Z. Phys.Chem.220(2006)125-l3l / DOI 10.152412pch.2006.22l.1.125 München @ by OldenbourgWissenschaftsverlag,

The Carboxymethyl Dextran Shetl is an Important Modulator of Magnetic Nanoparticle Uptake in Human Cells By M.Schwalbe''*, and J. H. Clementr

N. Buske', M. Vetterleinr, K. Höffken', K. Pachmann',

' Departmentof Internal Medlcihe II, Friedlich Schiller University Jena, Erlanger Allee l0l, D-07V40Jena,Germany 2 MagneticFluids,KöpenickerLandstlaße203, D- 12437Berlin, Gelmany (ReceivedAugirst 31, 2005; acceptedin revisedfolm October l2,2OO5)

Magnetic Nanoparticles / Carboxymethyl Dextran / Tumor Cells / Leukocytes In oncologyand hematologythe separationof tumor cells tlorn healthycells in peripheral blood is a vital problern. We could show previously,that enrichmentof tumor cells from peripheralblood is possibleby using magneticnanoparticleswith a carboxymethyl dextran(CMD) shell. Long-term storageof CMD nanopalticleseliminatedthe differential labeling of tumor cells and leukocyteswhich might be due to an alteration of the carboxymethyl dextran shell. Incubation of stored CMD nanopalticleswith freshly pleparedcarboxyrnethyldextranrestoredthe differentiallabeling. In contrast,enzymatic degradationof the carboxymethyldextran shell with dextranaseabolishedthe cell+ype specificlabeling.Thus, an intact carboxyrnethyldextlan shell is crucial tbr the cell+ype specificinteractionof the CMD nanoparticlesand living cells.

l.Introduction Cancer is one of the most severe diseasesof mankind. A serious aspect of cancer is its ability to spread and form metastases.Once the primary tumor is established,tumor cells may dissociate fror-nthe tumor and disseminate to other parts of the body via the circulation. Multiple researchefforts are focused to the treatment of the primary tumor or the rnetastasesusing magnetic nanoparticle technologies (e.9. hyperthemia or drug targeting). We focus on the tumor cells circulating in the peripheral blood. The frequency of cirrculatingtumor cells among nonnal blood cells is assumed to *

Corresnondins

attthor F-rnail'

mrntr cchurrlhpt'ömo't,rni-i--4

.t-

126

M. Schwalbeeral.

be in the range of 10*3-10-7. In oncology and hematology the detection and quantification of these disseminated tumor cells become an important diagnostic tool in order to monitor the primary tumor and to guide therapeutic decisions.Circulating tumor cells are routinely assayedby using highly sensitive rnethods,e.g. polymerasechain reaction, immunohistochemistry,laser scanningcytometry [1-3]. An efficient, inexpensiveand universalmethod for the enrichment of circulating tumor cells from the peripheral blood of patients is therefore highly recommended. A successful approach to separate cells from sLrspensionsis to use magnetic nanoparticles covered with various coatings. The most popular are dextran and its derivatives [4]. It has been shown from several groups,.that carboxymethyl dextran (CMD) coated magnetic nanoparticlescan.inieract with living cells [5-7] and that the interaction is cell-type specific [8]. In order to understand the differential interaction of tumor cells and leükocyteswith CMD coated magnetic nanopiu'ticles, we analyzed the role of the carboxymethyl dextran shell in more detail.

2. Materials and methods 2.1 Magnetic nanoparticles The nanoparticles were produced by one of us (N.B.) and consisted of magnetite/maghemitecore. The TEM-size diama superpara-(felro)-magnetic eter of the core varied between 3 and 15 nm (average diameter 5 nm). The nanoparticleswere initially coated with carboxymethyl dextran (MW 1500020000) to yield a hydrodynamic diameter of the nanoparticle clusters of 200-300 nm. The saturation magnetizationranged from 4.5 to 6.2 mT. The Fe (II) content was in the range of 5-10 mg/ml and the Fe (III) content from 45-55 mglml. 2.2 Repetitive incubation of magnetic nanoparticles with carboxymethyl dextran The re-incubationof the magnetic nanoparticleswith CMD was performed as follows: CMD (MW 15000-20000; Sigma-Aldrich, Deisenhofen,Germany) was dissolved in water and added to the magnetic nanoparticle solution. After irlcubationfor 2 h at 31 "C the magneticnanoparticleswere collectedby magnetic force and washed 2 times with tap water. Finally, the nanoparticles were tleated with ultrasonicto minimize aggregateformation. 2.3 Dextranase treatment of CMD nanoparticles 25 pl of the nanoparticle solution were treated overnight with 5 U dextranase (1.6-o-D-glucan 6-glucanohydrolase,Sigma-Aldrich, Deisenhofen,Germany) at room temnerafllre.

Carboxymethyl Dextran CoatedMagnetic Nanoparticles

2.4 Cell line and peripheral

127

blood leukocytes

The breast cancer cell line MCF-7 was obtained from American type culture collection (ATCC, Rockville, USA). The cell line was cultivated under standard conditions with DMEM with l07o fetal calf serum in humidified air. For incubation experiments the adherent MCF-7 cells were detached with Trypsin/EDTA. Peripheral blood leukocytes wele prepared by erythrocyte lysis (Qiagen, Hilden, Germany) from whole blood samples of healthy volunteers. The leukocyte pellet was washed with elythrocyte lysis buffer twice and then resuspendedin PE buffer (phosphate-buff'ersaline (PBS), 2 mmol EDTA).

2.5 Incubation

of cells with magnetic nanoparticles

MCF-7 cells (l x 106 cells per 500 pl) or peripheral blood leukocytes (2.5 x 106 cells per 5p0pl) were inoculated in short term incubation (0 to 20min) with 2.5 pr,lCMD coatedmagneticnanoparticlesin PE at 37 "C. After treatment magnetically labeled cells were separated using a SuperMACS device and MS columns (Miltenyi-Biotech, Bergisch-Gladbach,Gernrany).The separated cells were designed as positive fraction (retained in column) and the efflux as negative fraction. Cells from both fractions were quantified automatically (ParticleCount & Size Analyser 22,Beckman-Coulter, Krefeld, Germany).

3. Resultsand discussion Carboxymethyl dextran coated magnetic nanoparticles are widely used for labeling and detecting cells. During application the narroparticlesattaclr to the cell surface and are incorporatedinto the cell. lntracellular particles are surrounded by membranousstructurcs,which points to endocytotic mechanisms ofuptake t9-131. The attachmentand subsequentincorporationofthe nanoparticles allows the separationof the cells in a magnetic field. The interaction of the CMD nanopalticleswith various cell line cells and primar-ycells showed, that it is cell-type specific.Most of the tumor cells fiom variousorigins showed a more pronouncedinteraction over an incubation time up to 20 min than primary cells like leukocytes from peripheral blood [8]. During our expelimentswe observedthat the CMD magneticnanoparticles lost their ability to differentiatebetween tumor cells and leukocytes(Fig. la). After a storage period of 3 to 5 months the kinetic behavior of the stored CMD nanopalticleswas similar to magnetic nanoparticleswithout any shell. These nanoparticlesinteract within a few minutes with nearly all turnor cells. The same holds tme for the leukocytes prepared frorn peripheral blood showing that the cell-type specificity was lost, too (Fig. I b). We designatedthis phenomenon "aging" as and supposethat it is causedby the disruption of the carboxymethyl dextran shell.

t28

M. Schwalbeet al.

,F

I

ti O

o

incubation time [min] Fig. 1. CMD coatedmagneticnanoparticlesinteractdifferentially with MCF-7 cells and leukocytesin a tin-re-depenclant manner.a MCF-7 cells show an immediate interaction with CMD nzrnopat'ticles with a further increaseof cells in the positive fraction up to netly 90Voafter 20 min. Leukocytes from peripheral blood exhibit a delayed interaction with a maxinturnof cells in the positivefractionof 65Voafter20 min. b After 3 to 5 month of storageCMD nanoparticlesloose their ability to differentially iabel MCF-7 cells and leukocytes.Bars represent4 experiments+ SD. MCF-7 l; leukocytes!.

Dextran is a natural product. Because of the size of the molecule and the free hydroxyl groups complex high-molecular and three-dimensional structures are formed. Thus, it is highly improbable, that each single nanoparticle is surrounded by its own carboxymethyl dextran shell. It is more likely, that single nanoparticles or nanoparticle aggregates are embedded in the complex carboxynrethyl clextran network. During storage, the network might be degraded over lime and more and more nanoparticles are liberated. In order to proof the

129

CarboxymethylDextranCoatedMagneticNanoparticles 100

;

l

80

100

I

I

90

90

-

I I

70

70 I

DU

60 H

50

o

40 30 20 10

50 40

I

30

l-

20

I

d

10 0

4

8

1

2

incubation time [min]

4

8

1

2

incubation time Imin]

Fig.2. The differpntialinteraction of CMD r.nagnetic nanoparticles with tuurorcellsa1d leukocytes can bb restored. a 'Aged" nanoparticles were reincubated with fr-eshly preparedcMD. Theyshowthesametime-course of sepzrrable cellsas newlyobtainedcMD (Fig.1a).b Aftel long-telmstorageLlntreated nanopafiicles CMD nangparticles showed a similarmaximltmlabelingof lelrkocytes andMCF-7cellswithina sholttinte.Barsrepr€sent4 experiments + SD.MCF-7I; leukocytes tr

hypothesis,that the carboxymethyldextranshell is responsiblelbr the cell-type and time-dependantinteraction of the nanoparticlesand the cells we incubated the "aged" nanoparticleswith a freshly preparedcarboxymethyl dextran solution and used these nanoparticlesfor labelling experiments.The reincubated nanoparticlesled to a similar labelling kinetic as newly obtained CMD nanoparticles.The tumor cells showed a more rapid uptake of the nanoparticles than the leukocytes with the rnost pronounced dit'l'erenceafter 4min of incubation (Fig. 2a).In contrast,the presenceof nanoparticleswithout a cMD shell causeda rapid paralleled labelling of turnor cells zrswell as leukocytes (Fig. 2b). If the carboxymethyl dextran is responsiblelbr the crifferencesbetween tumor cells and leukocytes,a degradationof the cMD shell of newly obtained nanoparticlesshould cause the same results. Ther-efore,the cMD coated magnetic nanoparticleswere treated with dextranasein order to enzymatically degrade the carboxyrnethyl dextran. After the digest the treated nanoparticlesshowed the same behaviour as the "aged" nanoparticlesand the nanoparticleswithout CMD shell. The tumor cells and the leukocytes were labelled rapidly and to a high extent (Fig. 3). In conclusion, we could show that the carboxymethyl clextranshell plays an important role for the interaction of rnagnetic nanoparticleswith cells. The cMD shell is sufficient to enrich tumor cells frorn leukocytes without the use of antibodies coupled to the nanoparticle shell and destruction of

130

M. Schwalbeetal.

b

100 90

100 90

e- 80

e- 80

o l u

6 7 0

H o o

Hoo

e 5 0 E ^ n

P 5 0 -^ 40

q 3 0

a 3 0

- 2 O

o 2 0

3 r o

E r o

.L

4

8

1

2

incubation time [min]

'leukocytes

4

8

1

2

incutration time [min]

MCF.7

Fig.3. Breakdown of the carboxymethyl shell abolishesthe differential interaction kinetics of rnagneticnanopafticleswith cells. CMD nanoparticleswere treated with dextranase overnight and used for incubation experiments. After that treatment leukocytes do not show the delayed interaction kinetic as with untreated CMD nanoparticles.MCF-7 cells were mor€ rapidly labeled with dextranase-treatednanoparticles.Bars represent3 experimenß a SD. a. Leukocytes:Q dextranasetreatment;E control; b. MCF-7: N dextranase treatment;I control.

type specific labelling could be overcome by reincubation with carboxymethyl dextran. Long-term studies are necessary to evaluate the effect of repeated incubations of CMD nanoparticle charges with frestlly prepared carboxymethyl dextran.

Acknowledgement (DFG) PriThis work wassupportedby the DeutscheForschungsgemeinschaft ority programmI | 04, grantCl 2l2ll-2.

References l. K. Pantel,J. R. Izbicki, M. Angstwurm, S. Braun, B. Passlick,O. Karg, O. Thetter, and G. Riethmüller,CancerRes. 53 (1993) 1027. 2. G.G.Wulf, B. Jürgens, T. Liersch, W. Gatzemeier, H. Rauschecker, C.Buske, M. Hufner, W. Hiddemann, and B. Wörmann, J. Cancer Res. Clin. Oncol. 123

Calboxyrnethyl Dextr.anCoatedMagnetic Nanoparticles

131

3' K. Pachmann' J. H. clernent,c.-p.schneider, B. wiilen, o. c.rnara,u. pachmann, andK. Höflken,Clin. Chem.Lab.Med.43 (2005)617. 4. L.M. Lacava, Z.G.M. R. B. Azevedo, !9aya, M. F.daSilva,O. Silva,S.B. Chaves, F. Pelegrini,C. Gansau, N. Buske,D Sabolovic,and p. C. Vforrl*,'Äiopirys.J. g0 (2001')2483. p. 5. A. E. Trickett,O I and M. R. Vowels,Bone Manow f91A, R. Lam_po_Tang, Transplant. 7 (1991)t99. 6. J. P.HancockandJ. T. Kemshead, J. Immunol.Methods164(1993)51. 7. B. I. Haukanes andC. Kvarn,Biotechnology ll (t993) 60. 8 {, H. Clement,M. Schwalbe, gr.rk"t K: Wogn"" M. Schnabelrauch, p. Görnerr, -}l K. o. Kliche, K. pa'chrnann, w. weitschies,utr?r. Höffken,r. cancei Res.clin. Oncol.,in -press. 9. A.Jordan, P.Wur.st, S_.!9lrt B. Tesche,H. Fahlirg, T. Mitrovics, T. Vogl, l{ and.R.-Felix, Int. J. Hyper:therrnia _^ J. Cervos-NaVarro, 12 ltV}ey rc5. 10' A. M' Koch'.F. Reynolds,4vl. F. Kircher,ri. p. tvterue,n w"i.ri",r"i andL. Josephson,Bioconjugate Chern.14 (2003)I | 15. 11' c' wilhelm, c. Billotey,J. Roger,J. N. pons,J.-c. Bacri, andF.Gazeau,Biomaterials24 (2003),1001. 12. Cl'ilhehn, F. Gazeau,J. Roger,J. N. pons,anclJ._C.Bacri, Langm'ir Ig (2002) 13. K. Wagner,A. Kautz,M.Röde., M. Schwalbe,K. pachmann, J. H. Clernent,and M. Schnabelrauch, Appl. Organorner. Chern.l{i (2004\ 514.