to dealwith well-characterized cell lines, suggested ourlaboratory to critically consider all the existent dataon SR-4987 and update them by new investigationsconcerning growth parameters as plating efficiency(PE) and population doubling time (PDT), its in vitroplasticity in comparison to other mesenchymal stemcells, and its drug sensitivity in comparison to tumorcell lines. As SR-4987 cells originated from mesen-chymal stem cells, we studied if the main differenti-ation multipotency described for MSCs (Dennis et al.1999) was maintained and found that under appropri-ate stimulation this mesenchymal cell line maintainsonly in part its multipotency (being able to differen-tiate toward chondro-osteogenic lineages but not intoadipogenic). A very interesting characteristic of SR-4987 cells is their ability to differentiate spontane-ously into osteoblast-like cells, and their resistance topaclitaxel (PTX) correlated to high P-glycoprotein(P-gp) expression.

Materials and methods

Drugs

Aspirin (ASA) was purchased from Cayman (Tallin,Estonia); indomethacin (IMC), 5-fluorouracil (5-FU),and doxorubicin hydrochloride (DOX) from EnzoLife Sciences (Lausen, Switzerland); valproic acid(VPA) and camptothecin (CPT) from Sigma-Aldrich

(St Louis, MO, USA); PTX from Serva (Mannheim,Germany); and verapamil (VP) from Abbott. All thestock solutions, with the exception of VPA and VP,were prepared in DMSO at a concentration of100 mg/ml (ASA), 400 mg/ml (IMC), 0.2 mg/ml (5-FU), 5 mg/ml (DOX and PTX), and 10 mg/ml (CPT);VPA stock was prepared in PBS (50 mg/ml). VP waspurchased as solution for i.v. injection (Isoptin) at aconcentration of 2.5 mg/ml. All the stock solutionswere stored at −20°C except the VP that was storedat +4°C. Working solutions were prepared freshaccording to the experimental design by serialdilutions in culture medium.

Cell lines

SR-4987 is a bone marrow mesenchymal cell lineestablished in our laboratory from a long-term bonemarrow cell culture of BDF/1 mice (Pessina et al.1992, 1997) and currently available by ATCC (CRL-2028). The cell line, cultured in RPMI 1640 mediumsupplemented with 5% fetal bovine serum and 2 mML-glutamine (EuroClone, UK), was weekly trypsi-nized by 0.05% trypsin/0.02% EDTA (EuroClone,UK) and splitted 1:5 to 1:10 into 25-cm2 flasks(Corning, USA). For our differentiation experiments,cells from passage 120 to 160 were used.

Human cell lines MOLT-4 (acute lymphoblasticleukemia) (Minowada et al. 1972) and HT-29 (colo-rectal adenocarcinomas) (Fogh and Trempe 1975)have been kindly supplied by Dr. Maura Ferrari(IZSLER, Brescia). Cells were maintained in IMDMsupplemented with FBS 10% (MOLT-4) or FBS 5%(HT-29) by weekly 1:3 passages.

Determination of PDT and PE

PDT has been evaluated as suggested by McAteer andDevis (1994). SR-4987 cells were cultured in 24multi-well plates at three different starting densities(1,000–5,000 and 10,000 cells/well), in quadruplicate.PDT value was calculated measuring the cell growthafter 72, 96, 120, and 144 h from the seeding, and thecells were counted by Trypan blue in a Burkerchamber.

Plating efficiency has been evaluated by culturingcells in 35-mm Petri dishes (Nunc, Germany) at threedifferent densities: 100–200 and 1,000 cells/dish (intriplicate). After 72 h at 37°C, 5% CO2, by an

170 Cell Biol Toxicol (2011) 27:169–180

inverted microscope (20–25× magnification) wescored the number of colonies adherent to the Petrisurface.

Mesenchymal stromal cells from human bone marrow

Mononuclear cells from human bone marrow (BM-MSC) were purchased frozen in liquid nitrogenfrom Lonza (Switzerland) and stored at −120°Cuntil use. Then the cells were thawed according toa procedure detailed for umbilical cord blood(Pessina et al. 2004) previously used. Briefly, cellswere quickly thawed at 37°C and then transferred toa conical tube by adding IMDM+10% FBS (Euro-Clone) and DNase I 10 U/ml (Serva, Germany).After centrifugation (15 min at 200×g), most of thesupernatant was removed and cell pellet resuspendedin 1–2 ml of the remaining medium. Fresh mediumwas added to the cell suspension and then centri-fuged as above. Cell pellet was resuspended in non-hematopoietic expansion medium (Miltenyi Biotec,Germany) and, after evaluation of cell number andviability, cells were plated in 25-cm2 flasks (Corn-ing) at 2×106 cells/ml. After 24 h at 37°C, 5% CO2,floating cells were discarded and fresh medium wasadded to adherent cells. Medium was changedweekly until cells reached 80% confluency. Then,cells were trypsinized and plated in differentiationmedia according to the conditions described for theirdifferentiation.

Osteo-, chondro-, and adipo-differentiation

SR-4987 cells were tested for their capacity todifferentiate into osteocytes, chondrocytes, and adi-pocytes according to the methods suggested byPittenger et al. (1999) with some modifications. Aspositive control, we used human bone marrowmesenchymal cells (hMSC) prepared as describedabove. Briefly, cells were plated into 35-mm Petridishes (Nunc) at 50 cells/cm2 density in 1 ml/Petri ofStemline Mesenchymal Stem Cell Expansion Medium(Sigma-Aldrich) supplemented with 3% FBS and4 mM L-glutamine; after 72 h of incubation at 37°C,5% CO2, culture medium was replaced with specificdifferentiation media described below.

According to the method suggested by Tropel et al.(2004), to induce osteogenic differentiation, SR-4987cells were cultured for 14 days in Stemline supple-

mented with 10 nM dexamethasone, 10 mM glycerol-2-phospate, and 300 nM L-ascorbic acid. Differenti-ation medium was replaced every 3–4 days. Culturemonolayers were then fixed for 5 min in cooledmethanol at −20°C and then processed for alkalinephosphatase staining, using SIGMA FAST BCIP/NTB (Sigma-Aldrich). The presence of more maturestage of osteoblasts was checked by alizarin stainingthat identifies calcium in cultured cells.

Spontaneous osteogenesis was evaluated by main-taining SR-4987 in growth medium without supple-ments. Both spontaneous and induced osteogenesiswere also evaluated in the presence of different drugs:two anti-inflammatory, ASA (100 and 300 μg/ml) andIMC (50 and 150 μg/ml), one anti-tumor, PTX (10,30, and 50 ng/ml), and an anti-convulsant and mood-stabilizing drug, VPA (50 and 150 μg/ml).

To induce chondrocytic differentiation, cells werecultured in hMSC chondrogenesis induction medium(Provitro, Germany) according to the micro-massmethods (Johnstone et al. 1998; Tallheden et al.2003) with some modifications. Cultures wererefeeded every 4 to 5 days, and after 21 days the cellmass was spread and smeared on slides and cellsfixed for 30 min at room temperature with formalinsolution, neutral buffered 10% (Sigma-Aldrich), andthen processed both by toluidine staining and byimmunocytochemistry technique for aggrecan. Brief-ly, for toluidine staining, fixed cells were stained bytoluidine blue 0.1% (dissolved in ethanol 70%) for3 min and then washed with distilled water. Forimmunocytochemistry, cells were permeabilized by0.3% Triton X-100 for 45 min, treated 10 min withH2O2, incubated overnight at 4°C with a mouse anti-human aggrecan monoclonal antibody (1:100; Milli-pore, USA) and then 1 h at room temperature withgoat anti-mouse HRP-conjugated antibody (1:50;Santa Cruz Biotechnology, CA, USA). Cells werefinally developed with 3,3′-diaminobenzidine solution(MP Biomedicals, Irvine, Canada), 0.5 mg/ml inbidistilled water, for 10 min.

To induce adipocytic differentiation, cells werecultured in Stemline supplemented with indomethacin200 μM (Alexis Biochemicals, USA), isobutylme-thylxanthine 0.5 mM (Applichem, Germany), dexa-methasone 1 μM, hydrocortisone 1 μM, and insulin10 μg/ml (all from Sigma-Aldrich). After 10 days ofculture, cells were fixed for 30 min at roomtemperature with formalin solution, neutral buffered

Cell Biol Toxicol (2011) 27:169–180 171

10% (Sigma-Aldrich), and then processed for Oil redO staining and Sudan Black (Sigma-Aldrich).

Effect of drugs on cell proliferation

The effect of the four anti-cancer drugs (PTX, DOX,CPT, and 5-FU), two anti-inflammatory drugs (ASA,IMC), and an anti-convulsant (VPA) on cell prolifer-ation of SR-4987 has been studied in 96 multi-wellplates (Sarstedt, Germany) in comparison to the effectexerted on other cell lines HT-29 and MOLT-4.Briefly, 1:2 serial dilutions of 2× concentrations ofthe drugs were prepared in 50 μl of culture medium/well according to different ranges as determined inpreliminary experiments not reported here. Accordingto the experimental design, to each well were added50 μl containing 0.5×103 (for SR-4987) or 103 cells(for HT-29 and MOLT-4). After 7 days of culture at37°C, 5% CO2, cell viability was evaluated by MTT(3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetra-zolium) assay as previously described (Mossman1983; Pessina et al. 1994). For each cell line anddrug, the inhibitory concentrations (IC) 50 and 90were determined according to the Reed and Muench(1938) formula.

Study of cell cycle and P-gp expression by FACSanalysis

DNA content for cell cycle phase detection wasperformed as follows: Cells at T0 and after 24 h oftreatment with 2 μg/ml PTX were trypsinized,washed, suspended in saline GM solution (glucose6 mM, NaCl 150 mM, KCl 5 mM, Na2HPO4 1 mM,KH2PO4 1 mM, EDTA 0.7 mM), and fixed with 96%(v/v) ethanol for 1 h at 4°C. After PBS wash, cellswere suspended in propidium iodide 50 μg/ml inPBS. Cells were incubated overnight at 4°C andanalyzed by flow cytometry (FacsVantageSE; Becton-Dickinson, USA). For evaluation of P-gp expression,cells were first incubated with a monoclonal antibodymouse anti-human P-gp (clone C1; Ylem, Italy) for30 min a 4°C, washed with PBS, and incubated with aFITC-conjugated goat anti-mouse immunoglobulin(Becton-Dickinson) for further 30 min at 4°C. Cellswere then washed once and analyzed by flowcytometry using a specific software (CellQuest Pro;Becton-Dickinson). Data reported refer to at leastthree independent experiments expressed as the ratiobetween intensity detected with the specific antibodyand the basal fluorescence with isotypic antibody.

Table 1 Main biological characteristics of mesenchymal stromal cells SR-4987

References

Tissue of origin Long-term bone marrow from a female BDF/1 mouse Pessina et al. 1992Morphologykaryotype

Fibroblast-like

Tetraploid

Clonogenicitytumorigenicity

Poor clonogenicity in agar (0.6%)

Induction of sarcoma in syngeneic mice

Markers Positivity for: CD5, CD44, CD45R (B220), (CD73), vimentin,GM3

Pessina et al. 1997

Negativity for: ras, fms, GM1a ganglioside

Negative for reverse transcriptase

Cytokinesproduction

Secreting M-CSF Pessina et al. 1992, 1997Expressing mRNA for bFGF, IL-7, GM-CSF, and SCF

Cytokinessensitivity

bFGF and aFGF dramatically enhance agar clonogenicity Pessina et al. 1995, 1998M-CSF, G-CSF, GM-CSF, IL-3, IL-7, TNFα, PDGF, and EGFdo not modulate clonogenicity

Cholera toxin produces a dramatic increase in intracellular cAMPwithout inhibiting cell growth

Applications Metabolic conversion of doxorubicin Gribaldo et al. 1999; Takagi et al. 1999,2005; Opavsky et al. 2001Bioassays of bFGF

Construction of three-dimensional hematopoietic culture system

Gene transfection studies


Results

Growth characters

The main characters of SR-4987 cells are summarizedin Table 1. This line expresses both features ofstromal cell and others that are more specific ofhemolymphopoietic cell line. SR-4987 cells arenegative for reverse transcriptase (RT) and for theexpression of p-21 ras, although they are tumorigenicin vivo (Pessina et al. 1997). The new investigationson the cell growth kinetics show that SR-4987 cellshave a mean PDT value of 24.5±5.4 h with aninoculum of 10,000 cells/cm2. As expected, the PDTvalue was found to be dependent on the cell density atthe starting seeding (number of cells/cm2) and increasessignificantly over the density of 10,000 cells/cm2.Plating efficiency, expressed as the percentage of cellsable to adhere on the substrate to give colonies at 72 h,was 2.87±1.19%.

Differentiation capacity

The reference differentiation pattern of BM-MSC isreported in Fig. 1, whereas Fig. 2 shows the differen-tiating capacity of SR-4987 cells. As human MSCsfrom bone marrow, SR-4987 cells also have thecapacity to differentiate into osteoblasts and chondro-cytes (Fig. 2b–e), but are unable to differentiate intomature adipocytes (Fig. 2h, i). Of interest is that SR-4987 cells are capable of differentiating spontaneouslyinto osteoblasts because in the absence of specificstimulation with osteogenic medium the cell monolay-er evidenced many focal plaques of spontaneousosteogenesis (one of it is documented in the Fig. 2f, g).

Drug sensitivity of SR-4987 proliferationto anti-cancer

The sensitivity of SR-4987 to four anti-cancer drugswas evaluated by MTT assay in comparison to two

ALP staining

Negative controlChondrogenesis

Blue Toluidine Aggrecan

Alizarin red

Osteogenesis

Oil Red O Sudan Black B

Adipogenesis

a cb

d e f g

Fig. 1 Plasticity of human bone marrow mesenchymal stem cells(hu-MSCs). a Negative control (hu-MSC in culture mediumwithout stimulation and without staining). b, c Toluidine stainingand immunocytochemistry for aggrecan to evaluate chondro-differentiation (micropellet slide spread of cells cultured 20 days

in chondro medium). d, e Alkaline phosphatase (ALP) andalizarin staining in methanol-fixed cells to evaluated osteo-differentiation (14-day cultures in osteo medium). f, g Oil red andSudan Black staining on formalin-fixed cells to evaluate adipo-differentiation (10-day culture in adipo medium)


cell lines of different histological origin: an acutelymphoblastic leukemia (MOLT-4) and a colorectaladenocarcinoma (HT-29). The drugs produce a dose-dependent inhibition of cell proliferation with signif-icant different slopes that indicate differences insensitivity. To 5-FU, the three cell lines show a veryclose dose-dependent sensitivity whereas to DOX andCPT some degree of resistance is observed for HT-29cells. SR-4987 cells are sensitive to the effect of bothDOX and CPT but express an important significantresistance to PTX (Fig. 3). The IC50 values comparedin Table 2 evidence the high resistance of SR-4987 toPTX that is quantitatively expressed (for each drug)by the resistance index (RI) calculated as the ratiobetween the IC50 values found for SR-4987 and IC50

determined for the other cell lines (Fig. 4). The RI ofSR-4987 for PTX calculated on MOLT-4 and HT-29were 39.6 and 68.2, respectively.

Sensitivity of SR-4987 cell to the anti-proliferativeeffect of anti-inflammatory drugs and HDACinhibitor

The study of the inhibition produced by ASA, IMC,and VPA on the three cell lines indicates a similarsensitivity to ASA but difference to both VPA andIMC (Fig. 5). These differences are better evidencedby considering the IC50 values reported in Table 3and the RI reported in the histogram (Fig. 6). Theresistance of SR-4987 cells to the inhibitory effect of

Negative control

Alizarin red

a b c

d e

h i

f g

Oil Red O Sudan Black B

Adipogenesis

Chondrogenesis

Blue Toluidine Aggrecan

ALP staining Alizarin red

Osteogenesis Spontaneous Osteogenesis

ALP staining

Fig. 2 Plasticity of SR-4987 cells. a Negative control (SR-4987 without stimulation and without staining); b, c toluidinestaining and immunocytochemistry for aggrecan to evaluatechondro-differentiation (micropellet slide spread of cells cul-tured 20 days in chondro medium); d, e alkaline phosphatase(ALP) and alizarin staining in methanol-fixed cells to evaluated

osteo-differentiation (14-day cultures in osteo medium); f, galkaline phosphatase (ALP) and alizarin staining in methanol-fixed cells to show spontaneous osteo-differentiation (14-daycultures in RPMI medium without differentiation factors); h, iOil red and Sudan Black staining on formalin-fixed cells toevaluate adipo-differentiation (10-day culture in adipo medium)


VPA on cell proliferation is about 5.4 times higherthan that found for MOLT-4, although it is similar tothat found for HT-29. In general, the cell lines testedare more sensitive to IMC than to ASA, whereas SR-4987 line is 2.87 more resistant to IMC than HT-29.

Cell cycle analysis and P-gp expression on SR-4987

The cell cycle analysis and P-gp expression on SR-4987 has been evaluated before and after 24 h oftreatment with PTX at 2,000 ng/ml (corresponding toabout 34-fold the IC50 value). Before treatment (T0),

the percentage of cells in G0/G1 phase was morethan 70%, and at the end of treatment (T24), thepercentage decreases to 45% with a significant blockof cell percentage (40%) into S phase. Only a littleincrease of cells entering G2/M phase is observed(Fig. 7a).

SR-4987 cells express a high level of P-gp (T0)that is dramatically down-modulated by 24 h oftreatment with PTX. By using 20 μM VP, which isknown to be a specific inhibitor of the P-gp pump(Parekh et al. 1997), a significant decrease (from 58.6to 8.3 ng/ml) of the IC50 value of PTX on SR-4987proliferation was observed (Fig. 8).

Effect of paclitaxel and anti-inflammatory drugson SR-4987 osteogenesis

The effect of PTX, ASA, IMC, and VPA was alsoevaluated on the induced osteogenesis of SR-4987 cells.The treatment with ASA and IMC produces a decreasein osteogenic differentiation (Fig. 9b–e) that is moresignificant for IMC. This is in agreement with thehigher sensitivity of SR-4987 to IMC anti-proliferation

5-Fluorouracil

0

20

40

60

80

100

0,001 0,01 0,1 1 10

µg/ml

% p

rolif

erat

ion

Doxorubicin

0

20

40

60

80

100

0,01 0,1 1 10 100 1000ng/ml

% p

rolif

erat

ion

Camptothecin

0

20

40

60

80

100

0,01 0,1 1 10 100

ng/ml%

pro

lifer

atio

n

Paclitaxel

020406080

100

0,1 1 10 100 1000

ng/ml

% p

rolif

erat

ion

a b

c d

Fig. 3 Sensitivity of SR-4987 (bold line) to anti-cancer drugs(5-fluorouracil, paclitaxel, doxorubicin, camptothecin) in com-parison to HT-29 and MOLT-4. Each point of the curve reportsthe mean value of three independent experiments in which thesensitivity was studied by evaluating the anti-proliferationeffect of the drugs in a MTT assay. The cell proliferation is

expressed as percentage referred to control cultures in solvent(that did not receive drug). Drug concentrations are expressedin micrograms per milliliter (5-FU) or nanograms per milliliter(PTX, DOX, CPT). Black diamond suit SR4987, black squareMOLT-4, black up-pointing triangle HT-29

Table 2 IC50 values (ng/ml) for 5-FU, PTX, DOX, and CPTdetermined in SR4987,MOLT-4, andHT-29 cell lines (mean±SD)

SR-4987 MOLT-4 HT-29

5-FU 151.0±31 190.0±38 134.0±29

PTX 58.6±6.9 1.48±0.7 0.86±0.3

DOX 1.94±0.7 1.34±0.5 16.4±4.0

CPT 1.43±0.5 3.65±1.0 10.0±2.9


effect in respect to that of ASA. Also, PTX inhibitsosteogenesis that is yet significant at 10 ng/ml (Fig. 9h,i, l). Only VPA seems to give a significant increase inSR-4987 osteogenesis (Fig. 7f, g).

Discussion

SR-4987 cell line is one of the first models ofmesenchymal stromal cell established in vitro (Pessinaet al. 1992) derived from a long-term mouse bonemarrow culture. The fibroblast-like morphology, thevimentin expression, the secretion of M-CSF togetherwith their dramatic sensitivity to the FGFs indicates thestromal nature of these cells (Table 1). Nevertheless,the analysis of CD membrane markers indicated amixed phenotype (CD5+, CD44+, CD45R+, sIG+,CD73+) that is also consistent with that of a B celllineage. This co-expression of stromal and B cellcharacteristics supports the hypothesis of the commonstromal–hematopoietic origin of this cell line. Themesenchymal origin of SR-4987 cells is confirmedby this study demonstrating their ability to differ-entiate into bone and cartilage as done by MSCexpanded from bone marrow. Surprisingly, thesecells differentiate spontaneously toward osteoblast-like cells and in the meantime have their stemnessreduced because they lost the capacity to matureinto adipocytes in appropriate differentiating cultureconditions (Fig. 1), as verified by using both OilRed S and Sudan Black staining. These results agree

5-FU 5-FU

PTX

PTX

DOX DOXCPT CPT0

10

20

30

40

50

60

70

80

Versus MOLT-4 Versus HT-29

Res

ista

nce

inde

x

Fig. 4 Resistance index of SR-4987 for 5-fluorouracil,paclitaxel, doxorubicin and camptothecin. Each column repre-sents the resistance index (RI) calculated for each drug as theratio between the mean value of IC50 determined in SR-4987cells and that found in MOLT-4 and HT-29 cells

Acetylsalicylic Acid

0

20

40

60

80

100

0,1 1 10 100 1000

µg/ml

% p

rolif

erat

ion

Indomethacin

0

20

40

60

80

100

0,1 1 10 100 1000

µg/ml

% p

rolif

erat

ion

Valproic Acid

0

20

40

60

80

100

1 10 100 1000µg/ml

% p

rolif

erat

ion

BA

C

Fig. 5 Sensitivity of SR-4987 (bold line) to acetylsalicylicacid, indomethacin, and valproic acid in comparison to HT-29and MOLT-4 cell lines. Each point of the curve reports themean value of three independent experiments in which the

sensitivity was studied by checking the anti-proliferation effectof the drugs in a MTT assay. Drug concentrations are expressedin micrograms per milliliter. Black diamond suit SR4987, blacksquare MOLT-4, black up-pointing triangle HT-29


with the recent observations of Røsland et al. (2009)suggesting that, after transformation, mesenchymalstem cells can reduce their potential to differentiate(e.g., into adipocytes and chondrocytes) maintainingthe ability to differentiate along the osteogeniclineage. Furthermore, our data agree with the reportssuggesting that SR-4987 cells are very sensitive tonative bovine bone morphogenetic proteins. For thisreason, some authors suggested to use SR-4987 inmicroassay for the measurement of osteoinductiveactivity (Ulmanen et al. 2002).

Although SR-4987 cells are negative for thereverse transcriptase and p-21-ras expression andshow poor clonogenicity in soft agar, these cells wereable to give sarcomas in syngeneic strain. Thischaracter of transformed cell line able to producetumors in vivo provides even a model to explore themechanisms to induce differentiation in a therapeuticcontext of experimental sarcomas. To investigate this

aspect, our study analyzed the in vitro drug sensitivityof SR-4987 in comparison to two cell lines ofdifferent histological origin (hematopoietic and epi-thelial) (see Fig. 3). SR-4987 cells are sensitive to 5-FU at the same levels of other tumor cells (with just alittle more sensitivity to DOX and CPT) but show ahigh resistance to PTX (RI=39.6 versus MOLT-4 andRI=68.2 versus HT-29). PTX exerts its action bybounding polymerized tubulin and promoting micro-tubules formation with the result to produce theirstabilization against disassembly and inhibiting cellmitosis (Schiff et al. 1979). By studying the PTXcytotoxicity on SR-4987 at 24 h, we were not able to

Table 3 IC50 values (ng/ml) for ASA, IMC, and VPAdetermined in SR4987,MOLT-4, andHT-29 cell lines (mean±SD)

SR-4987 MOLT-4 HT-29

ASA 541.4±112 335.9±91 694.9±130

IMC 93.5±25 75.9±27 32.5±11.2

VPA 410±101 75.5±29 519.6±107.0

ASA

ASA

IMC

IMC

VPA

VPA

0

1

2

3

4

5

6

Versus MOLT-4 Versus HT-29

Res

ista

nce

inde

x

Fig. 6 Resistance index of SR-4987 for acetylsalicylic acid,indomethacin, and valproic acid. Each column represents theresistance index (RI) calculated for each drug as the ratiobetween the mean value of IC50 determined in SR-4987 cellsand that found in MOLT-4 and HT-29 cells

G0/G1 S G2/M G0/G1 S G2/M0

10

20

30

40

50

60

70

80

T0 T24

T0 T24

Cel

l cyc

le p

hase

(%

)

1

1,2

1,4

1,6

1,8

Stai

ning

rat

io v

ersu

s is

otyp

e co

ntro

l

a

b

Fig. 7 Cell cycle analysis (a) and P-gp expression (b) in SR-4987 cells before and after PTX treatment. The histogram (a)reports the percentage of cells in each cycle phase evaluated byFACS before treatment (T0) and 24 h after treatment with2,000 ng/ml of PTX (T24). Histogram (b) shows the expressionof P-gp by SR-4987 cells at T0 and T24 evaluated by FACS andreported as the ratio between fluorescence intensity measuredon cells treated with specific labeled antibody and that of cellstreated with isotype control antibody


determine the IC90 values being higher than10,000 ng/ml (data not reported). However, our datashow that 24 h of treatment with 2,000 ng/ml inhibitscell division by blocking cells in S phase, but theirviability remains very high (87.9±9.0%). Our previ-ous studies demonstrated that SR-4987 cells are able

to bio-transform both anti-cancer drugs as DOX andpesticides by promoting a detoxification from thesemolecules. The capacity of SR-4987 to metabolizeDOX is higher than that of primary cultures ofhepatocytes and bone marrow (Gribaldo et al. 1999)and result in a non-toxic metabolite (Pessina et al.1999). Nevertheless, SR-4987 cells are not moreresistant to DOX than other cell lines, whereas theyare resistant to PTX (Fig. 3). It has been reported thatosteosarcomas can develop resistance to taxols thathas been explained as related to MDR/p-glycoproteinmechanism (Burns et al. 2001) and/or regulated bythe steroid-xenobiotic receptor (SXR) (Synold et al.2001). Even SR-4987 cells express high levels of P-gp that is down-modulated only by the treatment withhigh PTX concentrations (Fig. 6), whereas thetreatment with verapamil dramatically increases SR-4987 sensitivity to PTX. This is consistent with amechanism of resistance related to a P-gp/MDRsystem that gives SR-4987 cells the capacity todetoxify taxanes by an active transport of the drugout of the cells. Of course, we cannot exclude thatother mechanisms may be involved to accountresistance in these cells (as altered metabolism ofthe drug, decreased sensitivity to cell death-inducing

0

20

40

60

80

100

00,7

81,5

63,1

36,2

512

,5 25 50 100

200

paclitaxel (ng/ml)

SR

-498

7 vi

abili

ty (

%)

RI=39.64

RI=5.37

Fig. 8 Modulation of SR-4987 paclitaxel resistance byverapamil (VP). Each point of the curve reports the mean valueof three independent experiments in which the sensitivity wasstudied by checking the anti-proliferation effect of PTX in aMTT assay in the absence and in the presence of verapamil.Drug concentrations are expressed in micrograms per milliliter.RI resistance index expressed as the ratio between IC50 valueof SR-4987 and IC50 value of MOLT-4 for paclitaxel. Blackup-pointing triangle paclitaxel, black square paclitaxel+20 μMverapamil

e Indometacin

d

50 µg/ml 150 µg/ml

Paclitaxel

h

5 ng/ml

i

10 ng/ml

l

30 ng/ml

c Aspirin

b


a Control

f gValproic acid


Fig. 9 Effect of acetylsalicylic acid, indomethacin, valproicacid, and paclitaxel on the in vitro osteogenic differentiationof SR-4987 cells. Cells were cultured in osteogenic mediumas described in “Materials and methods”. Control cultures (a)osteomedium without drugs. The other cell culture received

osteogenic medium containing two different amounts of drugsas follows: ASA—100 μg/ml (b), 300 μg/ml (c); IMC—50 μg/ml (d), 150 μg/ml (e); VPA—50 μg/ml (f), 150 μg/ml(g); PTX—5 ng/ml (h), 10 ng/ml (i), 30 ng/ml (l)


stimuli, and alteration of microtubule due to beta-tubulin mutations) that will be important to check(Orr et al. 2003).

As SR-4987 is a mesenchymal stromal cell linethat is used for different in vitro biotechnologyapplications (Takagi et al. 1999; Opavsky et al.2001; Takagi 2005), we thought it is very importantto have information also on drug sensitivity of thesecells both to anti-inflammatory and HDAC inhibitors.In fact, in literature, exist heterogeneous dataconcerning mesenchymal stem cells sensitivity tothese molecules.

In vivo inhibition of cox-2 has been described todown-regulate osteoclasts and osteoblasts by favoringadipocyte formation (Kellinsalmi et al. 2007), andexperiments in mouse model (ovariectomy-inducedosteoporosis) showed that aspirin increases osteogen-esis of bone marrow MSC by aiming on telomeraseactivity and inhibiting osteoclasts (Yamaza et al.2008). In vitro experiments with ASA reportedinhibition of MSC proliferation by action on WNT/beta catenin signal pathway (Wang et al. 2006), andaccording to Chang et al. (2007) anti-inflammatorydrugs act on MSC by affecting cell cycle regulatorsbut not prostaglandin-related mechanisms. As ourdata demonstrate that SR-4987 cells are more sensi-tive to IMC than to ASA, it seems that in these cellsthe prostaglandin-related pathway is more sensitivethan that related to cyclo-oxygenase. Furthermore,SR-4987 differentiation into osteoblasts is increasedby sub-toxic ASA concentrations and a dramaticincrease of osteogenesis is induced by VPA. Thisseems to be in agreement with the observation of Choet al. (2005) that explained the osteogenic activity ofHDAC inhibitors as a consequence of the modifica-tion of the methylation status of DNA produced byVPA treatment in vitro.

Our study significantly improves the knowledgeon SR-4987 biology, opening interesting perspectivesfor further investigation on established mesenchymalcell lines. By taking into account the above data andthe previous data collected on this cell line, it appearsevident that SR-4987 provides a very interesting cellmodel useful for studying the plasticity of mesen-chymal stem cells addressed to the osteogenesis (bothin vitro and in vivo), for investigating metabolicfunctions of detoxification related with the role ofMSC, and also the mechanism related to taxolresistance.

References

Burns BS, Edin ML, Lester GE, Tuttle HG, Wall ME, WaniMC, et al. Selective drug resistant human osteosarcomacell lines. Clin Orthop Relat Res. 2001;383:259–67.

Chang JK, Li CJ, Wu SC, Yeh CH, Chen CH, Fu YC, et al.Effects of anti-inflammatory drugs on proliferation, cyto-toxicity and osteogenesis in bone marrow mesenchymalstem cells. Biochem Pharmacol. 2007;74:1371–82.

Cho HH, Park HT, Kim YJ, Bae YC, Suh KT, Jung JS.Induction of osteogenic differentiation of human mesen-chymal stem cells by histone deacetylase inhibitors. J CellBiochem. 2005;96:533–42.

Dennis JE, Merriam A, Awadallah A, Yoo JU, Johnstone B,Caplan AI. A quadripotential mesenchymal progenitor cellisolated from the marrow of an adult mouse. J Bone MinerRes. 1999;14:700–9.

Fogh J, Trempe G. New human tumor cell lines. In: Fogh J,editor. Human tumor cells in vitro. New York: Plenum;1975. p. 115–41.

Friedenstein AJ, Piatetzky-Shapiro II, Petrakova KV. Osteo-genesis in transplants of bone marrow cells. J EmbryolExp Morphol. 1966;16:381–90.

Gribaldo L, Catalani P, Erminio M. Metabolism of doxorubicinin long-term bone marrow cultures and SR-4987 stromalestablished cell line. Drug Metab Drug Interact.1999;15:279–91.

Haynesworth SE, Baber MA, Caplan AI. Cell surface antigenson human marrow-derived mesenchymal cells are detectedby monoclonal antibodies. Bone. 1992;13:69–80.

Johnstone B, Hering TM, Caplan AI, Goldberg VM, Yoo JU. Invitro chondrogenesis of bone marrow-derived mesenchy-mal progenitor cells. Exp Cell Res. 1998;238:265–72.

Kellinsalmi M, Parikka V, Risteli J, Hentunen T, Leskelä HV,Lehtonen S, et al. Inhibition of cyclooxygenase-2 down-regulates osteoclast and osteoblast differentiation andfavours adipocyte formation in vitro. Eur J Pharmacol.2007;572:102–10.

McAteer JA, Devis J. Basic cell culture technique and themaintenance of cell lines. In: Davis JM, editor. Basic cellculture. Oxford: Oxford University Press; 1994. p. 93–148.

Minowada J, Onuma T, Moore GE. Rosette-forming humanlymphoid cell lines. I. Establishment and evidence fororigin of thymus-derived lymphocytes. J Natl Cancer Inst.1972;49:891–5.

Mossman T. Rapid colorimetric assay for cellular growth andsurvival: application to proliferation and cytotoxicityassays. J Immunol Meth. 1983;65:55–63.

Opavsky R, Haviernik P, Jurkovicova D, Garin MT, CopelandNG, Gilbert DJ, et al. Molecular characterization of themouse Tem1/endosialin gene regulated by cell density invitro and expressed in normal tissues in vivo. J Biol Chem.2001;276:38795–807.

Orr GA, Verdier-Pinard P, McDaid H, Horwitz SB. Mecha-nisms of Taxol resistance related to microtubules. Onco-gene. 2003;22:7280–95.

Parekh H, Wiesen K, Simpkins H. Acquisition of taxolresistance via P-glycoprotein- and non-P-glycoprotein-mediated mechanisms in human ovarian carcinoma cells.Biochem Pharmacol. 1997;53:461–70.


Pessina A, Mineo E, Neri MG, Gribaldo L, Colombi R,Brambilla P, et al. Establishment and characterization ofa new murine cell line (SR-4987) derived from marrowstromal cells. Cytotechnology. 1992;8:93–102.

Pessina A, Gribaldo L, Mineo E, Neri MG. In vitro short-termand long-term cytotoxicity of fluoroquinolones on murinecell lines. Ind J Exp Biol. 1994;32:113–8.

Pessina A, Piccirillo M, Mineo E, Tassi E, Brambilla P, NeriMG. A simple quantitative bioassay for basic fibroblastgrowth factor using SR-4987 cells as target. Methods MolCell Biol. 1995;5:230–6.

Pessina A, Neri MG, Mineo E, Piccirillo M, Gribaldo L,Brambilla P, et al. Expression of B cell markers on SR-4987 cells derived from murine marrow stroma. ExpHematol. 1997;25:536–41.

Pessina A, Mineo E, Neri MG, Piccirillo M, Valore L, Giuliani A.SR-4987 and L1210 cell lines: two models in which choleratoxin susceptibility does not correlate with cAMP accumula-tion and ganglioside content. Cell Mol Biol. 1998;44:933–40.

Pessina A, PiccirilloM,Mineo E, Catalani P, Gribaldo L,MarafanteE, et al. Role of SR-4987 stromal cells in the modulation ofdoxorubicin toxicity to in vitro granulocyte-macrophageprogenitors (CFU-GM). Life Sci. 1999;655:513–23.

Pessina A, Eletti B, Croera C, Savalli N, Diodovich C, GribaldoL. Pancreas developing markers expressed on humanmononucleated umbilical cord blood cells. BiochemBiophys Res Commun. 2004;323:315–22.

Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R,Mosca JD, et al. Multilineage potential of adult humanmesenchymal stem cells. Science. 1999;284:143–7.

Reed LJ, Muench H. A simple method of estimating fiftypercent endpoints. Am J Epidemiol. 1938;27:493–7.

Røsland GV, Svendsen A, Torsvik A, Sobala E, McCormack E,Immervoll H, et al. Long-term cultures of bone marrow-

derived human mesenchymal stem cells frequently under-go spontaneous malignant transformation. Cancer Res.2009;69:5331–9.

Schiff PB, Fant J, Horwitz SB. Promotion of microtubuleassembly in vitro by taxol. Nature. 1979;277:665–7.

Synold TW, Dussault I, Forman BM. The orphan nuclearreceptor SXR coordinately regulates drug metabolism andefflux. Nat Med. 2001;7:584–90.

Takagi M. Cell processing engineering for ex-vivo expansionof hematopoietic cells. J Biosci Bioeng. 2005;99:189–96.

Takagi M, Sasaki T, Yoshida T. Spatial development of thecultivation of a bone marrow stromal cell line in porouscarriers. Cytotechnology. 1999;31:225–31.

Tallheden T, Dennis JE, Lennon DP, Sjögren-Jansson E, CaplanAI, Lindahl A. Phenotypic plasticity of human articularchondrocytes. J Bone Joint Surg Am. 2003;85:93–100.

Tropel P, Noe D, Platet N, Legrand P, Benabid AL, Berbera F.Isolation and characterisation of mesenchymal stem cellsfrom adult mouse bone marrow. Exp Cell Res.2004;295:395–406.

UlmanenM, Birr E, Hietala O, LindholmTS. Determination of thebiological activity of BMPs in cell cultures is unreliable. In:Lindholm TS, editor. Advances in skeletal reconstructionusing bone morphogenetic proteins. Singapore: WorldScientific; 2002. p. 53–61.

Wang Y, Chen X, Zhu W, Zhang H, Hu S, Cong X. Growthinhibition of mesenchymal stem cells by aspirin: involve-ment of the WNT/beta-catenin signal pathway. Clin ExpPharmacol Physiol. 2006;33:696–701.

Yamaza T, Miura Y, Bi Y, Liu Y, Akiyama K, Sonoyama W, etal. Pharmacologic stem cell based intervention as a newapproach to osteoporosis treatment in rodents. PLoS ONE.2008;3:e2615.


Date post:	04-Nov-2023
Category:	Documents
Upload:	unimi
View:	0 times
Download:	0 times

A mesenchymal stromal cell line resistant to paclitaxel that spontaneously differentiates into...

Documents