Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1268
Tumor Cell Targeting ofStabilized Liposome Conjugates
Experimental studies using boronated DNA-binding agents
BY
ERIKA BOHL KULLBERG
ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2003
“You've got to work hard, you've got to work hard.If you want anything at all”Depeche mode
This thesis is based on the following papers, which will be referred to in thetext by their roman numerals I-VI
I E. Bohl Kullberg, N. Bergstrand, J. Carlsson, K. Edwards, M.Johnsson, S. Sjöberg and L. Gedda. Development of EGF-conjugated liposomes for targeted delivery of boronated DNA-binding agents. Bioconjugate Chemistry 13:737-43. (2002)
II E. Bohl Kullberg, J. Carlsson, K. Edwards, J.Capala, S. Sjöbergand L. Gedda. Introductory Experiments on Ligand Liposomes asDelivery Agents for Boron Neutron Capture Therapy.International Journal of Oncology, In Press, (2003)
III E. Bohl Kullberg, M. Nestor and L. Gedda. Tumor-cell targetedEGF-liposomes loaded with boronated acridine: uptake andprocessing. Pharmaceutical Research. 20:229-236. (2003)
IV Q.Wei, E. Bohl Kullberg and Lars Gedda. Trastuzumab-conjugated boron containing liposomes for tumor-cell targeting;Development and cellular studies. Submitted. (2003)
V E. Bohl Kullberg, Q. Wei, J. Capala, V. Giusti and L. Gedda.BNCT of cultured glioma cells using EGF-receptor targetedliposomes. Manuscript (2003)
VI E. Bohl Kullberg, Q. Wei and L. Gedda. Altered EGFbiodistribution in mice after liposome conjugation. Manuscript(2003)
Reprints were made with kind permission from American Chemical Society(I), International Journal of Oncology (II) and Kluwer Academic/ PleunumPublishers (III)
Contents
1. Introduction and background.......................................................................... 11.1 Tumor targeting........................................................................................ 21.2 Liposomes ................................................................................................ 4
1.2.1 Commercially available liposome formulations ............................. 41.2.2 Tumor targeting liposomes .............................................................. 7
1.3 The EGFR-family and its ligands............................................................ 81.3.1 Receptors .......................................................................................... 81.3.2 Ligands ............................................................................................. 91.3.3 Signaling......................................................................................... 101.3.4 In cancer development ................................................................... 101.3.5 For cancer therapy.......................................................................... 11
1.4 BNCT ..................................................................................................... 131.4.1 History ............................................................................................ 131.4.2 Compounds for BNCT................................................................... 151.4.3 BNCT and liposomes..................................................................... 16
2. Aims .............................................................................................................. 19
3. Materials, methods and techniques .............................................................. 203.1 Liposomes .............................................................................................. 203.2 Boronated DNA-binding compounds.................................................... 213.3 Cellular models ...................................................................................... 233.4 Boron determinations............................................................................. 243.5 Neutron irradiation at the BNCT facility in Studsvik........................... 243.6 Animals .................................................................................................. 24
4. Preparation of liposome conjugates ............................................................. 254.1 Micelle-transfer procedure..................................................................... 254.2 Radiolabeling of the targeting agent ..................................................... 264.3 Optimization of micelle-transfer conditions for EGF-liposomes usingmaleimide-PEG-DSPE (paper I) ................................................................. 274.4 Optimization of conditions for trastuzumab-liposomes using NHS-PEG-DSPE (paper IV) ................................................................................. 284.5 Preparation of EGF-liposomes using NHS-PEG-DSPE (paper V)...... 294.6 3H-labeled liposome conjugates (paper III and IV).............................. 294.7 Comments .............................................................................................. 29
5. Cell experiments ........................................................................................... 315.1 Test of receptor specificity .................................................................... 315.2 Time-dependent uptake ......................................................................... 325.3 Retention ................................................................................................ 345.4 Membrane-bound and internalized conjugate....................................... 365.5 Intracellular localization using fluorescence......................................... 365.5 Optimization of boron uptake................................................................ 395.6 Comments .............................................................................................. 40
6. BNCT experiments on cultured glioma cells............................................... 426.1 Experimental details .............................................................................. 426.2 Dose calculations ................................................................................... 436.3 Cell survival after neutron irradiation ................................................... 436.4 Comments .............................................................................................. 45
7. Biodistribution of EGF and EGF-liposomes ............................................... 467.1 Experimental procedures ....................................................................... 467.2 Circulation time...................................................................................... 477.3 Uptake in liver and kidneys ................................................................... 477.4 Uptake in other organs........................................................................... 497.5 Comments .............................................................................................. 50
8. Summary and future work ............................................................................ 518.1 Summary ................................................................................................ 518.2 Future work ............................................................................................ 52
Acknowledgements........................................................................................... 53
References ......................................................................................................... 55
Abbreviations
AR AmphiregulinBMRR Brookhaven medical research reactorBNCT Boron neutron capture therapyBNL Brookhaven national laboratoryBOPP Boronated protoporphyrineBPA borophenylalanineBSH Sulfhydryl boraneCAT Chloramine-TCBA carboranylalanineCEA Carcinoembryonic antigenCPA carboranylpropylamineDOX DoxorubicinDSPC Disteaoryl phosphatidylcholineDSPE Disteaoryl phosphatidylethanolamineEGF Epidermal growth factorEGFR Epidermal growth factor receptorETA Pseudomonas exotoxin AFab Monovalent antibody fragmentGBM Glioblastoma multiformeGM1 monosialogangliosideHB-EGF Heparin binding EGFHER-2-4 Human EGFR 2-4HRG HeregulinHTR Hitachi training reactorICP-AES Inductively coupled plasma-atomic
emission spectrometryICP-MS Inductively coupled plasma-mass
spectrometryID Injected doseLDL Low density lipoproteinsmAb Monoclonal antibodyMPS Mononuclear phagocyte systemNRG NeuregulinPDT Photodynamic therapyPEG Polyethylene glycol
PPM Part per million (µg/g)RID RadioimmunodiagnosticsRIT RadioimmunotherapyScFv Single-chain antibody fragmentTGF-α Transforming growth factor alphaWSA Water soluble acridineWSP Water soluble phenantridine
1
1. Introduction and background
Cancer therapy today is efficient in treating solid tumors in placesreachable for surgery or radiotherapy, the major treatment modalities todayfor primary tumors and large metastases. Chemotherapy is also effective intreatment of residual and spread disease in some tumor types, for examplelymphomas. However, these treatment modalities cannot cure a largenumber of patients due to location of the tumor, the presence ofdisseminated cells or recurrence of a drug resistant disease. Targetedtherapy might be of help when other curative treatments fail. Tumortherapy seeking out the disseminated cells in the bloodstream andlymphatic vessels and finding the residual cells after surgery is anappealing approach gaining interest.
Figure 1. The principle of tumor targeting with different carrier molecules. Thetargeting agent binds to targets on the cell surface and the toxic substances canexecute their actions. A) An antibody with a toxic agent. B) A polymer chainconjugated to a targeting agent and loaded with toxic agents. C) A targetedliposome loaded with toxic agents.
tumor cell
nucleus
targets
A
B
C
2
1.1 Tumor targetingCells have on their surface specific molecules designed to regulate severalprocesses such as differentiation and growth. These surface molecules canbe overexpressed on tumor cells and are therefore referred to as tumorassociated antigens. Tumors are also known to overexpress receptors for,for example, growth hormones, vitamins and lipids. These overexpressedstructures can be targeted and used for therapy by an antibody or a receptorligand to which a toxic substance of a radionuclide has been coupled.
Figure 2. A tumor cell with some of the more frequently targeted structures andexamples of their targeting agents.
Antibodies, most frequently monoclonal antibodies, mAbs, are used totarget tumor specific structures in several ways. In radioimmunotherapy,RIT, the radionuclide attached to the antibody is chosen to deliver localradiation energy in order to kill the targeted cells efficiently. The mostfrequently used radionuclides for therapy are β-emitters like 131I and 90Y. Inradioimmunodiagnostics, RID the same targeting principle is applied butthe radionuclides are chosen to emit X-ray and gamma suitable for externaldetection, for example 111In and 99mTc. So far, successful therapy has beenaccomplished with haematopoetic tumors, such as non-Hodgkin'slymphoma. Antibodies targeting the tumor antigens CD-19, CD-20, CD-22or CD-37 have been used with 131I or 90Y and have shown good specificityand therapeutic results (1, 2). A humanized antibody, rituximab, directed
Tumor associated antigens CD-19 CD-20, CD-22,CEA
Ligand: no natural, specific antibodies
nucleus
Growth factorreceptorLigand: growth factors
Vitamine receptorLigand: folate
Metal-receptorLigand: transferrin
HER-2 Ligand: no natural,specific antibodies
tumor cell
LDL-receptorLigand: lipoproteins
3
towards CD-20 have shown good treatment effects (60% response) not justas a radiolabeled antibody, but also in itself. A non-humanized version ofthis antibody, ibritumomab, has shown response rates of 80 % whenlabeled with 90Y (1, 2).
Tumor cells can also be eradicated using antibodies conjugated to toxicsubstances, such as ricin, genistein and pseudomonas exotoxin A (ETA).Antibodies can be used for immunotherapy by themselves or conjugated toa superantigen to evoke a more powerful immunoreaction towards thetargeted tumor structure.
The use of antibodies has been hampered by the fact that most mAbs arederived from mouse and can therefore evoke an immune response towardsthe injected murine antibody thereby disabling further injections. Bychanging the non-binding parts of an antibody to human parts, a humanizedchimeric antibody is created, being much more tolerated for repeateddosing. In some cases when a smaller targeting agent is needed, only thebinding part of the antibody, the Fab fragment can be used. The smallestparts of the antibody, the variable regions, so called single-chainfragments, ScFv, can also be used for targeting.
If the targeted structure in question has a natural ligand, then this ligand,or a derivative of it, can be used for targeting. For the overexpressedepidermal growth factor receptor, EGFR, the ligand EGF can be used fortargeting. More about this receptor and its ligand can be read below. Thevitamin folate receptor is often overexpressed on various types of tumors,such as ovarian, colorectal and endometrial carcinomas (3), and the folicacid or folate has been used for targeted delivery of both radionuclides anddrug carriers (3). Neuroendocrine tumors often overexpress thesomatostatin receptor, and a somatostatin analogue, octreotide, has beenused for both imaging and therapy of this kind of tumors (4). Many typesof tumors also overexpress receptors for low density lipoproteins, LDLs.This has awakened the interest for use of LDLs as delivery vehicles forchemotherapeutics. Experiments have been performed to loadanthracyclins into LDL particles with promising results regarding stabilityand toxicity (5, 6).
There are several ways to deliver the toxic agents with targeted therapy.The simplest are radiolabeled antibodies and ligands. To increase theamount of radionuclides or toxic agents, carriers can be attached to thetargeting agent. Possible carriers are polymers such as dextranes,liposomes, and chelates. There are advantages and disadvantages for theuse of large carrier molecules, the main advantage being the fact that moretoxic agents can be loaded. Compared to a small peptide or ligandmolecule, the larger constructs have a completely different circulationpattern and usually longer circulation times. This can be beneficial, giving
4
the construct more time to find disseminated tumor cells, but a large sizemight also give limited passage through capillary walls and thereforehamper the possibilities to target the tumor cells in normally vascularizedtissue. Smaller molecules, on the other hand, can have a too quick passagethrough the body and be excreted before finding the tumor cells to exerttheir toxicity (4).
1.2 LiposomesLiposomes are phospohlipid bilayer spheres composed of lipophilic doublemembranes with an aqueous core. Liposomes have been proposed as drug-delivery vehicles since the mid 1970´s (7). Hydrophilic drugs can beloaded in the aqueous core and lipophilic drugs in the double membrane.The early liposome in vivo experiments, using large (>200 nm), oftenmultilamellar liposomes (8), had problems with rapid removal from theblood stream by cells of the mononuclear phagocyte system MPS (9). Tocircumvent this problem, sterically stabilized liposomes were constructedand examined during the late 1980´s, where a polymeric coat was used toshield the liposomes from opsonization and recognition by the cells of theMPS (9, 10). Two main formulations of stabilized liposomes wereproposed: liposomes with monosialoganglioside, GM1 (9) and liposomeswith polyethylene glycol, PEG (10, 11).
Liposomes have been shown to gather in sites with increased capillaryblood-flow and leaky vasculature, such as inflammations (12, 13) andtumors (14-16). This tumor-homing effect is used for all commerciallyavailable liposome formulations today.
1.2.1 Commercially available liposome formulationsSome of the most potent drugs for cancer therapy are the anthracyclins:doxorubicin and daunorubicin. Unfortunately their use is constrained byhighly problematic systemic toxicities. For this reason the most studieddrug delivery systems are designed to enhance or preserve the toxicity ofanthracyclins against tumor cells but reduce the side effects for normaltissues, such as cardiotoxicity and bone marrow damage. Current activeloading methods make it possible to load 104 anthracyclin molecules intothe aqueous core of each liposome.
5
Figure 3. The structure of a stabilized liposome. A) The polar headgroup of thephospholipid. B) The lipophilic tails of the phospholipid. C) The polymer brush(PEG) stabilizing the liposome and making it less prone to uptake by the immunesystem. D) Lipophilic drugs that can be loaded in the lipid bilayer. E) Hydrophilicdrugs that can be loaded in the aqueous core.
The drug Myocet™ (Elan Pharmaceuticals) consists of doxorubicinenclosed in moderately sized liposomes (190 nm). Myocet™ gives limitedprolonged circulation compared to free drug but reduces the toxicity due toaltered biodistribution of doxorubicin. In trials where Myocet™ has beentested against free doxorubicin in metastatic breast cancer, the liposomaldrug exhibited less cardiotoxicity and neutropenia (17). However, itremains a controversy whether Myocet™ is equally effective as the freedrug (18).
For the drug DaunoXome® (Gilead Sciences) small liposomes (45 nm)loaded with daunorubicin are used. These liposomes have proven to extendcirculation times due to their small size and rigid bilayer. The drug hasbeen shown to be active against Kaposis sarcoma (19).
To increase the stability and circulation time the liposomes can be, asdescribed above, coated with a layer of polyethylene glycol. This is thecase for the doxorubicin loaded liposome formulation known asDoxil®/Caelyx® (Alza corporation). The liposomes are small (100 nm),rigid and coated with approximately 5 mol% PEG. They have been shown
A
B
C
D
E
6
to have a long circulation time, which increases accumulation in tumortissue. Doxil®/Caelyx® has proved to decrease side effects of doxorubicin,such as nausea and hair loss (alopecia) significantly, but it also inducessome new toxicities, the most noticeable being the palmar-plantareythrodysesthesia known as hand-foot syndrome. This syndrome is due tothe fact that the liposomes get stuck in the small capillaries of the palmsand soles, giving rise to high local doxorubicin concentrations. Thissyndrome is the usual dose limiting toxicity of Doxil®/Caelyx®. The drughas been shown to be effective against a number of solid tumor types, suchas Kaposis sarcoma (20, 21) and ovarian cancer (22, 23) and has also beentested for metastatic breast cancer with promising results (24, 25).
Table 1. Commercially available liposome formulations
Name Company Structure Drug IndicationMyocet™ Elan liposome Doxorubicin Metastatic
breast cancerDaunoXome® Gilead liposome Daunorubicin Kaposi´s
sarcoma (KS)Doxil®/Caelyx®
Alza PEG-liposome
Doxorubicin KS, ovariancarcinoma
AmBisome Gilead liposome Amphotericin B Anti fungal
Not only drugs for tumor therapy have been developed. AmBisome(Gilead Sciences) is a liposomal formulation of amphotericin B provenvery effective against fungal infections. Amphotericin B forms an ioniccomplex with the phospholipids in the bilayer and is not, compared to theformulations with anthracyclins described above, loaded in the aqueouscore. AmBisome was designed as very rigid, small, unilamellar liposomes(<100 nm) with long circulation times (26). It has been shown thatAmBisome liposomes preferentially bind to fungal cells and in some non-elucidated mechanism penetrate the fungal cell wall and can there executeits toxicities. The effectiveness of AmBisome relative to free AmphotericinB has been tested and showed to be slightly better for AmBisome fortreatment of leukemia induced fungal infection (27). The side effect profileof AmBisome was significantly better than that of Amphotericin B (27).
In conclusion it can be said that the results of the today availablecommercial formulations using liposomes are not dramatically bettercompared to the free drugs, but since the toxicities seem to be lower in all
7
cases a higher dose might be given, thereby potentiating the use ofliposome formulations.
1.2.2 Tumor targeting liposomesTo increase the tumor-specificity of liposomes a targeting agent can beattached. There have been several strategies for attachment of tumortargeting agents, but the prevailing and most successful is to attach thetargeting agent to the distal end of a PEG molecule on the outside of theliposome. This has proved to increase the targeting ability compared to ifthe targeting agent is attached to the lipid head group (28). There have beenseveral conjugation-chemical approaches to achieve this (29-32).
Several studies, both in vitro and in vivo, have been performed usingtargeted liposomes; so far none have performed clinical studies though.Among the most studied tumor targets with liposomes are the folatereceptor, Human EGF receptor 2 (HER-2), CD-19 and the transferrinreceptor.
A number of studies with folate-targeted liposomes loaded withanthracyclines (33, 34) has been performed showing that the folate receptoris suitable for liposome delivery with high specificity and internalizationabilities. Several other therapies using liposomes targeting the folatereceptor have been suggested, such as antisense delivery (35) andphotodynamic therapy (PDT) (36). Liposomes targeting the folate receptorloaded with boronated compounds have also been studied with promisingresults (37-39).
The HER-2 has been studied as a target since early 1990´s. Suzuki et al(40) studied doxorubicin loaded liposomes with antibodies targeting eitherthe p185 residue or the p125 residue, and it was shown that targeting p185was superior. All further studies have targeted this epitope on HER-2.Goren et al. (41) showed 1996 that the uptake in cell culture was 16 timesbetter for HER-2 targeted liposomes than non-targeted. Kirpotin et al (42)constructed immunoliposomes with Fab-fragments targeting HER-2, andthey showed good binding and proven endocytosis in vitro. They alsoshowed that the number of targeting molecules on each liposome could beoptimized for increased uptake. For binding 40 Fab/liposomes was optimal,but for internalization a plateau was reached at 10-15 Fab/liposome (42).Park and co-workers (43-45) have studied HER-2 targeting extensively andhave shown therapeutic efficacy in several animal studies. They havetested liposomes with both Fab-fragments and single chain fragments,ScFv, against the p185 epitope, and both conjugates showed equal effect.Immunoliposomes loaded with doxorubicin have been shown to be better
8
in animal studies than free doxorubicin, non-targeted liposomaldoxorubicin or HER-2 antibody treatment (trastuzumab).
Lopez de Menezes et al (46-48) have targeted the tumor antigen CD-19successfully on B-lymphoma cells both in vitro and in vivo. They have alsoperformed ex-vivo experiments targeting CD-19 positive B-cells ofmultiple myeloma patient blood. The in vivo studies in mice showed thatdoxorubicin in immunoliposomes targeting CD-19 gave much better resultsthan free doxorubicin or doxorubicin in non-targeted liposomes. As a testof the specificity, liposomes with a non-idiotypic antibody was used withvery limited uptake.
The transferrin receptor has been studied using transferrin-conjugatedliposomes. Sarti et al. (49) showed that transferrin liposomes interactedspecifically with cultured cells and that they were internalized via receptormediated endocytosis. Iinuma et al. (50) developed cisplatin loadedtransferrin liposomes that proved to increase the cisplatin levels ofdisseminated tumor cells in ascites significantly. It was also shown byelectron microscopy that gold labeled transferrin liposomes were locatedon the plasma membrane of cultured cells or in endosomes in the processof endocytosis.
1.3 The EGFR-family and its ligands
1.3.1 ReceptorsThe epidermal growth factor receptor, EGFR, is a 170 kDa transmembranereceptor present in many non-haematopoetic human tissues. It is composedof three major domains: a cystein rich extracellular domain connected via atransmembrane lipophilic segment to an intracellular protein tyrosinekinase domain activated by ligand binding. The type 1 subclass of thetyrosine kinase receptor family does not only involve EGFR (ErbB-1), butalso HER-2 (ErbB-2/Neu), HER-3 (ErbB-3) and HER-4 (ErbB-4). They allhave the same basic structure as described above with high degrees ofhomology between the different receptors (51). The most highly relatedstructure is the intracellular tyrosine kinase domain and the least related isthe intracellular carboxylic terminal. The c-terminal region contains mostof the tyrosines that, when autophosphorylated after activation, attract andbind specific substrates or adapter proteins involved in downstreamsignaling (51).
All EGF-receptors are involved in the mediation of proliferation anddifferentiation of normal cells and their importance in development has
9
been shown by the study of genetically modified mice (52). EGFR lossleads to embryonic or perinatal lethality with mice showing abnormalitiesin multiple organs. HER-2 knock-out mice died at mid-gestation due tomalfunction in heart development. This phenotype is also shared by HER-4knockouts. Mice that lack HER-3 develop a few days further but still havenon-functional hearts and neural crest defects. These data show that theEGFR family plays critical roles in modulating specific aspects ofvertebrate development. In the adult organism the receptors are alsonecessary. For example, mammary gland development is dependent onEGFR function and lactation is dependent on HER-2 and HER-4. Loss ofHER-2 has been shown to delay the onset of puberty (52).
Table 2. The EGF-receptor family and its ligands
Receptor ligandEGFR EGF, AR, TGF-α, BTC, EPR, HB-EGFHER-2HER-3 HRG (NRG 1-2)HER-4 BTC, EPR, HB-EGF, HRG (NRG1-4)
1.3.2 LigandsEGF-family hormones are initially synthesized as membrane-anchoredprecursors that are subsequently cleaved to release soluble hormone. Theyare as mature hormones 50-60 amino acids long proteins and share a stronghomology throughout 50 amino acids, in which the important feature is sixcharacteristically spaced cysteines that form three intramolecular disulfidelinkages. This defines a secondary structure comprising two sets of antiparallell β-sheet structures with little or no helical conformation (53, 54).
The family consists of epidermal growth factor (EGF), amphiregulin(AR), transforming growth factor alpha (TGF-α), betacellulin (BTC),epiregulin (EPR), and heparin binding EGF (HB-EGF) that bind EGFR;the three latter bind HER-4 as well. Neuregulins (NRG), of which the mostwell known is heregulin (HRG, NDF, NRG-1), bind to HER-3 and HER-4(52-54). No natural ligand for HER-2 has yet been discovered. Increasingevidence suggests that it functions mainly as a co-receptor. If heregulinbinds to HER-3/4 and the receptor heterodimerizes with HER-2, it has beenshown that the ligand associates closely enough with HER-2 to be cross-linked (51). It has also been shown that HER-2 potentiates the other'ssignals when forming heterodimers (55).
EGF was first described by S. Cohen and was purified from mousesalivary gland. It was found to promote growth and development of
10
epidermal cells and was therefore named mouse epidermal growth factor,mEGF. It has 53 amino acids and a molecular weight of 6 kDa (56). Itfollows the general structure described above. H. Gregory described thehuman version later in the 1970s. It was found in human urine andinhibited gastric acid secretion. It was therefore named urogastrone.Human EGF and mEGF are cross-reactive and show 70% similarity of theamino acids and the three disulfide bonds are formed at the same relativepositions (57).
1.3.3 SignalingWhen a ligand binds its receptor dimerization takes place, either withanother receptor of the same sort (homodimer) or with another receptor ofthe same receptor family (heterodimer). The dimerization starts a signalcascade that is dependent on the activating ligand, the receptor and thedimerization partner (54). A signal cascade starts with receptorphosphorylation of specific c-terminal sites that provide binding sites foradapter proteins such as Shc, Crk and Grb2, or kinases such as Src, Chk,and PI3K. All receptors can activate Shc or Grb2 to start the mitogen-activated protein (MAP) kinase pathway that results in DNA replicationand proliferation. The HER-3 receptor is non-functional whenhomodimerized but very potent when dimerized with HER-2 (54). In factHER-2 is the preferred heterodimerization partner for all EGFR-familyreceptors (55).
The receptor is down-regulated as a result of a response feedback loop.EGFR down-regulation of the receptor is mediated by internalization anddegradation in lysososmes, a process known as endocytosis. (58). Tyrosinekinase activity greatly enhances this process by stabilizing receptorassociation with the endocytosis apparatus. (59). Eps15 is an example of anEGFR specific substrate that is involved in coated pit mediatedinternalization and is needed for endocytosis (60). This specificity forEps15 of EGFR is probably why EGFR is the only EGFR-family receptorthat is readily encocytosed when activated by its natural ligands (61).
1.3.4 In cancer developmentEGFR is overexpressed in a variety of tumor tissues, for example ingliomas the EGF receptor gene has been shown to be amplified (62-64), aswell as EGF-receptor mRNA (63). EGF-receptors have also been shown tobe overexpressed in bladder carcinoma, where the expression wasassociated with the stage and grade of the tumor, indicating a poorprognosis (65). Also in breast cancer the overexpression of EGFR is
11
correlated with poor prognosis (51). Breast cancer cells have been shownto be potential targets for EGFR directed therapy, since EGF labeled with111In has been shown to be selectively radiotoxic to breast cancer cells (66).EGFR overexpression has also been found in lung cancer (67) and intumors of head and neck (68). Not only overexpression of the receptor hasbeen found, the ligands EGF and TGF-alpha have proved to beoverexpressed in gliomas as well. In a study by Ekstrand et al. all testedglioma tumors had mRNA expression of one or both of the ligands (69).HER-2 receptor is known to be overexpressed in adenocarcinoma,especially in the breast (70) and in the ovary (71).
The EGFR family receptors have been shown to play many roles in thedevelopment of cancer, which might explain why they are overexpressed intumors. Co-expression of HER receptors and ligands leads to receptoractivation and stimulation of tumor cell proliferation and apoptoticresistance, thus providing a survival advantage. The HER-2 receptorsignaling pathway has also been shown to impact neoangiogenes and tumorcell dissemination at several levels (70). In the breast cancer cells thatoverexpress HER-2, EGFR primarily forms heterodimers with HER-2, andthe EGFR-HER-2 heterodimeres are impaired in EGF-induced endocytosisand downregulation. The impaired endocytosis leads to sustained signalingin response to EGF and subsequently stimulates the overproliferation andtransformation of breast cancer cells (51, 72, 73).
Overexpression of HER-2 in breast carcinoma cells induces a mitogenicphenotype. Overexpression of HER-2 alone has been shown to besufficient to increase cell migration and invasion. The EGFR-familyreceptors have also proved to mediate resistance to chemotherapy;expression and activation of HER-2 confers resistance to cisplatinum andtamoxifen but increases sensitivity to anthracyclins (70).
1.3.5 For cancer therapyIn therapy there is a number of ways that the activation pathway of theoverexpressed EGF receptor system can be targeted. Antibodies can beused as antagonists of the ligands, peptides that inhibit receptordimerization can be used, as well as drugs that block tyrosine kinaseactivity. Growth factors or antibodies conjugated to toxins or radionuclidescan target the receptor structures (74).
Several EGFR-targeting monoclonal antibodies that have been able toinhibit proliferation of a variety of human cancer cell lines in culture and inxenograft models (75) have been developed. The most successful so far ismAb 225, which has been shown to target lung tumor and metastases in aphase I study. Unfortunately, an anti-mouse response was elicited, so a
12
chimeric version has been developed in order to be able to be injectedrepeatedly without evoking any immune response against the injectedantibody. Since the tumor response after treatment with EGFR blockingantibodies is cytostatic rather than cytocidal, a tumoricidal moiety can beattached to the antibody. MAb 225 has been used as a delivery agent fortoxins such as exotoxin A, ETA (75). Antibodies, most notably mAb 225,have also been shown to act as chemosensitizing agents in combinationwith chemotherapy (76).
The EGF molecule itself can be used as a targeting agent for cancertherapy. EGF-polylysines have been developed for oligonucleotidedelivery and they have shown positive results in A549 cells (77). Severalreports have used EGF as a targeting agent, multiple EGF-dextranconjugates have been developed for both radionuclide and boron deliveryto tumors (78-80). Clinical trials have been undertaken studying the uptakeof EGF-dextran in bladder carcinoma using bladder instillations (81). Theuse of EGF-chelates for radionuclide delivery has proved very promising(82) and clinical trials towards brain tumors are not far away.
There have been several trials performed using the EGFR tyrosine-kinase inhibitor ZD1839, Iressa, and it is shown to be effective towardsvarious tumors with high selectivity for the EGFR. Its actions are mostlyconsidered to be cytostatic, but cytocidal effects have been shown incombination with chemotherapeutics such as cisplatin and taxanes and incombination with radiotherapy as well (83).
Trastuzumab (Herceptin) is a humanized antibody against HER-2 thathas proved to inhibit tumor growth in certain tumors (71, 84). Themechanisms behind this inhibition have been studied quite extensively andseveral hypotheses have been made. It has primarily been shown thattrastuzumab arrests cells in the resting G1 phase of the cellcycle (85).Trastuzumab is also known to cause DNA strand breaks in HER-2overexpressing cell lines BT-474 and SKBR-3. This might be the reasonfor synergy effects with chemotherapy, such as doxorubicin (86). Theantibody is also known to mediate apoptosis for example in SKBR-3 cells(87).
Trastuzumab have been most frequently studied for therapy of HER-2positive metastatic breast cancer, both as a single therapeutic agent (84)and in combination with chemotherapeutics (88). It was shown that theantibody inhibited tumor growth when used alone but had synergistic oradditive effect when used in combination with the most commonanticancer drugs (88). Unfortunately, the combination of trastuzumab andanthracyclin, though very potent towards tumor cells, proved to be toocardiotoxic (88). To prevent this toxic effect trials have been startedstudying the synergy of trastuzumab and Doxil®/Caelyx® (89).
13
1.4 BNCT Boron neutron capture therapy, BNCT, is a binary therapy system which is very appealing. The tumor cells are loaded with high amounts of boron and therafter irradiated with thermal neutrons (0.025 eV), resulting in the production of highly cell-toxic ionizing particles of short range with localization within the tumor cell.
Some atoms are known to have large cross sections for absorption of thermal or epithermal neutrons and 10B is one of them with a cross section of 3837 barn (1 barn = 10-28 m2). The naturally occurring boron is a mixture of the two isotopes 10B (20%) and 11B (80%). 11B has a 106 times smaller cross section for thermal neutrons. Therefore only 10B-enriched compounds are considerable in a therapeutic situation. Upon neutron capture 10B can undergo two reactions.
Figure 4. Boron neutron capture reactions
In order to be effective, at least 10-30 µg 10B need to be delivered per
gram tumor tissue corresponding to a concentration of 10-30 ppm. To obtain this concentration approximately 109 10B atoms need to be delivered to each tumor cell (90). The required dose is given by the dose limitations due to captures in naturally occurring elements in tissue. The most abundant are 1H and 14N with cross sections of 0.33 and 1.81 barns, respectively. The captures in 1H produce mainly gamma which gives a background dose to normal tissues in a therapeutic situation. 14N captures produces protons with a range of 10-11 µm, which can give dose limiting radiation effects in the irradiated area. A neutron fluence of about 1012-1013 thermal neutrons/cm2 is approximately the upper limit for normal tissue.
1.4.1 History The use of 10B for medical purposes was depicted as early as 1936 by Locher (91). The first clinical trial of BNCT was initiated at Brookhaven
10B + 1nth[11B]
4He2+ (1.78 MeV) R= 9.7 µm7Li3+ (1.01 MeV) R= 4.8 µm
4He2+ (1.47 MeV) R= 8.0 µm7 Li3+ (0.84 MeV) R= 4.2 µm γ (0.48 MeV)
(6.3%)
(93.7%)
14
national laboratory (BNL) by Farr and Sweet in 1951. From 1959 to 1961 aseries of patients received BNCT at the Brookhaven Medical ResearchReactor (BMRR) of BNL (92). The malignancy chosen to study wasglioblastoma multiforme, GBM, a localized brain malignancy with lowsurvival. Despite aggressive efforts, no powerful treatment modality hasemerged. Another group of patients was treated at the MassachusettsInstitute of Technology (MIT) during 1959-1961 (93). These trials usedfour different boron compounds. Clinical results from these studies weredisappointing and the last clinical BNCT trial for decades in the USA wasperformed in 1961 (92, 93). The disappointing results of all trials abovewere thought to depend on poor penetration of the thermal neutron beamand too little boron in the tumor; the tumor to blood ratios were less thanone. Experiments using higher fluencies to ensure therapeutic levels atdepth caused severe damage to the scalp in some patients; this might havebeen due to high boron concentrations in the blood (92, 93).
In Japan BNCT experiments were started in 1968. Dr Hatanaka, whohad collaborated with Dr Sweet, began clinical trials at the HitachiTraining Reactor (HTR) using sulfhydryl borane Na2B12H11SH (BSH).Almost 150 patients with various forms of brain malignancies, mostlyGBM, were treated. To ensure thermal neutrons in the tumor, Hatanaka etal irradiated their patients with open skulls. This was also done to preventdamage to the scalp, as seen in the US studies. Hatanaka showed somesuccess with his studies including 9 patients with more than 10-yearsurvival. Six of these were able to live a normal life without any signs ofdisease (94).
Hatanaka's results, though not undisputed, awakened the interest forBNCT in USA and Europe again. During the 1980's interest was focusedon the compound borophenylalanine (BPA). It is a boronated amino acidthat can be taken up in the cells by the naturally occurring amino acidtransport system. One of the major problems with BPA is low solubility,which can be overcome by complexion with fructose (95).
A new higher energy neutron beam, with epithermal neutrons, wasdeveloped at BMRR. The epithermal neutrons are slowed down to thermalneutrons in the skull and the brain tissue. The use of BPA for clinical trialsof GBM was approved by the FDA in 1994 and between 1994 and 1999 53patients were treated. The results from the first 38 patients indicated nosevere BNCT related toxicities. It was also shown that the time toprogression was comparable to that after conventional treatment (96). In1996 another study using BPA was initiated at MIT. In the MIT study 22patients had been treated by 1999 and mixed results were obtained. Twopatients exhibited a complete radiographic response, and 13 of 17 patientshad measurable reduction in tumor volume for the first months after
15
irradiation, after which the disease either stabilized or increased. A numberof acute side effects were noted, in particular effects due to increasedintracranial pressure (97, 98). In Petten, the Netherlands, a study usingBSH was started in 1997. After 26 treated patients it could be determinedthat no dose-limiting toxicities had been observed and no conclusionsregarding the efficiency of the treatment compared to conventionaltreatment could be drawn (99, 100). Another study using BPA was initiatedin 1999 in Finland, in which 18 patients without earlier radiation treatmenthave been treated. The one-year survival is estimated to be 61%. Adifferent protocol accepting patients with previous radiotherapy has alsostarted and no serious acute BNCT-related adverse effects has beenencountered (101).
In 2001 the first BNCT study in Sweden was initiated at the StudsvikR2-0 reactor using an epithermal neutron beam, and for the first 17 patientsno severe BNCT related acute toxicities have been observed. Thecompound used is BPA and high blood boron levels have been reached(102, 103). It is still too early to evaluate the efficacy of the treatment.
1.4.2 Compounds for BNCTThe compounds mainly used for clinical studies so far are the lowmolecular weight compounds BSH and BPA. In order to deliver moreboron, carborane compounds were designed. A carborane is a boron cagecontaining 10 boron atoms. The first carborane compound constructed wasan analog to BPA known as carboranylalanine, CBA. CBA was shown toimprove boron uptake compared to BPA in vitro but not in vivo (104, 105),probably due to the lipophilicity of the carboranyl group (104).
Boron-containing nucleosides have also been of interest, asbiosynthetically active tumor cells need building blocks for their DNA.Boron-containing moieties, either as a single dihydroxyboryl group or as acarboranyl moiety, have been inserted into pyrimidine nucleosides (106).
Other low molecular weight compounds have been of interest, such asboron containing porphyrines, of which the most studied is a boronatedprotoporphyrin known as BOPP (107). Since the protoporphyrin is asensitizer for near infrared irradiation, the boronated protoporphyrin can beused for dual treatment, both BNCT and photodynamic therapy, PDT.BOPP was shown to augment boron uptake compared to BSH and BPA(107, 108) and bind selectively to glioma cells in vivo (109).
Proximity to DNA increases the lethality of the capture reaction (90)and therefore large interest has been put into the development of DNA-intercalating boron compounds based on the well known DNA-intercalating groups phenylphenantridine (110) and acridine (111). A
16
compound based on naphtalemide has also been developed (112). Boronrich oligomeric posphate diesters have also been shown to have DNA-binding properties (113, 114).
All compounds described so far are not tumor specific, or their tumorspecificity is based on the increased metabolism of tumor cells. For tumorspecific delivery, efforts have been made to develop boron-carryingantibodies or antibody fragments (114, 115). The problem with boroncarrying antibodies is to obtain enough boron on each antibody, at least 103
boron atoms per antibody is needed (115), and still retainimmunoreactivity. If instead a carrier molecule with the possibility todeliver large amounts of boron is attached to the targeting molecule thisproblem might be circumvented. The most studied carrier molecules fortumor specific delivery of boron are starburst dendrimers (116), lowdensity lipoproteins, LDL (117, 118), boronated dextrans (79, 80) or boroncarrying liposomes.
1.4.3 BNCT and liposomesLiposomes have been proposed as delivery agents for BNCT and haveduring the last decades been studied both with and without a targetingagent on the liposome.
Hawthorne and co-workers reported the use of liposomes as deliveryagents for boron in 1992 (119). They showed that liposomes with meandiameters of 70 nm or less were capable of encapsulating highconcentrations of water soluble ionic boron compounds, and that they wereable to deliver this boron to subcutaneous tumors in mice. The boronconcentrations reached over 15 ppm and the tumor to blood ration was over3. Further studies (120) showed even better tumor uptake, 30-50 ppm andtumor /blood ratio of 5. They obtained similar results using liposomeswhere lipophilic boron compounds were encapsulated in the membrane(121). By using liposomes combining the two approaches, with bothhydrophilic and lipophilic boron compounds, tumors in mice could besuccessfully targeted (122). The boron concentration reached 50 ppm and atumor to blood ratio of 5-10 was obtained.
Metha et al (123) have studied BSH in liposomes with and withoutPEG. They were shown to have significant improvement in circulationtime compared to free BSH after tail vein injection in mice. The circulationtime proved to be the highest for the PEG-stabilized liposomes.
Moraes et al (124) developed liposomes loaded with the compound o-carboranylpropylamine (CPA). The results showed that the compoundcould be loaded into liposomes at a concentration of 104 molecules/vesicle.Both PEG-stabilized and conventional liposomes were studied. It was
17
shown in cell culture that CPA toxicity decreased after liposomeentrapment (CPA concentration 0.15 mM).
Table 3. Experiments with liposomes for boron delivery
Liposomeformulation
Boron compound Target tissue Reference
Non-targeted Ionic, watersoluble compounds
Tumors in mice (119,120)
Non-targeted Lipophilic boroncompounds
Tumors in mice (121)
Non-targetedw/wo* PEG
BSH Mice without tumors (123)
Non-targetedw/wo* PEG
CPA Cultured glioma cells andlymphocytes
(124)
Anti CEA-liposomes
Ionic water solublecompounds
Pancreatic cancer cells invitro and tumors in mice
(126,127)
Non targetedliposomes
Ionic water solublecompounds
Breast cancer cells invitro
(125)
Folate-liposomeswith PEG
Lipophilic boroncompounds
KB cells in vitro (39)
Folate-liposomeswith PEG
Anionic boroncompounds
KB cells in vitro (37)
Folate-liposomeswith PEG
Anionic boroncompounds
Transplanted lung cancerin mice
(38)
* w/wo , with or without
The experiments described above used liposomes without a targetingagent. Yanagie et al. (125-127) were the first to examine targetingliposome as a mean for boron delivery. They targeted the carcino-embryonic antigen, CEA, on pancreatic cancer cells and experiments incell-culture showed that liposomes targeting CEA and enclosing 10B-compounds could, after neutron irradiation, inhibit cell-growth. The cell-growth was inhibited to approximately 30 % after irradiation with 5�1012
nth/cm2. It was also shown, using the same conjugate, that pancreatic cancertumor cells transplanted to mice could be growth inhibited after CEA-10B-liposome targeting and neutron irradiation. The same group has shown thatthe growth of breast cancer cells could be inhibited by use of liposomeswith 10B and neutron irradiation. No targeting agent was used in this study(125).
Recently Lee and co-workers in a series of publications studied the useof the folate-receptor for liposomal delivery of boron. Pan et al. (37)showed in a study that as much as 1500 µg boron /109 cells could bedelivered. However, the uptake could not be blocked by adding an excessof folate. Using a different boron compound, specificity could be obtained,
18
and the boron concentration was still very high (around 500 µg /109 cells).In the next study (38) it was shown that boron could be delivered toimplanted tumors in mice using folate receptor targeting liposomes. Thebiodistribution showed tumor to blood ratios up to 6 after 96 h. In anotherstudy a lipophilic boron compound was used and the uptake of folate-targeted liposomes in KB-cells was examined (39). It was shown that itwas possible to deliver 587 µg boron/109 cells with high specificity. Thecorresponding uptake in cells not expressing the folate receptor was lessthan 10%.
19
2. Aims
The aim of this thesis is to develop liposome-ligand conjugates withspecificity for EGFR or HER-2, and to study these conjugates in cellculture with regard to stability, uptake specificity and intracellularprocessing. Further, to evaluate the cell killing potency of the liposomeconjugates in an in-vitro system after neutron activation. Finally, toevaluate the use of liposome conjugates for systemic injection by studyingthe biodistribution of EGF-liposomes compared to EGF in mice.
20
3. Materials, methods and techniques
Some of the most important materials and methods used in this thesis arepresented here. For a more thorough description the reader is referred tothe enclosed papers.
3.1 LiposomesLiposomes are, as previously mentioned, phosphlipid bilayer spheres. Theones used in this study were composed of disteaoryl phosphatidylcholine( D S P C ) , ( 5 7 % ) , c h o l e s t e r o l ( 4 0 % ) a n d disteaorylphosphatidylethanolamine-polyethylenglycol (DSPE-PEG), (3%). Thisgiving rigid, stable and long circulating liposomes (128).
The liposomes were prepared by freeze-thawing and extrusion. Thelipids were dissolved in chloroform and dried to a lipid film that washydrated and heated to 60˚C and thereafter frozen in liquid nitrogen. Thefreezing and heating was repeated seven times. The liposomes were thenextruded ten times through a 100 nm filter to obtain liposomes of similarsize. The size 100 nm was chosen because liposomes of that size have beenshown to be the most stable in blood circulation (128).
Figure 5. Cryo-TEM picture of PEG stabilized liposomes loaded with WSA. Thedrug can be seen as the dark globular spots inside the liposomes. Bar is 100 nm.Photo kindly supplied by Markus Johnsson.
21
To get the boronated compounds into the liposomes, active loadingtechniques based on pH-gradients were used. Briefly, the liposomes havebeen prepared using a buffer with low pH (pH 4) and after extrusion the pHon the outside of the liposomes was raised to 7-8. The compounds forloading, in this case weak bases, were added to the solution and entered thecore due to equilibrium. As a result of the lower pH on the inside, thecompounds were protonated and thus trapped inside. This trapping ofcompound made the gradient behave like a pump, and approximately 98%of added compound entered the liposome core. The molar loading ratios forthe compound used in the papers herein were 0.2:1 and 0.1:1 (compound:liposome), giving approximately 104 molecules per liposome. As much as105 molecules loaded per liposome have been performed duringdevelopment of the production (129).
Figure 6. The principle for pH-gradient loading of drugs. Figure kindly providedby Markus Johnsson.
3.2 Boronated DNA-binding compoundsIt has been shown that if the boron is located close to the DNA lessamounts are needed (90). Therefore, DNA-binding boronated compoundshave been developed. Two of them, water soluble boronated phenantridine,WSP, and water soluble boronated acridine, WSA, have been used in thisthesis (110, 111) and were loaded into liposomes using the pH gradientloading technique described above (129).
WSP is composed of a phenylphenantridine chromophore coupled to aboron cage (10 boron atoms) and was used in paper II. The mostcommonly known phenylphenantridine is ethidium bromide, a widely usedDNA stain in life science. The phenantridine ring system intercalates in themajor groove, and the amino groups at position 3 and 8 are located insidethe helix and interact horizontally with the sugar phosphate backbone. Thephenyl group and the substituent at position 5 are externally positioned in
Drug = B:
Addition of drugTitration
Na2CO3
pHout = 4 pHout = 7-8
A3-
pHin = 4 pHin = 4
H+H+ H+
A3-
H+ H+H+
H+H+
H+
A3-
[(BH+)3A3-]
22
the minor groove. The substituent at position 5, where the boron cage isattached in WSP, has little or no effect on DNA binding (130, 131). Thephenantridinium chromopohore is fluorescent with an excitation around546 nm.
WSA is composed of 9-aminoacridine, a boron cage with 10 boronatoms and spermidine tails that have been added to increase the watersolubility and the affinity for DNA. The compound has been used in paperI-V. The acridine chromophore is known to interact with DNA in differentways. In one case the molecule is stacked between G-C base pairs with the9-aminogroup pointing into the minor groove. In another case the drugmolecule is intercalated asymmetrically between the bases of one strandonly (130, 132). This depicts that chains attached to the 9-aminogroup (asthe boron cage and spermidine tails in the case of WSA) do not interferewith the interaction (133). Boron-containing acridines were synthesized asearly as 1967 (134), and the acridine moiety has also been proposed as ananti tumor agent labeled with, for example, 125I (135-137). The most well-known acridine compounds are the vital dye Acridine Orange and the drugProflavine. WSA is fluorescent and was studied using excitation at 488 nm.
Figure 7. The chemical structures of the DNA binding compounds WSP andWSA.
23
As a comparison regarding DNA targeting we have also studieddoxorubicin, a well-known anti-neoplastic agent that intercalates DNA.Doxorubicin has been extensively studied in liposomes. The commerciallyavailable liposome formulation Doxil®/Caelyx® uses doxorubicin asactive agent. Doxorubicin was used in paper III and is also fluorescent withexcitation 488 nm.
The toxicity of WSA and WSP were tested, using clonogenic survival incell culture. WSP was shown to decrease survival significantly even at low(1 µg/ml) concentrations, while WSA was much better tolerated even up to20 µg/ml. The toxicity was reduced for both compounds if they wereenclosed in PEG-stabilized liposomes. For WSP the reduction in toxicitywas striking: even at the concentration 20 µg/ml, only a slight reduction insurvival could be seen (138).
3.3 Cellular modelsThe work in this thesis is largely based on studies of cultured tumor cells.The receptor specificity and intracellular processing of the conjugates canbe studied in a cell-culture system, but more complex aspects, such asdistribution and tumor selection, need to be investigated in studies usinganimal models. Tumor cells can be grown as monolayer culture on thebottom of culture dishes, as cell suspensions in culture medium in flasks, oras spheroids, three-dimensional cellular clusters. For this thesis three cell-lines have been used, all express the targeted receptors to a high extent.
A glioma cell line, U-343 MGa Cl2:6, has been used in paper I-III andV. The cell line is known to overexpress EGF-receptors and hasapproximately 105 receptors per cell (139). These cells have been used formonolayer culture, in pellet cell-suspension and in roller-flasks. These cellswere also used for clonogenic survival after experimental BNCT in paperV.
A squamous carcinoma cell line, A-431, has been used in paper III andwas chosen because of its large overexpression of EGF-receptors,approximately 106 receptors per cell (140). This cell line has been used formonolayer culture only.
A breast-cancer cell line, SK-BR-3, overexpressing the HER-2 receptor,the average number of receptors is 106, was used in paper IV. The cellswere used in monolayer culture and in roller flasks.
All cell lines were grown in Ham´s F-10 medium, supplemented with10% foetal calf serum, L-glutamine (2 mM) and PEST (penicillin 100IU/ml and streptomycin 100 µg/ml) using humidified air at 37˚C with 5%CO2.
24
3.4 Boron determinationsIn order to measure the boron concentration in samples the cells weredigested in HNO3 under heat and pressure (141). After digestion thesamples were diluted in MilliQ water with 5% HNO3 and then measuredusing either inductively coupled plasma-mass spectrometry (ICP-MS) orinductively coupled plasma-atomic emission spectrometry (ICP-AES).
In both procedures the liquid sample was injected in the inductivelycoupled plasma and the sample was atomized and ionized by the hightemperature of the plasma (6000 K). Detection of which atoms werepresent was then performed with mass spectrometry or atomic emissionspectrometry. The detection limits for boron of the instruments used areabout 5 ppb and 50 ppb, respectively. ICP-MS was used for borondeterminations in paper II-V and ICP-AES was used in paper V.
3.5 Neutron irradiation at the BNCT facility inStudsvikA facility for BNCT has been built at the R2-0 research reactor at Studsvik.The filter/moderator system used for this study is the same that is currentlyemployed for BNCT clinical trials (102, 103). The transport of neutronsgenerated by the reactor core, with a mean energy of 2 MeV acrossdifferent materials (the filter), results in a final spectrum enhanced in theenergy range between 0.4 eV and 10 keV. This is the range that is usuallyreferred to as epithermal energy range. The slowing-down process (i.e. theshift of the neutron spectrum toward lower energies) is mainly driven bythe combination of different layers of aluminum and Teflon, while theremoval of the thermal neutron component, which cannot go deep into thetissue, is performed by a 6Li filter positioned just at the end of the beam.The cell sample irradiations of this study took place in a 20�20�15-cm3
PMMA phantom at a depth of 3 cm, where it had previously been foundthat the peak of the thermal neutron distribution produced is located.
3.6 AnimalsMice were used to study the biodistribution of EGF and EGF-liposomes inpaper VI. Female NMRI mice were used, and they were housed in acontrolled environment and fed ad libitum. Mice are known to have a largequantity of EGF receptors primarily in the liver (142), which makessystemic injection for tumor targeting with EGF less useful.
25
4. Preparation of liposome conjugates
The thesis is based on targeted stabilized liposome conjugates(summarized in table 4), and examines their construction and theirbehavior in vitro and in vivo. The most important aspects regarding thepreparation and the optimizations made are presented in this chapter. Formore thorough descriptions on the methods and materials used the reader isreferred to the enclosed papers.
Table 4. The liposome conjugates used in paper I-VI
Liposome conjugate Conjugation method Used in paper125I-EGF-liposome Maleimide-PEG-DSPE I125I-EGF-liposome-WSA —"— I-III125I-EGF-liposome-WSP —"— II125I-EGF-liposome-DOX —"— IIIEGF-3H-liposome-WSA —"— III125I-EGF-liposome-WSA NHS-PEG-DSPE V125I-EGF-liposome —"— VI125I-Trastuzumab-liposome-WSA —"— IVTrastuzumab-3H-liposome-WSA —"— IV
4.1 Micelle-transfer procedureLiposome conjugates were prepared using the micelle transfer (post-insertion) technique (figure 8) (31, 143, 144). Briefly, liposomes areprepared and loaded with the desired compound separately, and thetargeting agent is added afterwards. The targeting agent is conjugated tothe distal end of a PEG3400-DSPE lipid. The ligand-PEG-DSPE lipids formmicelles in solution that can be mixed with the liposomes. At a highenough temperature the ligand-PEG-DSPE lipids then transfer from themicelles into the liposome membrane.
26
Figure 8. Schematic drawing of the micelle-transfer method. EGF/trastuzumab isattached to DSPE-PEG-maleimide/NHS lipids in micelles. The EGF /trastuzumablipids, in the form of micelles, are mixed with preformed liposomes, and theEGF/trastuzumab-PEG-DSPE molecules are thereby incorporated into theliposome membranes.
The use of this method is appealing in the sense that there is no need tochange the whole conjugation procedure if a new targeting agent or a newload is desired. Another advantage is that the targeting agent is known tobe situated in the outer membrane. Further, the fact that PEG3400 is used forthe ligand conjugation and PEG2000 is used for stabilization in the liposomemembrane gives an extra long spacer-arm for the ligand, making sure thatit is extended for maximal contact with the receptors.
4.2 Radiolabeling of the targeting agentEGF or Trastuzumab was labeled with 125I using the Chloramine-T (CAT)method. This is a direct labeling method in which the CAT oxidizes theiodine ion so it becomes positively charged. The iodine ion then undergoeselectrophilic substitution with tyrosine residues. The reaction can takeplace under physiological pH when labeling with 125I and can therefore beperformed with largely preserved biological activity of the labeled protein.
EGF/trastuzumab
maleimide/NHS-PEG-DSPE micelles
EGF/trastuzumab-lipid micelle
preformed liposome EGF/trastuzumab-liposome
EGF/trastuzumab-lipid micelle
27
4.3 Optimization of micelle-transfer conditions forEGF-liposomes using maleimide-PEG-DSPE (paper I)125I-EGF was modified with Traut´s reagent (2-iminothiolane) to get a free–SH group for conjugation. The modified 125I-EGF-SH was conjugated tomaleimide-PEG-DSPE for 24 h in room temperature. After purification,the EGF-lipid was transferred to the preformed liposomes as describedabove. The transfer conditions were optimized in paper I. The optimizedparameters were temperature, time and initial PEG concentration in thepreformed liposome (figure 9).
The times used for incorporation were 1, 4 and 24 h. For unloadedliposome no significant increase was apparent after 1 h, Therefore 1 h wasconsidered optimal. For the WSA loaded liposomes longer incorporationtime, 15 h, was used for the experiments in paper II since the lowincorporation yield made us want to maximize the number of incorporatedEGF-lipids. However, it was later concluded that the increase in yield withtime was marginal for the WSA loaded liposomes as well, and the longertime only constituted an increased risk for EGF degradation. Therefore 1 hincorporation time was used for the experiments in paper III.
The temperatures studied were room temperature (RT), 37˚C and 60˚C.It was concluded that 60˚C was by far the most efficient transfertemperature for all experiments, and it was shown that the degradation wasnot extensive enough to defend use of a lower temperature.
The initial PEG-concentration in the preformed liposomes was alsostudied: 0, 3 and 5 % PEG was used. No clear difference could be noted forthe three concentrations, but the yield seemed to be a little lower for thehighest concentration at short times. It is beneficial to have PEG initially inthe liposomes for stability, therefore 3 % PEG was used in the preformedliposomes for all experiments. It was also found that after incorporation ofthe ligand-PEG-lipids the PEG concentration in the outer layer wasapproximately 5%, which is found to be optimal for stability (145).
It is interesting to note that the WSA-loaded liposomes exhibit an overall lower yield compared to unloaded liposomes. This phenomenon isshown for all EGF-targeted WSA-loaded liposomes. The incorporationyield in WSP and doxorubicin loaded liposomes (which have lowerconcentrations of loaded drug) is more similar to unloaded liposomes. Forthe WSA-loaded liposomes the final concentration was 10-15EGF/liposome.
The conjugate was shown to be stable; at 37˚C 79% of the EGF-associated radioactivity remained in the liposome fraction after one weekand at 4˚C 90% remained after 3 weeks.
28
Figure 9. Incorporation of 125I-EGF-PEG-DSPE lipid molecules into preformedliposomes. The percentage of added lipid incorporated in the liposome membranewas measured using the 125I-label on EGF. The amount of added EGF lipid was2.7% to total liposomal lipid.A: 0% PEG in the preformed liposomes, B: 3% PEG-DSPE to total lipid, C: 5%PEG-DSPE to total lipid, D: WSA loaded liposomes with 3% PEG-DSPE, to totallipid. Error bars represent maximal errors from double experiments. Scale on y-axis represents percentage incorporated EGF-lipid.
4.4 Optimization of conditions for trastuzumab-liposomes using NHS-PEG-DSPE (paper IV)125I-trastuzumab solution was added to NHS-PEG-DSPE, and afterhydration at 60˚C for 5 min, to allow for the lipids to get into micelles, themixture was kept at room temperature for 1 or 3 h conjugation. No increasein yield could be seen after 1 h. Therefore 1 h conjugation was consideredsufficient. 125I-trastuzumab-PEG-DSPE micelles were purified andthereafter mixed with the preformed liposomes at 60˚C, for 1, 4 or 24 h. Noresults were obtained from incorporation for 24 h at 60�˚C since thetreatment caused the samples to stick to the column and no separation waspossible. Therefore 4 h was considered to give the best yield. The emptyliposomes gave higher yields than the WSA loaded for both 1 and 4 h as
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
0 1 0 2 0 3 0
transfer time (h)
RT
37˚C
60˚C
D
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
0 1 0 2 0 3 0
transfer time (h)
RT
37˚C
60˚C
A
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
0 1 0 2 0 3 0
transfer time (h)
RT
37˚C
60˚C
B
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
0 1 0 2 0 3 0
transfer time (h)
RT
37˚C
60˚C
C
29
was previously shown with EGF-liposomes. The final yield fortrastuzumab-liposome-WSA was approximately 35 antibodies/liposome.
The conjugate was stable. After two weeks in 37˚C still 95% of theligand-associated radioactivity remained in the liposome fraction.
4.5 Preparation of EGF-liposomes using NHS-PEG-DSPE (paper V)The preparation procedure of the EGF-liposome conjugate used in paper Vwas slightly different from the one in paper I-III (optimized in paper I). Asfor the trastuzumab-liposome conjugates NHS-PEG-DSPE was usedinstead of maleimide-PEG-DSPE and no thiolation of EGF was necessary.The times and temperatures used for conjugation and incorporation oftrastuzumab-lipids into liposomes were adapted for these EGF-liposomesas well. Since no modification of the protein was necessary onepurification step could be removed, this together with the slightly higherconjugation and incorporation yields obtained made the over all yield abouttwice as good as for the procedure in paper I. The final result concentrationwas 20-30 EGF/liposome.
4.6 3H-labeled liposome conjugates (paper III and IV)To prepare radiolabeled liposomes, 3H-cholesteryl ether was used. This is anon-exchangeable, non-degradable lipid marker that is used to study, forexample, the cellular uptake of liposomes (146, 147). The 3H-cholesterolwas added to the lipid mixture before freeze-thawing and was thusincorporated in the lipid membrane. For the attachment of a ligand themethods described above was used.
4.7 CommentsWe have been able to develop stabilized liposome conjugates for targetingof EGFR and HER-2. The conjugates showed good stability. The amountof targeting agents per liposome, 10-30 for the EGF-liposome conjugatesand around 35 for trastuzumab-liposome conjugates, is assumed to beenough for a good cellular binding and uptake, at least when antibody-liposome conjugates have been studied (29, 42). The increase from 10 to30 EGF per liposome did not yield any extra cellular uptake either (datanot shown).
30
The clear difference in incorporation yield between WSA loaded andnon-loaded liposomes is interesting. A similar study published by Ishida etal (144) reports less incorporation yield into Doxil®/Caelyx® liposomesbut conclude that it is the increased PEG concentration in these liposomes(5% compared to their unloaded liposomes with 4% PEG) that hinders thelipids from transferring to the liposome membrane. We do not exclude thepossibility that the PEG concentration in the preformed liposomesinfluences the amount of incorporation but believe that the liposome loadalso plays a large part, at least in this study.
31
5. Cell experiments
In the studies of uptake and intracellular processing of the conjugatesmonolayer cell cultures were used. Most of the experiments (summarizedin table 5) have been performed analyzing the radioactive label on thetargeting agent (125I-EGF or 125I-trastuzumab), the radioactive label in theliposome membrane (3H-cholesterol) and the boron from the liposome load(WSA or WSP).
Table 5. Performed in vitro experiments
Liposome conjugate Cell line Experiment Paper125I-EGF-liposome U343 Displacement I125I-EGF-liposome-WSA,EGF-3H-liposome-WSA
U343A-431
Displacement,Time dependent uptakeRetentionInternalization studiesFluorescence studies
I, IIII, III
125I-EGF-liposome-WSP U343 Time dependent uptakeFluorescence studies
II
EGF-liposome-DOX U343A-431
Fluorescence studies III
125I-EGF-liposome-WSA U343 Uptake in cell-suspensionClonogenic survival
V
125I-trastuzumab-liposome-WSAtrastuzumab-3H-liposome-WSA
SKBR-3 DisplacementTime dependent uptakeRetentionInternalization studiesFluorescence studiesUptake in cell-suspension
IV
5.1 Test of receptor specificityTests where the receptors have been blocked by an excess of native non-radiolabeled targeting agent have been studied for both EGF andtrastuzumab conjugates. Displacement curves where increasing
32
concentrations of the blocking agent have been added show receptorspecificity for both liposome conjugates (figure 10, A, B). The unspecificbinding was shown to be approximately 10-20 % for EGF-liposomes and30% for trastuzumab-liposomes.
Figure 10. Displacement of A) 125I-EGF-liposome-WSA and B) 125I-trastuzumab-liposome-WSA after 4 h incubation. The maximal uptake was set to 1.
5.2 Time-dependent uptakeThe time dependent uptake has been studied with and without blockedreceptors for both 125I and 3H labeled conjugates (figure 11 A-D). It wasshown for both 1 2 5I-EGF-liposome and 1 2 5I-trastuzumab-liposomeconjugates that the uptake of 125I reached a plateau after 8-24 h (figure 11A and C). In the case of 1 2 5I-trastuzumab-liposomes the uptake evendecreased after 24 h. When looking instead at the uptake of 3H-liposomeconjugates the uptake is continuous throughout the incubation times (figure12 B and D). The most probable explanation for this is that in the case of125I, EGF or trastuzumab is degraded and excreted so that the measuredradioactivity does not correspond to the total uptake. The liposome label,3H-cholesterylether, instead is known to be a non-exchangeable, non-degradable marker (147), so this label should represent the true uptake ofligand-liposome conjugates.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0.001 0.1 1 0 1000 100000
ng EGF
A-431
U-343
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0.0001 0.01 1 100
µg trastuzumab
SKBR-3
A B
33
Figure 11. Time dependent uptake of A) 125I-EGF-liposome-WSA. B) EGF-3H-liposome-WSA. C) 125I-trastuzumab-liposome-WSA. D) trastuzumab-3H-liposome-WSA. Maximal uptake was set to 1. Unspecific uptake was determinedby addition of an excess of EGF (A, B) or trastuzumab (C, D) to block thereceptors.
The time dependent uptake of boron is shown in table 6. It was shownthat in monolayer culture that the uptake of boron increases with time toapproximately 6 ppm after 24 h incubation for EGF-liposome-WSA. EGF-liposome-WSP only reached 2 ppm and thus did not show any increasefrom the uptake with non-targeted liposomes. The uptake of WSA andWSP per se were 6.6 and 10.5 ppm, respectively. The choice was tocontinue further studies with WSA instead of WSP. One possibleexplanation for the difference might be that only half the amount of boroncould be loaded into the liposomes when using WSP as compared to WSA.With trastuzumab-liposome-WSA the monolayer culture uptake after 4 hincubation reached 12 ppm.
trastuzumab- 3 H-liposome-WSA
0
0.2
0.4
0.6
0.8
1
1.2
0 2 0 4 0 6 0
incubation time (h)
+trastuzumab
-trastuzumab
1 2 5I-trastuzumab-liposome-WSA
0
0.2
0.4
0.6
0.8
1
1.2
0 2 0 4 0 6 0
incubation time (h)
+trastuzumab
-trastuzumab
1 2 5I-EGF-liposome-WSA
0
0.2
0.4
0.6
0.8
1
1.2
0 1 0 2 0 3 0incubation time (h)
+EGF(U343) -EGF(U343) +EGF(A-431) -EGF(A-431)
EGF- 3H-liposome-WSA
0
0.2
0.4
0.6
0.8
1
1.2
0 1 0 2 0 3 0incubation time (h)
+EGF(U343) -EGF(U343)
+EGF(A-431) -EGF(A-431)
BA
C D
34
Table 6. Uptake of boron in monolayer cell culture (ppm).
Conjugate Cell line 1h 4h 24hWSP U343 10.49 +/- 1.88Liposome-WSP —"— 1.80 +/- 0.22EGF-liposome-WSP
—"— 2.21 +/- 0.22
WSA —"— 6.61 +/- 1.18Liposome-WSA —"— 0.29 +/- 0.07EGF-liposome-WSA
—"— 1.0 +/- 0.03 2.4 +/- 0.3 6.0 +/- 1.7
EGF-liposome-WSA
A-431 1 +/- 0.04 1.4 +/- 0.03 4.5 +/- 1.5
Trastuzumab-liposome-WSA
SKBR-3 12 +/- 0.4
5.3 RetentionThe retention of the conjugates after 24h incubation was studied withregard to the targeting agent, the liposome, and the load. For both EGF andtrastuzumab conjugates the retention of 3H and boron was very good(figure 12 A-C), corresponding to the retention of the liposome and WSA.For EGF-liposome-WSA approximately 80% of the boron remained inboth cell lines after 48 h incubation. For trastuzumab-liposome-WSA 70%of the boron remained after 48 h. The 3H-liposome label remained almostas long for both conjugates. There was a different result when the retentionof 125I-EGF and 125I-trastuzumab was studied. For 125I-EGF-liposome-WSAapproximately 50 % was rapidly excreted during the first 4 hours and after48 h only 20-30% remained. For 125I-trastuzumab-liposome-WSA theligand was also rapidly excreted and 18 % remained after 48 h.
35
Figure 12. Retention of 125I-EGF-liposome-WSA, EGF-3H-liposome-WSA andEGF-liposome-WSA(boron) in A) U343 MGa Cl2:6 cells and B) A-431 cells.Figure C) 1shows 125I-trastuzumab-liposome-WSA, trastuzumab-3H-liposome-WSA and trastuzumab-liposome-WSA(boron) retention in SKBR-3 cells. Theretention was, in all cases, studied after 24 h incubation and the uptake at 24 h wasset to 100 %.
Retention in SKBR-3 cells of trastuzumab-liposome-WSA
0
2 0
4 0
6 0
8 0
100
120
0 1 0 2 0 3 0 4 0 5 0 6 0time (h)
125I-trastuzumab-liposome-WSAtrastuzumab-3H-liposome-WSAtrastuzumab-liposome-WSA (boron)
C
Retention in U343 cells of EGF-liposome-WSA
0
20
40
60
80
100
120
0 10 20 30 40 50 60
time (h)
A
Retention in A-431 cells of EGF-liposome-WSA
0
2 0
4 0
6 0
8 0
100
120
140
0 1 0 2 0 3 0 4 0 5 0 6 0time (h)
125I-EGF-liposome WSA EGF-3H-liposome-WSA
EGF-liposome-WSA (boron)
B
36
5.4 Membrane-bound and internalized conjugateThe internalization of the conjugates was studied using the acid-washtechnique developed by Haigler (148). In short, the cells were incubatedfor the various times and thereafter washed to remove unbound conjugate.Glycine-HCl acid (pH 2.5) was added to the cell dishes for 6 min at 4˚Cand is thought to disrupt the binding of receptors and ligands on the cellsurface, i.e., the membrane bound conjugate ends up in the acid fraction.After another wash with the acid, 1 M NaOH was added to detach the cells.The radioactivity in the base fraction consisted of the internalizedconjugate.
It was shown for both conjugates that the majority was internalizedfairly rapidly. For EGF-liposome-WSA there was a difference between thetwo cell-lines tested. In U-343 MGa Cl 2:6 over 90% of the conjugate wasinternalized even after short times. In A-431 the majority of the conjugatewas internalized but only after longer times. This difference might beexplained by the fact that A-431 cells recycle their EGF receptors.
Trastuzumab-liposome-WSA was also shown to be mostly internalized,about 70% of the total cell-associated radioactivity was internalized whenboth the label on the liposome and on the antibody were studied (figure13).
5.5 Intracellular localization using fluorescenceBoth boronated compounds WSA and WSP are, as mentioned earlier,fluorescent and this could be used for intracellular localization studies.When incubating U-343 MGa Cl2:6 cells with the compounds per se, anuclear staining could be seen indicating that the compound in itself isDNA-binding (figure 14 A, B). When enclosing the compound in targetedliposomes the staining is mostly cytoplasmic (figure 14 C,D). Thisphenomenon was most pronounced for cells incubated with EGF-liposome-WSA, which showed no nuclear staining at all. Even after prolongedincubation time the WSA did not seem to reach the nucleus. The sameexperiments have also been performed with trastuzumab-liposome-WSAand no clear nuclear staining was apparent (figure 14 F).
37
Figure 13. Internalization of A) 125I-EGF-liposome-WSA in U343 cells. B) EGF-3H-liposome WSA in U343 cells. C) 125I-EGF-liposome-WSA in A-431 cells. D)EGF-3H-liposome-WSA in A-431 cells. E) 125I-trastuzumab-liposome-WSA inSKBR-3 cells. F) trastuzumab-3H-liposome-WSA in SKBR-3 cells. Filled bars:internalized, empty bars: membrane-bound
Earlier publications have shown that doxorubicin loaded targetedliposomes have nuclear staining after cellular uptake (33, 48). Experimentsincubating cells with EGF-liposome-DOX (figure 14 E) and trastuzumab-
0
0.2
0.4
0.6
0.8
1
1 4 8 24 48
1 2 5I - trastuzumab-l iposome-WSA
part
of t
otal
max
imal
upt
ake
incubation time (h)
0
0.2
0.4
0.6
0.8
1
1 2 4 8 24
1 2 5I-EGF-liposome-WSA (A-431)
part
of t
otal
max
imal
upt
ake
incubation time (h)
0
0.2
0.4
0.6
0.8
1
1 4 8 24 48
t ras tuzumab-3 H-l iposome-WSA
part
of t
otal
max
imal
upt
ake
incubation time (h)
A B
DC
FE
0
0.2
0.4
0.6
0.8
1
1 2 4 8 24
EGF-3H-liposome-WSA (A-431)
part
of t
otal
max
imal
upt
ake
incubation time (h)
0
0.2
0.4
0.6
0.8
1
1 2 4 8 24
EGF-3H-liposome-WSA (U343)
part
of t
otal
max
imal
upt
ake
incubation time (h)
0
0.2
0.4
0.6
0.8
1
1 2 4 8 24
1 2 5I-EGF-liposome-WSA (U343)
part
of t
otal
max
imal
upt
ake
incubation time (h)
38
liposomes-DOX were therefore performed. These experiments showednuclear staining of doxorubicin. It can be concluded that the EGFR andHER-2 pathways are working if nuclear delivery is desired, but the releasefrom the liposomes might be a problem. This problem of compound releasehas been addressed by several groups proposing different liposomalformulations to make the liposomes release their content more readily afterinternalization (149-151).
Figure 14. Fluorescent images of A) WSP. B) WSA. C) EGF-liposome-WSP. D)EGF-liposome-WSA. E) EGF-liposome-DOX. F). Trastuzumab-liposome-WSA.after 24 h incubation. WSA and WSP concentration 1µg/ml, and DOXconcentration 2µg/ml.
39
5.5 Optimization of boron uptakeTo test if the uptake of boron using EGF or trastuzumab-liposome-WSAcould be increased from the monolayer incubation, incubation incellsuspension in centrifuge tubes and in roller flasks were tested (figure15).
Incubating monolayer culture with EGF-liposome-WSA, the boroncontent in the cells reached 7 ppm after 24 h incubation. If instead the cellswere incubated as loose pellets in a centrifuge tube on gentle shake, theboron concentration reached over 40 ppm after 24 h incubation. Moreoverthe background was significantly lower for the pellet incubation,approximately 10%, while for the monolayer incubation it was almost50%.
Unfortunately, the cells did not seem to survive the rather harshtreatment of incubating the cells as loose pellets, most probably due to thepoor circulation of incubation medium. Therefore cells were incubated insuspension in small siliconized flasks kept on constant roll to ensuremaximum medium circulation. Using this method boron concentration of55 ppm was reached after only 4 h incubation. After 24 h almost 90 ppmwas reached, but the cells had then started forming spheroids. Thebackground was low, 20%, using this incubation method as well.
The increase in uptake using the different incubation methods isprobably due to an increase in conjugate concentration (in the incubationmedium). There might also be a more favorable geometry for uptake byincubating the cells in suspension with more accessible receptors.
The fact that the background is lower is not easy to explain, but it can benoted that the background is at approximately the same quantitative levelfor all three incubation methods. Thus, it is not the background that hasbeen lowered but the specific uptake that has increased.
For trastuzumab-liposome-WSA the same increase in uptake whenchanging from monolayer culture to incubation in roller-flasks was seen. Inmonolayer culture the uptake reached 12 ppm after 4 h incubation, with abackground of 75%. In roller flasks experiments more promising resultswere obtained with an uptake increase to 132 ppm and a background of25%.
40
Figure 15. Boron uptake in cells after different incubation methods. The conjugateEGF-liposome-WSA was added to cell dishes, cells in suspension forming a loosepellet, or cells in suspension in roller flasks. Trastuzumab-liposome-WSA wasadded to cell dishes or cells in suspension in roller flasks. To test the specificity ofthe uptake, samples where the cell receptors had been blocked by addition ofexcess EGF or trastuzumab were also used. Incubation time was 4 and 24 h.
5.6 CommentsIt was shown that all studied conjugates bound specifically to theirreceptors, were internalized and had good intracellular retention.Internalization is especially important for BNCT since the short range ofthe particles makes the compound more toxic if the decay takes placewithin the cell than on the cellular membrane (90).
The specificity, studied by the displacement assay, shows unspecificbinding ranging from 10 to 30%. Other groups have shown similar results.Gabizon et al. showed that addition of free folate in excess inhibited 86%of the binding of folate-PEG-liposome constructs (152). Lee et al. reportedunspecific binding of about 30% using folate liposomes and folic acidblock (34). Lopez de Menezes et al report 50% reduction in uptake afteraddition of free anti-CD19 antibody compared to the uptake of anti-CD19-liposomes (46). The unspecific uptake will have to be studied in an in vivosituation to determine the uptake in healthy tissue. In a binary therapy likeBNCT unspecific uptake is of less problem, since it is only the irradiatedareas that are exposed to the toxic effects.
The uptake of boron, reaching 90 ppm for EGF-liposome-WSA and 132ppm for trastuzumab-liposome-WSA, is at levels that are of clinical
Boron uptake (ppm) for different incubation methods
0
2 0
4 0
6 0
8 0
100
120
140
160
EGF-liposome-WSA, dishes
EGF-liposome-WSA, pellet,cells removedby scraping
EGF-liposome-WSA, pellet,
cells removedby trypzination
EGF-liposome-WSA, roller-
bottle
trastuzumab-liposome-WSA,
dishes
trastuzumab-liposome-WSA,
rol ler-bot t le
4h +block4h -block24h +block24h -block
41
interest if 10B-enriched compounds were used. Other groups report uptakein cell culture of up to 1500 ppm. However, these high levels wereassociated with very high (1300 ppm) unspecific uptake (37).
.
42
6. BNCT experiments on cultured gliomacells
To study the efficacy of the EGF-liposome-WSA conjugate a BNCTexperiment was conducted, using clonogenic survival of cultured U-343MGa Cl2:6 cells. After incubation with EGF-liposome-WSA the cells wereirradiated with neutrons at the Studsvik R2-0 reactor. It is important to notethat the WSA used in this study was not 10B enriched and therefore only20% of the boron in the cells was capable of neutron capture. If 10Benriched compounds had been used the efficacy would have been higher.
6.1 Experimental detailsGlioma cells were incubated for 4 hours as a cell suspension in roller-flaskswith either EGF-liposome-WSA, EGF-liposome-WSA with excess of freeEGF, or culture medium. After incubation the medium was removed andthe cells were thoroughly washed to remove any unbound conjugate. Thecells were resuspended in a small volume of culture medium and kept onice throughout the irradiation to hinder any repair mechanisms. The cellswere irradiated with 0, 1, 2 or 3�1012 thermal neutrons/cm2. After neutronirradiation the cells were counted, diluted and plated for clonogenicsurvival assay.
The clonogenic survival assay is a powerful tool to determine thetoxicity of a compound or a treatment. A number of cells are plated in apetri dish, where only the surviving cells will continue to divide. After anumber of doubling times (approx. 10) the surviving cells will have formedcolonies that, after staining, are visible for the naked eye. The count of thecolonies compared with the number of seeded cells gives a value of theclonogenic ability. Normalization to the clonogenic ability of untreatedcells gives the survival.
43
6.2 Dose calculationsThe physical dose contributions by the thermal neutron beam and thecontributions from the boron uptake were calculated using Monte Carlosimulations (more information can be obtained in paper V), (Table 7). Themajor dose contributors for the thermal neutron beam were the photon dosewith 0.9148 Gy/1012 thermal neutrons/cm2, the nitrogen dose with 0.232Gy/1012 nth/cm2 and the fast neutron dose with 0.082 Gy/1012 nth/cm2. Thedose from the boron was with the uptake found in the study 0.823 Gy/1012
nth/cm2. The boron uptake was 55 ppm although only 20%, 11 ppm, was10B. If 10B enriched compound had been used, the dose from the boronwould have been 4.11 Gy/1012 nth/cm2, more than four times higher thanthe dose from the neutron beam alone.
Table 7 Fluence rates and physical doses
Fluence rates Dose (Gy/1012 nth/cm2)Thermal neutrons (nth): 6.35�109 cm-2s-1
Fast neutrons:1.89�109 cm-2s-1
Fast neutron dose 0.082Nitrogen dose 0.232Photon dose 0.9148Boron dose (11 ppm) 0.823Boron dose (55 ppm) 4.11
6.3 Cell survival after neutron irradiationThe survival of the glioma cells after neutron irradiation as a function ofneutron fluence is shown in figure 16. It was shown that, if the cells hadbeen incubated with EGF-liposome-WSA, only half the neutron fluencewas needed to the reduce the survival to 10% compared to irradiation ofuntreated control cells. After incubation for 4 h with EGF-liposome-WSA,the boron levels in the cells reached 55 ppm. Cells with blocked EGF-receptors reached 11 ppm resulting in an unspecific binding of approx.20%. This low binding confirms receptor specificity. The difference seenin survival of the liposomes with and without blocked receptors alsoconfirm specificity and the difference reflects the amount boron taken upby the cells.
44
Figure 16. Clonogenic survival following thermal neutron irradiation of cellstreated with EGF-liposome-WSA, EGF-liposome-WSA with pre-blockedreceptors or untreated cells. The data sets were in all cases fitted to an exponentialmodel, S=exp(-αfF) where F is the neutron fluence and αf is a fitting parameter.
Considering the survival as a function of dose, the results look similar ascan be seen in figure 17. The survival was still lowest for the cells treatedwith WSA-liposomes. Comparing with gamma irradiation it could be seenthat for 10% survival 2.9 Gy was needed for targeted liposomes and 5.6 Gyfor 137Cs gamma. 3.6 Gy was needed for neutron irradiation only. Theslight difference between blocked and non-blocked cells should be due todifferences in boron localization. The boron uptake with blocked receptorsis probably due to both unspecific cell binding (as is shown in paper III)and liposome leakage. It can be assumed that the uptake via receptormediated endocytosis, as in the case for EGF-liposome-WSA, shoulddeliver the boron to a more favorable intracellular position.
0.01
0.1
1
0 0.5 1 1.5 2 2.5 3 3.5
neutronsEGF-liposome-WSAEGF-liposome-WSA+EGF
surv
ival
neutrons (1012/cm2)
45
Figure 17. Survival as a function of dose. The neutron irradiated data sets havebeen fitted to a purely exponential curve, S = exp(-αdD) and the gamma irradiatedare fitted to a linear quadratic curve fit, S =exp(-αD-βD2), where D is the dose inGy and α and β are fitting parameters.
6.4 CommentsIt is possible to reduce survival of cultured glioma cells 2 orders ofmagnitude by incubation with EGF-liposome-WSA and subsequentneutron irradiation. There was also a clear difference in survival if anexcess of EGF blocked the uptake.
Previous studies of cell survival after neutron activation of liposomedelivered boron also show decreased cell survival. Yanagie et al (125, 127)have performed in vitro studies of liposome-encapsulated 10B. They showthat the cell growth of pancreatic cancer cells and breast cancer cells couldbe lowered to approximately 30 % after neutron irradiation with 5�1012
neutrons/cm2. The cell growth was assayed by 3H-TdR incorporation. Wehave managed, at 3�1012 neutrons/cm2, to inactivate 99% of the cells.
With a 10B uptake of 11 ppm and a neutron flux of 3�1012 neutrons/cm2
1% of the cells still survive. This means that for a tumor with 105 cells1000 will survive and continue to grow. If 10B-enriched compounds couldbe used, the survival would be much lower, hopefully by several orders ofmagnitude.
0.01
0.1
1
0 2 4 6 8 10
137Cs gamma137 Cs gamma +EGF-liposome-WSAneutrons onlyneutrons+ EGF-liposome-WSA +EGFneutrons +EGF-liposome-WSA
surv
ival
Dose (Gy)
46
7. Biodistribution of EGF and EGF-liposomes
As shown earlier the EGFR is a suitable target for tumor therapy.Unfortunately, the biodistribution of EGF has proved that it is a not verysuccessful candidate for systemic injection because of its short circulationtime, mainly due to its rapid uptake in liver and kidneys (153). It has beenreported that after 2.5 min only 7% of 125I-EGF remain in the bloodstreamin rats (154). The hepatocytes in the liver are known to have a largenumber of EGF-receptors (142) that effectively binds and degrades EGF.
In mice the sinusoidal fenestrations lining the vessels in the liver areapproximately 100 nm (155). That means that a particle needs to be smallerthan 100 nm to leave the vessels and reach the hepatocytes. If the EGFwere to be coupled to a particle of this size or larger, maybe the binding toEGFRs of the hepatocytes might be circumvented. We wanted toinvestigate whether the attachment of a liposome to EGF would increasecirculation time of EGF and decrease the binding to the liver due to its size.The liposomes with similar size, as the epithelial fenestrations of the liver,should reduce the hepatocytic uptake of EGF, while some non-parenchymal uptake of the EGF-liposomes would still be present due to theliposome uptake by Kupffer cells.
7.1 Experimental proceduresMice were injected in the tail vein with 125I-EGF-liposome conjugate or125I-EGF and after various times the animals were sacrificed. Blood, urineand organs were collected according to a standard protocol with 20 organs.The radioactivity of the organs was measured using a gamma counter afterweight determinations.
To see if the EGFR-specific uptake of 125I-EGF and 125I-EGF-liposomecould be blocked, an excess of EGF was injected into the mice 15 minprior to 125I-EGF or 125I-EGF-liposome injection. The mice were sacrificed15 min after the second injection. Blood and organs were collectedaccording to standard protocol, weighed, and thereafter the radioactivitywas measured.
47
7.2 Circulation timeAs is shown in figure 18 the circulation time of 125I-EGF increased if 100nm PEG-stabilized liposome were attached. 15 minutes after injection of125I-EGF 3.5% injected dose per g (ID/g) was still in the blood stream. Onehour later 3% ID/g remained. If a liposome was attached, the blood levelchanged to 25% ID/g after 15 min and 14% ID/g after 1h. If the EGF-receptors were blocked by injection of an excess of EGF, the blood level of125I-EGF after 15 min was higher than for non-blocked mice, probably asan effect of blocked EGF-receptors in the liver. No difference in bloodlevel could be seen for 125I-EGF-liposomes after blocking. The circulationhalftime of 125I-EGF-liposomes is longer than 125I-EGF but lower thanresults obtained in other studies with liposomes which often have halftimesof several hours (12, 16). This can be explained by the fact that in thisstudy the targeting agent is followed, not the liposome. If the liposomeswere radiolabeled we could expect longer halftimes.
Figure 18. Levels of 125I-EGF (open bars) and 125I-EGF-liposomes (filled bars) inblood expressed as % ID/g. Error bars represent maximal errors, 4 animals pertime point.
7.3 Uptake in liver and kidneysThe uptake of 125I-EGF in liver decreased significantly when 100 nm PEG-stabilized liposomes were attached to EGF as is shown in figure 19 A. For125I-EGF the uptake was 51% ID/g 15 min after injection, and 1h afterinjection it had decreased to 3% ID/g, indicating rapid degradation. For125I-EGF-liposomes the uptake 15 min after injection was 9% ID/g, and theradioactivity was not excreted as rapidly as for 125I-EGF. When thereceptors were blocked the uptake of 125I-EGF altered dramatically. From
0
10
20
30
40
0.25h 0.25h +EGF 1 h 2h 4h 8h 24h
Blood
%ID
/g
48
the previously shown uptake of 51% ID/g after 15 min, the uptakedecreased to 2% ID/g indicating blockage of EGFR specific uptake tohepatocytes in liver. This dramatic difference was not seen when 125I-EGF-liposomes were used. The uptake was somewhat lowered if the receptorswere blocked, indicating low EGFR specific binding to liver for 125I-EGF-liposomes. However, we can not exclude that there is some hepatocytespecific uptake, even though the slight difference after blockage indicatesthat the liver uptake of 125I-EGF-liposomes is by non-parenchymal cells toa high extent.
Figure 19. Levels of 125I-EGF (open bars) and 125I-EGF-liposomes (filled bars) inliver and kidney expressed as % ID/g. Error bars represent maximal errors, 4animals per time point.
In the kidney there were also vast differences in the uptake patterns of EGFand EGF-liposomes (figure 19 B). The uptake of 125I-EGF was at 51% ID/g15 min after injection and after 1h this had decreased to 8% ID/g. If thereceptors were blocked, the uptake doubled due to the higher blood levels
0
10
20
30
40
50
60
0.25h 0.25h +EGF 1 h 2h 4h 8h 24h
L i v e r
%ID
/g
A
0
20
40
60
80
100
120
0.25h 0.25h +EGF 1 h 2h 4h 8h 24h
Kidney
%ID
/g
B
49
as a result of blocked organs, for example the liver. It was clear that EGFwas excreted through the kidneys. For 125I-EGF-liposomes the kidneyuptake was as low as 4% ID/g 15 min after injection and slowlydecreasing. No difference could be seen with blocked EGF receptors.These results indicate that EGF-liposomes are not excreted via the kidneys.Probably the size of the liposome disables it from renal excretion.
7.4 Uptake in other organsThere was an uptake in submaxillary gland of both 125I-EGF and125I-EGFliposomes (table 8, 9). The accumulation of 125I-EGF was more rapid, butthe same levels were reached for both the peptide and the conjugate. Thesubmaxillary gland, being the main site of EGF production in mice,expresses EGF receptors (156), and it was shown that the uptake of 125I-EGF could be blocked. Since the blocking experiment only was performedwith 15 min incubation time, no difference after blocking could be seen for125I-EGF-liposomes. The uptake of 125I-EGF-liposomes in submaxillarygland did not reach its highest value until after 4h.
In the spleen there was a higher uptake of the liposome conjugate thanof 125I-EGF (table 8, 9). This is probably due to the fact that liposomes arecleared by MPS uptake and the spleen has an important role in the MPS.There was no difference in the uptake if EGF-receptors were blocked,indicating that the splenic uptake is not mediated by EGFR.
The uptake of 125I in thyroid reached approximately the same levels forboth 125I-EGF and 125I-EGF-liposomes (table 8, 9). The uptake increasedwith time corresponding to the release of 125I and 1 2 5I-tyrosine fromdegraded 125I-EGF and 125I-EGF-liposomes.
Table 8. Uptake of 125I-EGF in selected organs (%ID/g)
Time Submax. Spleen Thyroid*
15 min 16 +/- 6 3.9 +/- 1.2 0.3 +/- 0.515 min + EGF 5.2 +/- 1.5 2.1 +/- 0.3 0.19 +/- 0.051 h 24 +/- 19 1.8 +/- 0.6 1.1 +/- 0.42 h 13 +/- 5 1.8 +/- 0.1 2.2 +/-1.34h 14 +/- 6 0.88 +/- 0.10 3.9 +/- 1.58 h 8.1 +/- 3.3 0.58 +/- 0.14 4.5 +/- 1.824 h 0.42 +/- 0.53 0.06 +/- 0.04 7.8 +/- 4.7* The values are expressed as %ID
50
Table 9. Uptake of 125I-EGF-liposomes in selected organs (%ID/g)
Time Submax. Spleen Thyroid*
15 min 0.58 +/- 0.42 9.8 +/- 7.1 0.03 +/- 0.0215 min + EGF 0.63 +/- 0.18 8.1 +/- 3.2 0.03 +/- 0.011 h 4.7 +/- 0.9 11 +/- 5 0.24 +/- 0.092 h 7.3 +/- 1.4 6.5 +/- 4.6 0.59 +/- 0.054h 16 +/- 7 6.3 +/- 4.9 1.7 +/- 0.98 h 5.9 +/- 3.6 2.3 +/- 1.0 4.8 +/- 3.224 h 0.46 +/- 0.21 1.2 +/- 0.6 5.6 +/- 1.4* The values are expressed as %ID
7.5 CommentsWe were able to alter the biodistribution of 125I-EGF by conjugation to 100nm PEG-stabilized liposomes. The most important differences inbiodistribution are the reduced uptake in the liver and the kidneys whenEGF is conjugated to liposomes, as well as the elevated blood levels overtime. The circulation half-time of 125I-EGF was prolonged when attachingit to liposomes, since the blood-levels were much higher for 125I-EGF-liposomes. To decrease liver and kidney uptake and to increase the bloodconcentration of EGF-conjugate is of greatest importance for the use ofEGF as a targeting agent in nuclide therapy or other targeting modalitiesutilizing EGF against cancer. The approach presented here will probablyreduce liver and kidney toxicity, while giving time for the EGF conjugateto reach the tumor cells.
51
8. Summary and future work
8.1 SummaryThis thesis has mainly dealt with the development of targeted liposomeconjugates and their behavior in cell culture and in mice.
Liposome conjugates with high stability targeting EGFR and HER-2have been constructed. They showed receptor specificity and internalizeinto the tumor cells after binding. The cellular retention of the liposomeconjugates after 24 h incubation was also found to be good. Thus, thedescribed liposome systems seems of interest of tumor therapy. There ispotential to load the liposomes with various toxic agents such as toxins,radionuclides and nuclides for neutron activation.
The compounds WSA and WSP have previously been loaded intoliposomes at high levels with low leakage (129). We have shown that theuse of EGF and trastuzumab conjugated liposomes loaded with WSAresults in high boron levels after incubation in a cell suspension culture.The boron levels reached in vitro are high enough to be of clinical interest.Further, by applying enriched 10B the situation will be even better. WithEGF-liposome-WSA, the high boron levels were used to kill culturedglioma cells. After neutron irradiation survival was studied, using theclonogenic survival assay. Significant differences in survival were seen forcells incubated with conjugate and cells that were only neutron irradiated.
The in vivo studies, comparing the uptake of 125I-EGF and 125I-EGF-liposomes in selected organs in mice, showed that the uptake in liver andkidneys was significantly lower for 125I-EGF-liposomes and the circulationtime in blood was longer. The fact that the 125I-EGF-liposome conjugatesdoes not assemble in the liver to the same high extent as free 125I-EGFmakes this conjugate promising for systemic injection.
52
8.2 Future workThe most interesting areas to continue the work with stabilized liposomeligand conjugates, in my point of view, are the following:
1 . Further study of EGF-liposomes in vivo, using animals withxenografted tumors to study tumor specific delivery, will beperformed to conclude whether EGF-liposomes are suitable forsystemic drug delivery.
2 . The high boron level reached in cell culture for trastuzumab-liposome-WSA conjugates, 132 ppm, makes this conjugate apromising candidate for experimental BNCT. It would also be veryinteresting to perform BNCT on tumor bearing animals, using bothEGF-liposome and trastuzumab-liposome conjugates. If 10B enrichedcompounds could be used the therapeutical results would bedramatically improved when using WSA loaded liposomes.
3 . Further development of the liposome conjugates with new DNA-binding compounds based on doxorubicin is in progress. Doxorubicin,which has been shown to be specifically delivered to the DNA afterliposome targeting, is of specific interest for labeling with the Augerelectron emitting nuclide 125I. Auger electron emitting 125I gives highcell damage ( about one double strand break per decay), if the nuclideis situated within the DNA (157).
4 . It would also be exciting to study whether the intact antibody oftrastuzumab-liposomes has a negative impact on the tumor specificdelivery. Previous studies by Park et al have shown that Fab- or ScFv-conjugated liposomes targeting HER-2 have at least as goodcharacteristics as the antibody conjugated liposomes (158) in vitro andin vivo.
5. Targeted liposomes loaded with radiometals, for example 111In, wouldalso be of interest to study. Radiometals are known to have goodintracellular retention, and being able to deliver large amounts byusing loaded liposomes would be of great value for imaging andpossibly also therapy.
53
Acknowledgements
The work in this thesis have been conducted at the division of BiomedicalRadiation Sciences (BMS), Uppsala University, with financial supportmainly from Cancerfonden. I would like to thank everyone who has helpedme along the way and a send a special thank you to:
Dr. Lars Gedda, my supervisor. For being the perfect supervisor, you havealways taken part in my research with great interest and you have alwaysfound the time to talk to me no matter how trivial the issue.
Prof. Hans Lundqvist and Ass. Prof. Bo Sternerlöw, Co-supervisors, foralways being supportive and helpful.
Prof. Jörgen Carlsson, Head of the BMS department and examiner. Forcreating a wonderful warm and supportive atmosphere. It has been a purepleasure working at BMS. Thank you also for always taking such greatinterest in my work.
Quichun Wei, co-author, thank you for all help with the trastuzumab-workand for the animal experiments.
Marika Nestor, co-author, for working patiently and independently with allthose cell experiments and for putting up with me as a supervisor.
Prof. Katarina Edwards, Nill Bergström, Dr. Markus Johnsson and MariaArfvidsson, Co authors. For all help with liposomes, answering all myquestions and swiftly supplying me with new liposome batches on shortnotice. Thank you for a lot of fun as well.
Prof. Stefan Sjöberg and his Ph.D. students, past and present, for supplyingWSA and WSP.
Ass. Prof. Jacek Capala and Charlotta Persson. Thank you for all help withthe experiments at Studsvik.
54
Dr Peter Frisk, Ulla Johansson, Ass. Prof Ulf Lindh and Dr JeanPettersson. For helping me with a lot of ICP boron analyzes.
Dr Stefan Gunnarsson, for introducing me to the world of confocalmicroscopy.
Dr Anna Orlova and Ass. Prof. Vladimir Tolmachev. For instructions andhelp during my brief period as "labeling chemist" working with A3.
Veronika Asplund Eriksson, som håller ordning på allt och alla och ocksåhar varit en trevlig rumskamrat.
Maria Östh-Eklind för administrativ hjälp.
Åsa Liljegren Sundberg, för en massa kul hela tiden på BMS.
Gamla och nya doktorander, för en massa roliga saker som gör livet kul.Man får ALLTID minst ett skratt om dagen på BMS.
Mina föräldrar och min syster, tack för att ni alltid trott på mig, stöttat migoch pushat mig att göra mitt bästa.
Kullbergfamiljen, för att ni tagit mig till er och alltid orkat uppbådaintresse för hur det går med min "infekterade celler" m.m.
Mathias, min käre make… För att du är så bra och för att du får mig attkänna mig som den roligaste, vackraste och smartaste kvinnan på jorden.
55
References
1. Goldenberg, D. M. Targeted therapy of cancer with radiolabeled antibodies, JNucl Med. 43: 693-713. 2002.2. Postema, E. J., Boerman, O. C., Oyen, W. J., Raemaekers, J. M., and Corstens,F. H. Radioimmunotherapy of B-cell non-Hodgkin's lymphoma, Eur J Nucl Med.28: 1725-35. 2001.3. Sudimack, J. and Lee, R. J. Targeted drug delivery via the folate receptor, AdvDrug Deliv Rev. 41: 147-62. 2000.4. Carlsson, J., Forssell Aronsson, E., Hietala, S. O., Stigbrand, T., and Tennvall, J.Tumour therapy with radionuclides: assessment of progress and problems,Radiother Oncol. 66: 107-17. 2003.5. Masquelier, M., Vitols, S., Palsson, M., Mars, U., Larsson, B. S., and Peterson,C. O. Low density lipoprotein as a carrier of cytostatics in cancer chemotherapy:study of stability of drug-carrier complexes in blood, J Drug Target. 8: 155-64.2000.6. Masquelier, M., Tirzitis, G., Peterson, C. O., Palsson, M., Amolins, A.,Plotniece, M., Plotniece, A., Makarova, N., and Vitols, S. G. Plasma stability andcytotoxicity of lipophilic daunorubicin derivatives incorporated into low densitylipoproteins, Eur J Med Chem. 35: 429-38. 2000.7. Gregoriadis, G., Wills, E. J., Swain, C. P., and Tavill, A. S. Drug-carrierpotential of liposomes in cancer chemotherapy, Lancet. 1: 1313-6. 1974.8. Fidler, I. J., Raz, A., Fogler, W. E., Kirsh, R., Bugelski, P., and Poste, G. Designof liposomes to improve delivery of macrophage-augmenting agents to alveolarmacrophages, Cancer Res. 40: 4460-6. 1980.9. Gabizon, A. and Papahadjopoulos, D. Liposome formulations with prolongedcirculation time in blood and enhanced uptake by tumors, Proc Natl Acad Sci U SA. 85: 6949-6953. 1988.10. Klibanov, A. L., Maruyama, K., Torchilin, V. P., and Huang, L. Amphipathicpolyethyleneglycols effectively prolong the circulation time of liposomes, FEBSLett. 268: 235-7. 1990.11. Allen, T. M., Hansen, C., Martin, F., Redemann, C., and Yau-Young, A.Liposomes containing synthetic lipid derivatives of poly(ethylene glycol) showprolonged circulation half-lives in vivo, Biochim Biophys Acta. 1066: 29-36.1991.12. Bakker-Woudenberg, I. A., Lokerse, A. F., ten Kate, M. T., Mouton, J. W.,Woodle, M. C., and Storm, G. Liposomes with prolonged blood circulation andselective localization in Klebsiella pneumoniae-infected lung tissue, J Infect Dis.168: 164-71. 1993.13. Boerman, O. C., Dams, E. T., Oyen, W. J., Corstens, F. H., and Storm, G.Radiopharmaceuticals for scintigraphic imaging of infection and inflammation,Inflamm Res. 50: 55-64. 2001.
56
14. Papahadjopoulos, D., Allen, T. M., Gabizon, A., Mayhew, E., Matthay, K.,Huang, S. K., Lee, K. D., Woodle, M. C., Lasic, D. D., Redemann, C., and et al.Sterically stabilized liposomes: improvements in pharmacokinetics and antitumortherapeutic efficacy, Proc Natl Acad Sci U S A. 88: 11460-4. 1991.15. Forssen, E. A., Coulter, D. M., and Proffitt, R. T. Selective in vivo localizationof daunorubicin small unilamellar vesicles in solid tumors, Cancer Res. 52: 3255-61. 1992.16. Gabizon, A. A. Selective tumor localization and improved therapeutic index ofanthracyclines encapsulated in long-circulating liposomes, Cancer Res. 52: 891-6.1992.17. Batist, G., Ramakrishnan, G., Rao, C. S., Chandrasekharan, A., Gutheil, J.,Guthrie, T., Shah, P., Khojasteh, A., Nair, M. K., Hoelzer, K., Tkaczuk, K., Park,Y. C., and Lee, L. W. Reduced cardiotoxicity and preserved antitumor efficacy ofliposome-encapsulated doxorubicin and cyclophosphamide compared withconventional doxorubicin and cyclophosphamide in a randomized, multicenter trialof metastatic breast cancer, J Clin Oncol. 19: 1444-54. 2001.18. Williams, G., Cortazar, P., and Pazdur, R. Developing drugs to decrease thetoxicity of chemotherapy, J Clin Oncol. 19: 3439-41. 2001.19. Gill, P. S., Wernz, J., Scadden, D. T., Cohen, P., Mukwaya, G. M., von Roenn,J. H., Jacobs, M., Kempin, S., Silverberg, I., Gonzales, G., Rarick, M. U., Myers,A. M., Shepherd, F., Sawka, C., Pike, M. C., and Ross, M. E. Randomized phaseIII trial of liposomal daunorubicin versus doxorubicin, bleomycin, and vincristinein AIDS-related Kaposi's sarcoma, J Clin Oncol. 14: 2353-64. 1996.20. Northfelt, D. W., Dezube, B. J., Thommes, J. A., Miller, B. J., Fischl, M. A.,Friedman-Kien, A., Kaplan, L. D., Du Mond, C., Mamelok, R. D., and Henry, D.H. Pegylated-liposomal doxorubicin versus doxorubicin, bleomycin, andvincristine in the treatment of AIDS-related Kaposi's sarcoma: results of arandomized phase III clinical trial, J Clin Oncol. 16: 2445-51. 1998.21. Northfelt, D. W., Martin, F. J., Working, P., Volberding, P. A., Russell, J.,Newman, M., Amantea, M. A., and Kaplan, L. D. Doxorubicin encapsulated inliposomes containing surface-bound polyethylene glycol: pharmacokinetics, tumorlocalization, and safety in patients with AIDS-related Kaposi's sarcoma, J ClinPharmacol. 36: 55-63. 1996.22. Johnston, S. R. and Gore, M. E. Caelyx: phase II studies in ovarian cancer, EurJ Cancer. 37: S8-14. 2001.23. Muggia, F. and Hamilton, A. Phase III data on Caelyx in ovarian cancer, Eur JCancer. 37: S15-8. 2001.24. Ranson, M. R., Carmichael, J., O'Byrne, K., Stewart, S., Smith, D., andHowell, A. Treatment of advanced breast cancer with sterically stabilizedliposomal doxorubicin: results of a multicenter phase II trial, J Clin Oncol. 15:3185-91. 1997.25. Symon, Z., Peyser, A., Tzemach, D., Lyass, O., Sucher, E., Shezen, E., andGabizon, A. Selective delivery of doxorubicin to patients with breast carcinomametastases by stealth liposomes, Cancer. 86: 72-8. 1999.26. Adler-Moore, J. and Proffitt, R. T. AmBisome: liposomal formulation,structure, mechanism of action and pre-clinical experience, J AntimicrobChemother. 49: 21-30. 2002.27. Chopra, R. AmBisome in the treatment of fungal infections: the UKexperience, J Antimicrob Chemother. 49: 43-7. 2002.
57
28. Blume, G., Cevc, G., Crommelin, M. D. J. A., Bakker-Woudenberg, I. A.,Kluft, C., and Storm, G. Specific targeting with poly(ethylene glycol)-modifiedliposomes: coupling of homing devices to the ends of the polymeric chainscombines effective target binding with long circulation times, Biochim BiophysActa. 1149: 180-184. 1993.29. Allen, T. M., Brandeis, E., Hansen, C. B., Kao, G. Y., and Zalipsky, S. A newstrategy for attachment of antibodies to sterically stabilized liposomes resulting inefficient targeting to cancer cells [published erratum appears in Biochim BiophysActa 1995 Dec 13;1240(2):285], Biochim Biophys Acta. 1237: 99-108. 1995.30. Hansen, C. B., Kao, G. Y., Moase, E. H., Zalipsky, S., and Allen, T. M.Attachment of antibodies to sterically stabilized liposomes: evaluation, comparisonand optimization of coupling procedures, Biochim Biophys Acta. 1239: 133-44.1995.31. Zalipsky, S., Mulla, N., Harding, J., Gittelman, J., Guo, L., and DeFrees, S.Poly(ethylene glycol)-grafted liposomes with oligopeptide or oligosaccharideligands appended to the temini of the polymer chains, Bioconjug Chem. 8: 111-118. 1997.32. Torchilin, V. P., Levchenko, T. S., Lukyanov, A. N., Khaw, B. A., Klibanov,A. L., Rammohan, R., Samokhin, G. P., and Whiteman, K. R. p-Nitrophenylcarbonyl-PEG-PE-liposomes: fast and simple attachment of specificligands, including monoclonal antibodies, to distal ends of PEG chains via p-nitrophenylcarbonyl groups, Biochim Biophys Acta. 1511: 397-411. 2001.33. Goren, D., Horowitz, A. T., Tzemach, D., Tarshish, M., Zalipsky, S., andGabizon, A. Nuclear delivery of doxorubicin via folate-targeted liposomes withbypass of multidrug-resistance efflux pump, Clin Cancer Res. 6: 1949-57 2000.34. Lee, R. J. and Low, P. S. Folate-mediated tumor cell targeting of liposome-entrapped doxorubicin in vitro, Biochim Biophys Acta. 1233: 134-44. 1995.35. Wang, S., Lee, R. J., Cauchon, G., Gorenstein, D. G., and Low, P. S. Deliveryof antisense oligodeoxyribonucleotides against the human epidermal growth factorreceptor into cultured KB cells with liposomes conjugated to folate viapolyethylene glycol, Proceedings of the National Academy of Sciences of theUnited States of America. 92: 3318-3322. 1995.36. Qualls, M. M. and Thompson, D. H. Chloroaluminum phthalocyaninetetrasulfonate delivered via acid-labile diplasmenylcholine-folate liposomes:intracellular localization and synergistic phototoxicity, Int J Cancer. 93: 384-92.2001.37. Pan, X. Q., Wang, H., Shukla, S., Sekido, M., Adams, D. M., Tjarks, W.,Barth, R. F., and Lee, R. J. Boron-Containing Folate Receptor-TargetedLiposomes as Potential Delivery Agents for Neutron Capture Therapy, BioconjugChem. 13: 435-442. 2002.38. Pan, X. Q., Wang, H., and Lee, R. J. Boron delivery to a murine lungcarcinoma using folate receptor-targeted liposomes, Anticancer Res. 22: 1629-33.2002.39. Sudimack, J. J., Adams, D. M., Rotaru, J. H., Shukla, S., Yan, J., Sekido, M.,Barth, R. F., Tjarks, W., and Lee, R. J. Folate receptor-mediated liposomaldelivery of a lipophilic boron agent to tumor cells in citro for neutron capturetherapy, Pharm Res. 19: 1502-1508. 2002.40. Suzuki, S., Uno, S., Fukuda, Y., Aoki, Y., Masuko, T., and Hashimoto, Y.Cytotoxicity of anti-c-erbB-2 immunoliposomes containing doxorubicin on humancancer cells, Br J Cancer. 72: 663-8. 1995.
58
41. Goren, D., Horowitz, A. T., Zalipsky, S., Woodle, M. C., Yarden, Y., andGabizon, A. Targeting of stealth liposomes to erbB-2 (Her/2) receptor: in vitro andin vivo studies, Br J Cancer. 74: 1749-56. 1996.42. Kirpotin, D., Park, J. W., Hong, K., Zalipsky, S., Li, W. L., Carter, P., Benz, C.C., and Papahadjopoulos, D. Sterically stabilized anti-HER2 immunoliposomes:design and targeting to human breast cancer cells in vitro, Biochemistry. 36: 66-75. 1997.43. Park, J. W., Hong, K., Carter, P., Asgari, H., Guo, L. Y., Keller, G. A., Wirth,C., Shalaby, R., Kotts, C., Wood, W. I., and et al. Development of anti-p185HER2immunoliposomes for cancer therapy, Proc Natl Acad Sci U S A. 92: 1327-31.1995.44. Park, J. W., Hong, K., Kirpotin, D. B., Meyer, O., Papahadjopoulos, D., andBenz, C. C. Anti-HER2 immunoliposomes for targeted therapy of human tumors,Cancer Lett. 118: 153-60. 1997.45. Park, J. W., Hong, K., Kirpotin, D. B., Colbern, G., Shalaby, R., Baselga, J.,Shao, Y., Nielsen, U. B., Marks, J. D., Moore, D., Papahadjopoulos, D., and Benz,C. C. Anti-HER2 Immunoliposomes: Enhanced Efficacy Attributable to TargetedDelivery, Clin Cancer Res. 8: 1172-81. 2002.46. Lopes de Menezes, D. E., Pilarski, L. M., and Allen, T. M. In vitro and in vivotargeting of immunoliposomal doxorubicin to human B-cell lymphoma, CancerRes. 58: 3320-30. 1998.47. Lopes de Menezes, D. E., Kirchmeier, M. J., Gagne, J.-F., Pilarski, L. M., andAllen, T. M. Cellular trafficking and cytotoxicity of anti- CD19-targeted liposomaldoxorubicin in B lymphoma cells, Journal of liposome research. 9: 199-228. 1999.48. Lopes de Menezes, D. E., Pilarski, L. M., Belch, A. R., and Allen, T. M.Selective targeting of immunoliposomal doxorubicin against human multiplemyeloma in vitro and ex vivo, Biochim Biophys Acta. 1466: 205-20. 2000.49. Sarti, P., Ginobbi, P., D'Agostino, I., Arancia, G., Lendaro, E., Molinari, A.,Ippoliti, R., and Citro, G. Liposomal targeting of leukaemia HL60 cells induced bytransferrin-receptor endocytosis, Biotechnol Appl Biochem. 24: 269-76. 1996.50. Iinuma, H., Maruyama, K., Okinaga, K., Sasaki, K., Sekine, T., Ishida, O.,Ogiwara, N., Johkura, K., and Yonemura, Y. Intracellular targeting therapy ofcisplatin-encapsulated transferrin-polyethylene glycol liposome on peritonealdissemination of gastric cancer, Int J Cancer. 99: 130-7. 2002.51. Earp, H. S., Dawson, T. L., Li, X., and Yu, H. Heterodimerization andfunctional interaction between EGF receptor family members: a new signalingparadigm with implications for breast cancer research, Breast Cancer Res Treat.35: 115-32. 1995.52. Olayioye, M. A., Neve, R. M., Lane, H. A., and Hynes, N. E. The ErbBsignaling network: receptor heterodimerization in development and cancer, EmboJ. 19: 3159-67. 2000.53. Carpenter, G. and Cohen, S. Epidermal growth factor, J Biol Chem. 265: 7709-12. 1990.54. Riese, D. J., 2nd and Stern, D. F. Specificity within the EGF family/ErbBreceptor family signaling network, Bioessays. 20: 41-8. 1998.55. Graus-Porta, D., Beerli, R. R., Daly, J. M., and Hynes, N. E. ErbB-2, thepreferred heterodimerization partner of all ErbB receptors, is a mediator of lateralsignaling, Embo J. 16: 1647-55. 1997.
59
56. Cohen, S. Isolation of a Mouse Submaxillary Gland Protein AcceleratingIncisor Eruption and Eyelid Opening in the New-born Animal, J Biol Chem. 237:1555-1562. 1962.57. Gregory, H. Isolation and structure of urogastrone and its relationship toepidermal growth factor, Nature. 257: 325-7. 1975.58. Beguinot, L., Lyall, R. M., Willingham, M. C., and Pastan, I. Down-regulationof the epidermal growth factor receptor in KB cells is due to receptorinternalization and subsequent degradation in lysosomes, Proc Natl Acad Sci U SA. 81: 2384-8. 1984.59. Wiley, H. S., Herbst, J. J., Walsh, B. J., Lauffenburger, D. A., Rosenfeld, M.G., and Gill, G. N. The role of tyrosine kinase activity in endocytosis,compartmentation, and down-regulation of the epidermal growth factor receptor, JBiol Chem. 266: 11083-94. 1991.60. Carbone, R., Fre, S., Iannolo, G., Belleudi, F., Mancini, P., Pelicci, P. G.,Torrisi, M. R., and Di, F. P. P. eps15 and eps15R are essential components of theendocytic pathway, Cancer Research. 57: 5498-5504. 1997.61. Baulida, J., Kraus, M. H., Alimandi, M., Di, F. P. P., and Carpenter, G. AllErbB receptors other than the epidermal growth factor receptor are endocytosisimpaired, Journal of Biological Chemistry. 271: 5251-5257. 1996.62. Collins, V. P. Amplified genes in human gliomas, Semin Cancer Biol. 4: 27-32. 1993.63. Chaffanet, M., Chauvin, C., Laine, M., Berger, F., Chedin, M., Rost, N.,Nissou, M. F., and Benabid, A. L. EGF receptor amplification and expression inhuman brain tumours, Eur J Cancer. 28: 11-7. 1992.64. Frederick, L., Wang, X. Y., Eley, G., and James, C. D. Diversity and frequencyof epidermal growth factor receptor mutations in human glioblastomas, CancerRes. 60: 1383-7. 2000.65. Neal, D. E. and Mellon, K. Epidermal growth factor receptor and bladdercancer: a review, Urol Int. 48: 365-71. 1992.66. Reilly, R. M., Kiarash, R., Cameron, R. G., Porlier, N., Sandhu, J., Hill, R. P.,Vallis, K., Hendler, A., and Gariepy, J. 111In-labeled EGF is selectively radiotoxicto human breast cancer cells overexpressing EGFR, J Nucl Med. 41: 429-38. 2000.67. Rusch, V., Mendelsohn, J., and Dmitrovsky, E. The epidermal growth factorreceptor and its ligands as therapeutic targets in human tumors, Cytokine GrowthFactor Rev. 7: 133-41. 1996.68. Wilson, J. A., Rogers, M. J., Hawkins, R. A., Gilmour, H. M., and Maran, A.G. Epidermal growth factor receptors and oestrogen receptors in the head andneck, Clin Otolaryngol. 18: 66-8. 1993.69. Ekstrand, A. J., James, C. D., Cavenee, W. K., Seliger, B., Pettersson, R. F.,and Collins, V. P. Genes for epidermal growth factor receptor, transforminggrowth factor alpha, and epidermal growth factor and their expression in humangliomas in vivo, Cancer Res. 51: 2164-72. 1991.70. Eccles, S. A. The role of c-erbB-2/HER2/neu in breast cancer progression andmetastasis, J Mammary Gland Biol Neoplasia. 6: 393-406. 2001.71. Bookman, M. A., Darcy, K. M., Clarke-Pearson, D., Boothby, R. A., andHorowitz, I. R. Evaluation of monoclonal humanized anti-HER2 antibody,trastuzumab, in patients with recurrent or refractory ovarian or primary peritonealcarcinoma with overexpression of HER2: a phase II trial of the GynecologicOncology Group, J Clin Oncol. 21: 283-90. 2003.
60
72. Wang, Z., Zhang, L., Yeung, T. K., and Chen, X. Endocytosis deficiency ofepidermal growth factor (EGF) receptor-ErbB2 heterodimers in response to EGFstimulation, Mol Biol Cell. 10: 1621-36. 1999.73. Vieira, A. V., Lamaze, C., and Schmid, S. L. Control of EGF receptorsignaling by clathrin-mediated endocytosis, Science. 274: 2086-2089. 1996.74. Modjtahedi, H. and Dean, C. The receptor for EGF ant its ligands: Expression,prognostic value and target for therapy in cancer (Review), International Journal ofOncology. 4: 277-296 1994.75. Fan, Z. and Mendelsohn, J. Therapeutic application of anti-growth factorreceptor antibodies, Curr Opin Oncol. 10: 67-73. 1998.76. Mendelsohn, J. and Fan, Z. Epidermal growth factor receptor family andchemosensitization, J Natl Cancer Inst. 89: 341-3. 1997.77. Deshpande, D., Toledo, V. D., Thakkar, D., Liang, W., and Rojanasakul, Y.Enhanced cellular uptake of oligonucleotides by EGF receptor-mediatedendocytosis in A549 cells, Pharmaceutical Research. 13: 57-61. 1996.78. Carlsson, J., Blomquist, E., Gedda, L., Liljegren, Å., Malmström, P.-U.,Sjöström, A., Sundin, A., Westlin, J.-E., Zhao, Q., Tolmachev, V., and Lundqvist,H. Conjugate chemistry and cellular processing of EGF-Dextran, Acta Oncol. 38:313-321. 1999.79. Olsson, P., Gedda, L., Goike, H., Liu, L., Collins, V. P., Ponten, J., andCarlsson, J. Uptake of a boronated epidermal growth factor-dextran conjugate inCHO xenografts with and without human EGF-receptor expression, AnticancerDrug Des. 13: 279-89. 1998.80. Gedda, L., Olsson, P., Ponten, J., and Carlsson, J. Development and in vitrostudies of epidermal growth factor-dextran conjugates for boron neutron capturetherapy, Bioconjug Chem. 7: 584-91. 1996.81. Bue, P., Holmberg, A. R., Marquez, M., Westlin, J. E., Nilsson, S., andMalmstrom, P. U. Intravesical administration of EGF- dextran conjugates inpatients with superficial bladder cancer, Eur Urol. 38: 584-9. 2000.82. Liljegren Sundberg, Å., Orlova, A., Bruskin, A., Gedda, L., Carlsson, J.,Blomquist, E., Lundqvist, H., and Tolmachev, V. [111In] Bz-DTPA-hEGF:preparation and in vitro characterization of a potential anti-glioblastoma targetingagent, Cancer Biotherapy and Radiopharmaceuticals. In Press: 2003.83. Wakeling, A. E., Guy, S. P., Woodburn, J. R., Ashton, S. E., Curry, B. J.,Barker, A. J., and Gibson, K. H. ZD1839 (Iressa): an orally active inhibitor ofepidermal growth factor signaling with potential for cancer therapy, Cancer Res.62: 5749-54. 2002.84. Cobleigh, M. A., Vogel, C. L., Tripathy, D., Robert, N. J., Scholl, S.,Fehrenbacher, L., Wolter, J. M., Paton, V., Shak, S., Lieberman, G., and Slamon,D. J. Multinational study of the efficacy and safety of humanized anti-HER2monoclonal antibody in women who have HER2-overexpressing metastatic breastcancer that has progressed after chemotherapy for metastatic disease, J Clin Oncol.17: 2639-48. 1999.85. Yakes, F. M., Chinratanalab, W., Ritter, C. A., King, W., Seelig, S., andArteaga, C. L. Herceptin-induced inhibition of phosphatidylinositol-3 kinase andAkt Is required for antibody-mediated effects on p27, cyclin D1, and antitumoraction, Cancer Res. 62: 4132-41. 2002.86. Mayfield, S., Vaughn, J. P., and Kute, T. E. DNA strand breaks and cell cycleperturbation in herceptin treated breast cancer cell lines, Breast Cancer Res Treat.70: 123-9. 2001.
61
87. Moulder, S. L., Yakes, F. M., Muthuswamy, S. K., Bianco, R., Simpson, J. F.,and Arteaga, C. L. Epidermal growth factor receptor (HER1) tyrosine kinaseinhibitor ZD1839 (Iressa) inhibits HER2/neu (erbB2)-overexpressing breast cancercells in vitro and in vivo, Cancer Res. 61: 8887-95. 2001.88. Slamon, D. J., Leyland-Jones, B., Shak, S., Fuchs, H., Paton, V., Bajamonde,A., Fleming, T., Eiermann, W., Wolter, J., Pegram, M., Baselga, J., and Norton, L.Use of chemotherapy plus a monoclonal antibody against HER2 for metastaticbreast cancer that overexpresses HER2, N Engl J Med. 344: 783-92. 2001.89. Park, J. W. Liposome-based drug delivery in breast cancer treatment, BreastCancer Res. 4: 95-9. 2002.90. Hartman, T. and Carlsson, J. Radiation dose heterogeneity in receptor andantigen mediated boron neutron capture therapy, Radiother Oncol. 31: 61-75.1994.91. Locher, G. L. Biological effects and therapeutic possibilities of neutrons, TheAmerican Journal of Roentgenology and Radium Therapy. 36: 1-13. 1936.92. Farr, L. E., Haymaker, W., Konikowski, T., and Lippincott, S. W. Effects ofAlpha Particles Randomly Induced in the Brain in the Neutron-capture Treatmentof Intracranial Neoplasms, Int. J. Neurol. 3: 564-576. 1962.93. Sweet, W. H., Soloway, A. H., and Brownell, G. L. Boron-Slow NeutronCapture Therapy of Gliomas, Acta Radiologica. 1: 114-121. 1963.94. Hatanaka, H. and Nakagawa, Y. Clinical results of long-surviving brain tumorpatients who underwent boron neutron capture therapy, Int J Radiat Oncol BiolPhys. 28: 1061-6. 1994.95. Barth, R. F., Soloway, A. H., and Brugger, R. M. Boron Neutron CaptureTherapy of brain tumors: Past history, current status and future potential, CancerInvestigation. 14: 534-550. 1996.96. Chanana, A. D., Capala, J., Chadha, M., Coderre, J. A., Diaz, A. Z., Elowitz,E. H., Iwai, J., Joel, D. D., Liu, H. B., Ma, R., Pendzick, N., Peress, N. S., Shady,M. S., Slatkin, D. N., Tyson, G. W., and Wielopolski, L. Boron neutron capturetherapy for glioblastoma multiforme: interim results from the phase I/II dose-escalation studies, Neurosurgery. 44: 1182-92; discussion 1192-3. 1999.97. Busse, P. M., Harling, O. K., Palmer, M. R., Kiger, W. S., Kaplan, J., Kaplan,I., Chuang, C. F., Goorley, J. T., Riley, K. J., Newton, T. H., Santa Cruz, G. A.,Lu, X.-Q., and Zamenhof, R. G. A Critical Examination of the Results from theHarvard-MIT NCT Progrma Phase 1 Clinical Trial of NCT For IntracranialDisease. In: Research and Development in Neutron Capture Therapy, Essen, 2002,pp. 1123-1127.98. Palmer, M. R., Goorley, J. T., Kiger, W. S., Busse, P. M., Riley, K. J., Harling,O. K., and Zamenhof, R. G. Treatment planning and dosimetry for the Harvard-MIT Phase I clinical trial of cranial neutron capture therapy, Int J Radiat OncolBiol Phys. 53: 1361-79. 2002.99. Sauerwein, W. and Zurlo, A. The EORTC Boron Neutron Capture Therapy(BNCT) Group: achievments and future projects, Eur J Cancer. 38: S31-S34.2002.100. Wittig, A., Hideghéty, K., Paquis, P., Heimans, J., Vos, M., Goetz, C.,Haselsberger, K., Grochulla, F., Moss, R., Morrissey, J., Bourhis-Martin, E.,Rassow, J., Stecher-Rasmussen, F., Turowski, B., Wiestler, M., deVries, M. J.,Fankhauser, H., Gabel, D., and Sauerwein, W. Current Clinical Results of theEORTC-Study 11961. In: Research and Development in Neutron CaptureTherapy, Essen, 2002, pp. 1117-1122.
62
101. Joensuu, H., Kankaanranta, L., Seppälä, T., Auterinen, I., Kallio, M., Kulvik,M., Laakso, J., Vähätalo, J., Kortesniemi, M., Kotiluoto, P., Seren, T., Karila, J.,Brander, A., Järviluoma, E., Ryynänen, P., Pateau, A., Ruokonen, I., Tenhunen,M., Jääskeläinen, J., Färkkilä, M., and Savolainen, S. Preliminary Results of theFirst Finnish BPA-based BNCT Trial on Primary Glioblastoma. In: Research andDevelopment in Neutron Capture Therapy, Essen, 2002, pp. 1093-1096.102. Capala, J., Sköld, K., H-Stenstam, B., Munck af Rosenschöld, P., Giusti, V.,Persson, C., Wallin, E., Brun, A., Carlsson, J., Salford, L., Ceberg, C., Persson, B.,Pellettieri, L., and Henriksson, R. Clinical BNCT Studies in Sweden. In: Researchand Development in Neutron Capture Therapy, Essen (Germany), 2002, pp. 1101-1106.103. Capala, J., Stenstam, B. H., Sköld, K., Munck af Rosenschöld, P., Giusti, V.,Persson, C., Wallin, E., Brun, A., Franzen, L., Carlsson, J., Salford, L., Ceberg, C.,Persson, B., Pellettieri, L., and Henriksson, R. Boron neutron capture therapy forglioblastoma multiforme: clinical studies in Sweden, J Neurooncol. 62: 135-144.2003.104. Yong, J. H., Barth, R. F., Wyzlic, I. M., Soloway, A. H., and Rotaru, J. H. Invitro and in vivo evaluation of o-carboranylalanine as a potential boron deliveryagent for neutron capture therapy, Anticancer Res. 15: 2033-8. 1995.105. Pettersson, O. A., Olsson, P., Lindstrom, P., Sjoberg, S., Larsson, B. S., andCarlsson, J. Cellular binding of carboranylalanine and some effects of boronneutron capture. Analysis of cultured melanoma B16 cells, Acta Oncol. 33: 685-91. 1994.106. Al-Madhoun, A. S., Johnsamuel, J., Yan, J., Ji, W., Wang, J., Zhuo, J. C.,Lunato, A. J., Woollard, J. E., Hawk, A. E., Cosquer, G. Y., Blue, T. E., Eriksson,S., and Tjarks, W. Synthesis of a small library of 3-(carboranylalkyl)thymidinesand their biological evaluation as substrates for human thymidine kinases 1 and 2,J Med Chem. 45: 4018-28. 2002.107. Fairchild, R. G., Kahl, S. B., Laster, B. H., Kalef-Ezra, J., and Popenoe, E. A.In vitro determination of uptake, retention, distribution, biological efficacy, andtoxicity of boronated compounds for neutron capture therapy: a comparison ofporphyrins with sulfhydryl boron hydrides, Cancer Res. 50: 4860-5. 1990.108. Nguyen, T., Brownell, G. L., Holden, S. A., Kahl, S., Miura, M., and Teicher,B. A. Subcellular distribution of various boron compounds and implications fortheir efficacy in boron neutron capture therapy by Monte Carlo simulations, RadiatRes. 133: 33-40. 1993.109. Hill, J. S., Kahl, S. B., Kaye, A. H., Stylli, S. S., Koo, M. S., Gonzales, M. F.,Vardaxis, N. J., and Johnson, C. I. Selective tumor uptake of a boronatedporphyrin in an animal model of cerebral glioma, Proc Natl Acad Sci U S A. 89:1785-9. 1992.110. Ghaneolhosseini, H., Tjarks, W., and Sjoberg, S. Synthesis of BoronatedPhenantidinium Derivatives for Potential Use in Boron Neutron Capture Therapy(BNCT), Tetrahedon. 53: 17519-17526. 1997.111. Ghaneolhosseini, H., Tjarks, W., and Sjöberg, S. Synthesis of novelBoronated Acridines- and Spermidines as possible agents for BNCT, Tetrahedon.54: 3877-3884. 1998.112. Ghaneolhosseini, H. and Sjoberg, S. Synthesis of a boronated naphthalimidefor potential use in boron neutron capture therapy (BNCT), Acta Chem Scand. 53:298-300. 1999.
63
113. Nakanishi, A., Guan, L., Kane, R. R., Kasamatsu, H., and Hawthorne, M. F.Toward a cancer therapy with boron-rich oligomeric phosphate diesters that targetthe cell nucleus, Proc Natl Acad Sci U S A. 96: 238-41. 1999.114. Guan, L., Wims, L. A., Kane, R. R., Smuckler, M. B., Morrison, S. L., andHawthorne, M. F. Homogeneous immunoconjugates for boron neutron-capturetherapy: design, synthesis, and preliminary characterization, Proc Natl Acad Sci US A. 95: 13206-10. 1998.115. Primus, F. J., Pak, R. H., Richard-Dickson, K. J., Szalai, G., Bolen, J. L., Jr.,Kane, R. R., and Hawthorne, M. F. Bispecific antibody mediated targeting of nido-carboranes to human colon carcinoma cells, Bioconjug Chem. 7: 532-5. 1996.116. Barth, R. F., Yang, W., Adams, D. M., Rotaru, J. H., Shukla, S., Sekido, M.,Tjarks, W., Fenstermaker, R. A., Ciesielski, M., Nawrocky, M. M., and Coderre, J.A. Molecular targeting of the epidermal growth factor receptor for neutron capturetherapy of gliomas, Cancer Res. 62: 3159-66. 2002.117. Setiawan, Y., Moore, D. E., and Allen, B. J. Selective uptake of boronatedlow-density lipoprotein in melanoma xenografts achieved by diet supplementation,Br J Cancer. 74: 1705-8. 1996.118. Ji, B., Peacock, G., and Lu, D. R. Synthesis of cholesterol-carboraneconjugate for targeted drug delivery, Bioorg Med Chem Lett. 12: 2455-8. 2002.119. Shelly, K., Feakes, D. A., Hawthorne, M. F., Schmidt, P. G., Krisch, T. A.,and Bauer, W. F. Model studies directed toward the boron neutron-capture therapyof cancer: boron delivery to murine tumors with liposomes, Proc Natl Acad Sci US A. 89: 9039-43. 1992.120. Feakes, D. A., Shelly, K., Knobler, C. B., and Hawthorne, M. F. Na3[B20H17 NH3]: Synthesis and liposomal delivery to murine tumors, Proc Natl AcadSci U S A. 91: 3029-3033. 1994.121. Feakes, D. A., Shelly, K., and Hawthorne, M. F. Selective boron delivery tomurine tumors by lipophilic species incorporated in the membranes of unilamellarliposomes, Proc Natl Acad Sci U S A. 92: 1367-70. 1995.122. Hawthorne, M. F. and Shelly, K. Liposomes as drug delivery vehicles forboron agents, J Neurooncol. 33: 53-8. 1997.123. Mehta, S. C., Lai, J. C., and Lu, D. R. Liposomal formulations containingsodium mercaptoundecahydrododecaborate (BSH) for boron neutron capturetherapy, J-Microencapsul. 13: 269-79. 1996.124. Moraes, A. M., Santana, H. A., and Carbonell, R. G. Preparation andcharacterization of liposomal systems entrapping the boronated compound o-carboranylpropylamine, J. Microencapsulation. 16: 647-664. 1999.125. Yanagie, H., Kobayashi, H., Takeda, Y., Yoshizaki, I., Nonaka, Y., Naka, S.,Nojiri, A., Shinnkawa, H., Furuya, Y., Niwa, H., Ariki, K., Yasuhara, H., andEriguchi, M. Inhibition of growth of human breast cancer cells in culture byneutron capture using liposomes containing 10B, Biomed Pharmacother. 56: 93-9.2002.126. Yanagie, H., Tomita, T., Kobayashi, H., Fujii, Y., Nonaka, Y., Saegusa, Y.,Hasumi, K., Eriguchi, M., Kobayashi, T., and Ono, K. Inhibition of humanpancreatic cancer growth in nude mice by boron neutron capture therapy, Br JCancer. 75: 660-5. 1997.127. Yanagie, H., Tomita, T., Kobayashi, H., Fujii, Y., Takahashi, T., Hasumi, K.,Nariuchi, H., and Sekiguchi, M. Application of boronated anti-CEAimmunoliposome to tumour cell growth inhibition in in vitro boron neutroncapture therapy model, Br J Cancer. 63: 522-6. 1991.
64
128. Woodle, M. C. Sterically stabilized liposome therapeutics, Advanced drugdelivery reviews. 16: 249-265. 1995.129. Johnsson, M., Bergstrand, N., and Edwards, K. Optimization of drug loadingprocedures and characterization of liposomal formulations of two novel agentsintended for boron neutron capture therapy (BNCT), Journal of liposome research.9: 53-79. 1999.130. Neidle, S. and Abraham, Z. Structural and sequence-dependent aspects ofdrug intercalation into nucleic acids, CRC Crit Rev Biochem. 17: 73-121. 1984.131. Sobell, H. M., Tsai, C. C., Jain, S. C., and Gilbert, S. G. Visualization ofdrug-nucleic acid interactions at atomic resolution. III. Unifying structuralconcepts in understanding drug-DNA interactions and their broader implications inunderstanding protein-DNA interactions, J Mol Biol. 114: 333-65. 1977.132. Blake, A. and Peacocke, A. R. The interaction of aminoacridines with nucleicacids, Biopolymers. 6: 1225-1253. 1968.133. Drummond, D. S., Simpson-Gildemeister, V. F. W., and Peacocke, A. R.Interaction of aminoacridines with deoxyribonucleic acid: Effects of ionicstrength, denaturation and structure, biopolymers. 3: 135-153. 1965.134. Davis, M. A. and Soloway, A. H. Carboranes- III. Boron-ContainingAcridines, Notes. 10: 730-732. 1967.135. Martin, R. F. Induction of double-stranded breaks in DNA by binding with an125i-labelled acridine, Int J Radiat Biol Relat Stud Phys Chem Med. 32: 491-7.1977.136. Martin, R. F., Bradley, T. R., and Hodgson, G. S. Cytotoxicity of an 125I-labeled DNA-binding compound that induces double-stranded DNA breaks,Cancer Res. 39: 3244-7. 1979.137. Bohl Kullberg, E., Lundqvist, H., Orlova, A., Stenerlow, B., Tolmachev, V.,and Gedda, L. An 125I- aminoacridine compound for cancer radionuclide therapy:labeling and in vitro properties, Manuscript 2003.138. Bohl, E., Carlsson, J., Edwards, K., Sjöberg, S., and Gedda, L. SLT-particlesfor two-step targeting in Boron Neutron Capture Therapy. In: Frontiers in NeutronCapture Therapy, La Jolla, 1998, pp. 1069-1075.139. Westermark, B., Magnusson, A., and Heldin, C.-H. Effect of epidermalgrowth factor on membrane motility and cell locomotion in cultures of humanclonal glioma cells, Journal of Neuroscience Research. 8: 491-507. 1982.140. Haigler, H., Ash, J. F., Singer, S. J., Cohen, S., Yakes, F. M., Chinratanalab,W., Ritter, C. A., King, W., Seelig, S., and Arteaga, C. L. Visualization byfluorescence of the binding and internalization of epidermal growth factor inhuman carcinoma cells A-431, Proc Natl Acad Sci U S A. 75: 3317-21. 1978.141. Lindström, P., Olsson, P., Malmqvist, J., Pettersson, J., Lemmen, P., Werner,B., Sjöberg, S., Olin, Å., and Carlsson, J. New carborane-based compounds forboron neutron capture therapy: binding and toxicity of ANC-1, DAC-1 and B-Et-11-OMe in cultured human glioma and mouse melanoma cells, Anti-CancerDrugs. 5: 43-52. 1994.142. Soley, M. and Hollenberg, M. D. Epidermal Growth Factor (Urogastrone)-Stimulated Gluconeogenesis in Isolated mouse Hepatocytes, Archives ofBiochemistry and Biophysics. 255: 136-146. 1987.143. Uster, P. S., Allen, T. M., Daniel, B. E., Mendez, C. J., Newman, M. S., andZhu, G. Z. Insertion of poly(ethylene glycol) derivatized phospholipid into pre-formed liposomes results in prolonged in vivo circulation time., FEBS Lett. 386:243-246. 1996.
65
144. Ishida, T., Iden, D. L., and Allen, T. M. A combinatorial approach toproducing sterically stabilized (Stealth) immunoliposomal drugs, FEBS Lett. 460:129-33. 1999.145. Silvander, M., Johnsson, M., and Edwards, K. Effects of PEG-lipids onpermeability of phosphatidylcholine/cholesterol liposomes in buffer and in humanserum, Chem Phys Lipids. 97: 15-26. 1998.146. Pool, G. L., French, M. E., Edwards, R. A., Huang, L., and Lumb, R. H. Useof radiolabeled hexadecyl cholesteryl ether as a liposome marker, Lipids. 17: 448-52. 1982.147. Derksen, J. T., Morselt, H. W., and Scherphof, G. L. Processing of differentliposome markers after in vitro uptake of immunoglobulin-coated liposomes by ratliver macrophages, Biochim Biophys Acta. 931: 33-40. 1987.148. Haigler, H. T., Maxfield, F. R., Willingham, M. C., and Pastan, I.Dansylcadaverine inhibits internalization of 125I-Epidermal Growth Factor inBALB 3T3 cells, J Biol Chem. 255: 1239-1241. 1980.149. Mastrobattista, E., Koning, G. A., and Storm, G. Immunoliposomes for thetargeted delivery of antitumor drugs, Adv Drug Deliv Rev. 40: 103-127 1999.150. Drummond, D. C., Zignani, M., and Leroux, J. Current status of pH-sensitiveliposomes in drug delivery, Prog Lipid Res. 39: 409-60. 2000.151. Slepushkin, V. A., Simoes, S., Dazin, P., Newman, M. S., Guo, L. S., Pedrosode Lima, M. C., and Duzgunes, N. Sterically stabilized pH-sensitive liposomes.Intracellular delivery of aqueous contents and prolonged circulation in vivo, J BiolChem. 272: 2382-8. 1997.152. Gabizon, A., Horowitz, A. T., Goren, D., Tzemach, D., Mandelbaum-Shavit,F., Qazen, M. M., and Zalipsky, S. Targeting folate receptor with folate linked toextremities of poly(ethylene glycol)-grafted liposomes: in vitro studies, BioconjugChem. 10: 289-98. 1999.153. Mateo de Acosta, C., Justiz, E., Skoog, L., and Lage, A. Biodistribution ofradioactive epidermal growth factor in normal and tumor bearing mice, AnticancerRes. 9: 87-92. 1989.154. Jorgensen, P. E., Poulsen, S. S., and Nexo, E. Distribution of i.v. administeredepidermal growth factor in the rat, Regul Pept. 23: 161-9. 1988.155. Rieder, H., Meyer zum Buschenfelde, K. H., and Ramadori, G. Functionalspectrum of sinusoidal endothelial liver cells. Filtration, endocytosis, syntheticcapacities and intercellular communication, J Hepatol. 15: 237-50. 1992.156. Cobo, J., Hernandez, L. C., del Valle, M. E., Vijande, M., and Vega, J. A.Immunohistochemical localization of epidermal growth factor and its receptorduring odontogenesis in the rat, Eur J Orthod. 14: 333-8. 1992.157. Kassis, A. I., Harapanhalli, R. S., and Adelstein, S. J. Comparison of strandbreaks in plasmid DNA after positional changes of Auger electron-emitting iodine-125, Radiat Res. 151: 167-76. 1999.158. Park, J. W., Kirpotin, D. B., Hong, K., Shalaby, R., Shao, Y., Nielsen, U. B.,Marks, J. D., Papahadjopoulos, D., and Benz, C. C. Tumor targeting using anti-her2 immunoliposomes, J Control Release. 74: 95-113. 2001.
