Core-shell Macromolecules for Specific Cell Nucleus or/and Cell Matrix

Staining

Description

Fluorescence is ideally suited for observing the location of molecules in cells as it is non-invasive and can be detected with high sensitivity and signal specificity. The spectroscopic properties of fluorescence can be exploited to obtain information not only on the location of labelled macromolecules but also on the immediate molecular environment.

An ideal fluorescent dye for cell staining should combine water solubility, sufficient fluorescence quantum yield, high chemical and photostability, low toxicity, good biocompatibility and scalable production. The highly fluorescent perylene-SAθ.IO-tetracarboxydiimide (PDI) chromophore is a popular dye and pigment because of its excellent chemical, thermal and photochemical stability. (Y. Nagao, T. Misono, Dyes Pigm. 1984, 5, 171-188; H. Zollinger, Color Chemistry, VCH, Weinheim, 1987; R. M. Christie, Polym. Int. 1994, 34, 351-361 ).

Due to these outstanding properties, there have been several successful applications of PDI chromophores in diverse fields, such as dye lasers, photovoltaic cells, organic light-emitting diodes, light-harvesting complexes, photoreactive thin films, and solar cells.

Recently, attempts have been made to develop water-soluble and fluorescent PDI dyes for biological applications such as the in vitro staining of cells and proteins ( a) J. Qu, C. Kohl, M. Pottek, K. Mullen, Angew. Chem. 2004, 116, 1554-1557; Angew. Chem., Int. Ed. 2004, 43, 1528-1531 ; b) F. Yukruk, A. L. Dogan, H. Canpinar, D. Guc, E. U. Akkaya, Org. Lett. 2005, 7, 2885-2887; c) C. Kohl, T. Weil, J. Qu, K. Mullen, Chem. Eur. J. 2004, 10, 5297-5310. d) C. Jung, B. K. Mϋller, D. C. Lamb, F. Nolde, K. Mullen, C.

Brauchle, J. Am. Chem. Soc. 2006, 128, 5283-5291 ; e) A. Margineanu, J. Hofkens, M. Cotlet, S. Habuchi, A. Stefan, J. Qu, C. Kohl, K. Mullen, J. Vercammen, Y. Engelborghs, T. Gensch, F. C. De Schryver, J. Phys. Chem. B 2004, 108, 12242-12251. 8f). J. Qu ₁ N. G. Pschirer, D. Liu, A. Stefan, F. C. De Schryver, K. Muellen, Chemistry - A European Journal 2004, 10, 528- 537).

However, the combination of water solubility and high fluorescence quantum yields still represents a challenging goal since PDI dyes have a strong tendency to form aggregates in aqueous solution even at very low concentrations ( a) G. Schnurpfeil, J. Stark, D. Wohrle, Dyes Pigm. 1995, 27, 339-350; b) H. Langhals, German Patent DE-3703513, 1987; H. Langhals, W. Jona, F. Einsiedl, S. Wohnlich, Adv. Mater. 1998, 10, 1022-1024).

Therefore, there is considerable interest in the development of novel synthetic strategies toward water-soluble PDI derivatives with improved optical properties. In general, there are two major strategies for the introduction of functional groups into the PDI chromophore in order to achieve water solubility: functionalization of the imide region or functionalization of the bay region. It has been reported that the first strategy leads to poor water solubility and moderate fluorescence quantum yields in water. Moreover, recently, the uptake of such ionic PDIs by living cells in combination with no cytotoxicity raise the possibility to investigate cell uptake processes in a systematic way (J. Qu, C. Kohl, M. Pottek, K. Mullen, Angew. Chem. 2004, 116, 1554-1557; Angew. Chem., Int. Ed. 2004, 43, 1528- 1531 ).

All ionic PDI chromophores reported so far were found evenly distributed in the cellular cytoplasm, thus revealing a lack of binding specificity for subcellular components. Although there exist several fluorescent dyes for staining the cell nucleus by interaction with DNA, no histone-specific fluorescent dyes have been available so far. Moreover, no fluorescent dyes have been reported yet to specifically stain the extracellular cell matrix (ECM).

In view of the above problems, there is a need in the art for new improved fluorescent macromolecules, which can be applied in a biological environment and which reveal a good water-solubility and provide high fluorescence quantum yields.

Recently, the preparation of core shell macromolecules consisting of a non- fluorescent polyphenylene dendrimer core and a flexible polymer shell has been reported by V. Atanasov, V. Sinigersky, M. Klapper and K. Mullen in Macromolecules 2005, 38, 1672-1683. The authors obtained complex core shell structures with layers of different polarity and flexibility. The stiff core - flexible shell nanoparticles are investigated as support for metallocenes catalysing the olefin polymerisation. However, the described molecules are not fluorescent. Core shell macromolecules with a tailored profile, i. e. high structural perfection, photostability, water-solubility, and biological specificity have not been demonstrated by the prior art.

The inventors now found out that a fluorescent dye fulfilling the above demands can be provided by a macromolecule consisting of a fluorescent core, an oligophenylene shell and a charged flexible polymer shell.

Subject matter of the invention is a macromolecule of formula (I)

A B _X C _Y

wherein

A is a chromophore comprising a polycyclic aromatic hydrocarbon;

B _x is a first polymer shell, wherein B is an oligo(phenylene)-based polymer comprising 1-

100 phenyl rings, and x is an integer from 1-64; Cy is a second polymer shell, wherein C is a flexible polymer comprising 1 to 500 repeating units, and carrying at least one charged group; and

y is an integer from 4-32.

The first oligo(phenylene)-based polymer shell B _x is suitable to prevent the central chromophore A from aggregation in aqueous media, which usually leads to reduced fluorescence quantum yields.

The chromophore A preferably comprises 2 to 20 annelated phenylene rings. More preferred are polycyclic aromatic hydrocarbons comprising 4-11 or 5-8 annelated phenylene rings. Exemplary chromophores of the invention are based on polycyclic aromatic hydrocarbons selected from the group consisting of naphthalene, pyrene, perrylene, terrylene or quaterrylene and derivatives thereof, which may be substituted or unsubstituted. Mono- or diimide-derivatives thereof are particularly preferred.

naphthene pyrene perrylene terrylene quateπylene

quaterrylenediimide (QDI)

According to an embodiment of the invention, the polycyclic aromatic hydrocarbon is selected from naphtalene, naphtalene-monoimide, naphtalene-diimide, pyrene, perylene, perylene-monoimide, perylene- diimide, terrylene and derivatives thereof.

in the case of imide derivatives, the imide nitrogen is in each case independently bound to a radical R ¹ . Preferably, R ¹ is in each case independently an organic radical, in particular a secondary or tertiary aliphatic or aromatic, linear, branched chain or cyclic radical having typically 3-30 carbon atoms and optionally one or more heteroatoms which are preferably selected from N ₁ O and/or S. Examples for suitable radicals are in particular mono- or bicyclic aromatic or heteroaromatic radicals, for instance phenyl, pyridyl or naphthyl, which optionally bears one or more substituents.

Examples of suitable substituents are R ² , CN, NO ₂ , halogen (e.g. F, Cl, Br or I), OH, OR ² , OCOR ² , SH, SR ² , SCOR ² , SO ₂ R ² , CHO, COR ² , COOH, COOR ² , CONH ₂ , CONHR ² , CON(R ² ) ₂ , SO ₃ H, SO ₃ M, SO ₃ R ² , NH ₂ , NHR ² or N(R ² ) ₂ , wherein M is a cation, e.g. an alkali metal ion such as sodium, potassium, etc., and R ² is an optionally halogen-substituted linear or branched chain Ci-Ce-alkyl moiety.

Preferably, the chromophore A is selected from

1 ,6,7,12-tetraphenoxy-N,N'-(2,6-diisopropylphenyl)perylene-3, 4,9, 10- tetracarboxdiimide (PDI);

1 ,6,7,12-tetraphenoxy-N-(2, 6-diisopropylphenyl)-perylene-3,4-dicarboxy- imide (PMI);

1 ,6,9,14-tetraphenoxy-N,N'-(2,6-diisopropylphenyl)terrylene-3 ,4,11 ,12- tetracarboxdiimide (TDI);

1 ,6,9,14-tetraphenoxy-N-(2, 6-diisopropylphenyl)-terrylene-3,4-dicarboxy- imide (TMI); and i .e.H .Iθ-tetraphenoxy-N.N'^.e-diisopropylphenyOquaterrylene-S^.I S.M-

tetracarboxdiimide (QDI).

More preferably, A is

I.θJ.^-tetraphenoxy-N.N'-^.δ-diisopropylphenyOperylene- SAQ.IO- tetracarboxdiimide (PDI) or

1 ,6,9,14-tetraphenoxy-N,Nχ2,6-diisopropylphenyl)terrylene-3, 4, 11.12- tetracarboxdiimide (TDI).

The central chromophore A is surrounded by a first shell B _x , comprising x oligo(phenylene)-based polymers B which may be the same or different. According to the invention, B comprises 1-100 phenylene rings, preferably 1- 64 phenylene rings, more preferably 4-16 phenylene rings. Suitably, the oligo(phenylene)-based shell B _x is a rigid dendrimer shell, capable to suppress aggregation of the chromophore in aqueous solution, particularly preferred are first and second generation dendrimers.

The oligo(phenylene)-based polymers B may be bound to the chromophore for instance via a direct bond or via a bifunctional linker such as for example Ci-C ₆ alkylene, -O-, -S-, etc. Further examples of suitable linkers are given below.

The second polymer shell C _y enables the introduction of different numbers of polar functionalities. Thus it is possible to achieve tailor made properties with regard to charges, charge densities and water solubility. C _y comprises y flexible polymers C ₁ which may be the same or different, each comprising 1- 500 repeating units, preferably 2-200 repeating units, and carrying at least one charged group, y is an integer from 4-32, preferably 4-16.

One or more flexible polymers C may be covalently bound to B, in particular C may be bound via a direct bond or via a bifunctional linker. According to a preferred embodiment of the invention each B is bound to at least one C.

Bifunktional linkers suitable for the purpose of the invention are for example linear or branched chain alkylene, aralkylene, arylene, which may optionally be substituted with suitable substituents as defined above, -O-, -CO-, -COO-, -OR-, -COR-, -COOR-, -ROR-, -RCOR-, -RCOOR-, -NH-CO- -NH-, etc., wherein R is for example linear or branched chain alkylene, aralkylene or arylene.

The flexible polymer is preferably selected from the group consisting of linear or branched chain, substituted or unsubstituted polyethylene, polypropylene, polyacrylate, polymetacrylate, polyacrylamide and/or polymetacrylamide. [Examples of suitable substituents are as given above.

According to the invention a flexible polymer C carries at least one positively or negatively charged group. The charged group is preferably a charged group, which is charged in neutral media, for example at pH 7. Examples for negatively charged groups are carboxylic acid or sulfonic acid groups, -CO ₂ H or -SO ₃ H. Positively charged groups are preferably amines, for instance an amino group or an ammonium group, in particular a quatemized ammonium group, or an alkylated heteroaromatic nitrogen atom, in particular an N- alkylpyridinium, N-alkylquinolinium or N-alkylisoquinolinium group, wherein the alkyl moiety preferably has up to 6 carbon atoms and may optionally be substituted as described above. The charged group(s) can be located at the polymer main chain and/or at side chains.

Examples of flexible polymers according to the invention are: H poly(meth)acrylic acid ide

In addition to inducing water solubility, the charged flexible polymer shell also provides the macromolecules of the invention with improved binding properties that allow the application of the described compounds as colorants in cytochemistry and histochemistry.

A further embodiment of the invention relates to the preparation of the above-described charged core-shell macromolecules. Subject matter of the invention is a method for preparing a macromolecule of formula (I) comprising the steps i) providing a chromophore A, ii) building up a first polymer shell B _x around A, iii) building up a second polymer shell C ₇ around B _x .

Step (ii) preferably comprises one or more Diels-Alder [3+2] cycloaddition reactions and/or palladium(O) catalyzed coupling reactions such as Suzuki reactions. For this purpose, for instance a chromophore A may be used, which has one or more substituents providing reactive functionalities. Reactive functionalities may for example be alkynyl groups. In that case, the first polymer shell may be built up by means of a Diels-Alder reaction with e.g. cyclopentadienone or derivatives thereof. It is particularly preferred to use cyclopentadienone derivatives, which are substituted by one or more phenyl rings. Further, additional substituents may be present at the cyclopentadienone or at phenyl residues bound thereto, Preferably, such additional substituents provide suitable reactive groups, which can then be

applied in step (iii) for creating a second polymer shell.

In step (iii), flexible polymer chains are attached to the polyphenylene shell in order to build up an outer second polymer shell C _γ . Step (iii) involves for instance living anionic polymerization reactions and/or controlled radical polymerization.

Further, the present invention relates to the use of the above-described macromolecules of formula (I). According to an embodiment of the invention macromolecules bearing negatively charged groups (e.g. carboxylic acid groups) are capable to specifically stain those compartments of the nucleus where positively charged histones are enriched along with the genomic DNA. In the cell nucleus, DNA is packed into a dense structure called chromatin. Chromatin is largely made up of higher order arrangements of nucleosomes. In nucleosomes, DNA is tightly wrapped around a core of histone proteins. The protein complement of chromatin largely consists of histones, basic proteins which are positively charged at neutral pH, and non-histone chromosomal proteins which are predominantly acidic at neutral pH. ^{15a b> c} . DNA-histone binding is attributed to charge-charge interactions as well as to hydrogen bonds and salt links between DNA-phosphate oxygen atoms and protein basic and hydroxyl side chain groups. ^15d The inventors found out, that these negatively charged water-soluble core-shell macromolecules potentially interact with histones by mechanisms similar to the DNA-histone interactions. It is assumed that the interaction is due to the presence of negatively charged groups such as -COOH groups, which are capable to interact with histones by electrostatic interaction as well as by hydrogen bonds and salt links.

Due to their ease of synthesis and their high fluorescence quantum yields, these compounds of the invention offer an attractive alternative to conventional expensive antibodies or chromophores displaying broad emission spectra such as DAPI.

According to another embodiment of the invention, macromolecules bearing positively charged groups (e.g. amine groups) are capable to specifically stain the extracellular matrix (ECM).

All cells in animal tissue are surrounded by ECM or, in the case of epithelia, are bounded on one side by the basement membrane, a specialized ECM. The ECM provides mechanical strength and elasticity, absorbs shock from extracellular change, retention of water, transports signals between cells, binds growth factors, and interacts with cell-surface receptors. The ECM does not constitute an inert environment but is affected by and feeds back on cellular physiology. The ECM in general is composed of highly hydrated proteins and carbohydrates. Main protein components are collagen and elastin, the glycoproteins fibronectin and laminin, and proteoglycans containing long glycosaminoglycan (GAG) side chains (Fig. 1B). GAGs ₁ which can also occur in free form in the ECM, are build from characteristic repeating disaccharide units and are highly negatively charged due to the presence of uronic acids and sulphate modifications. The main GAGs are hyaluronan, chondrotin and dermatan sulfate, heparan sulfate, and keratan sulfate.

Before the present invention only few histochemical stains against components of the ECM were available, some of which require harsh staining conditions. At present, the ECM is mostly detected by immunofluorescence against individual ECM components. A large variety of mono- and polyclonal sera are available from various commercial and noncommercial sources. Generally, little is known about the species specificity of these antisera. Because proteoglycans are structurally heterogeneous, antisera raised against a given antigen, can yield non-overlapping staining patterns on tissue sections. A general fluorescent stain for ECM would, therefore, be useful for many purposes.

The present inventors found out that a positively charged core-shell macromolecule of the invention is able to bind to the highly negatively

charged ECM, and, thus, is applicable to label ECM in animal tissue.

Surprisingly, it was further found out, that beside the above described positively charged core-shell macromolecules of formula J also linear macromolecules comprising a chromophore A and a flexible polymer C comprising 1 to 500 repeating units, and carrying least one positively charged group, are suitable to selectively stain the extracellular matrix.

A and C are as defined above in context with macromolecules of formula (I). Preferred examples of chromophore A and flexible polymer C are as described above.

According to this embodiment of the invention, C is covalently bound to the chromophore A via a direct bond or via a bifunctional linker as defined above.

Linear macromolecules comprising a chromophore A and a flexible polymer can be prepared by a method involving living anionic polymerization reactions and/or controlled radical polymerization (NMP, ATRP, RAFT).

The present invention is illustrated by means of the following examples and figures.

Brief description of the figures

Fig. 1 shows PDI labeled star polymers containing peripheral carboxylic acid (P1 and P2) and hydroxyl-groups (P3) groups.

Fig. 2: (A) Absorption of P1 (dash) and P2 (line) in water;

(B) Normalized emission spectra of P1 (dash) and P2 (line) in water.

Fig. 3 shows the structure of the negatively charged PDI derivative 8.

Fig. 4 Dot blotting experiment: P1 and P2 bind to histones but not BSA.

Fig. 5 (A). Normalized emission spectra of histones (dotted line), P1 (dashed line) and mixture of P1 and histones (continuous line) in water;

(B). Normalized emission spectra of histones (dotted line), P2 (dashed line), the mixture of P2 and histones (continuous line) in water.

Fig. 6 Staining of Drosophila larval tissues by P1 (A to E), P2 (F), P3 (G), the DNA-specific detection agent DAPI (H), and anti-OMB (I). (A-E) P1 shows exclusively nuclear localization in the squamous epithelium of the wing (A) or leg imaginal disc (B), in the columnar epithelium of the antennal imaginal disc (C), in the fat body (D), in tracheal cells (E). P2 localizes to nuclei but also to the cell membrane (*) of fat body cells (F). No staining of larval tissue with P3 (trachea, G). Nuclear staining by DAPI (wing disc), (H). Nuclear staining by anti-OMB (trachea, I).

Fig. 7 Confocal images of P1 staining of Drosophila wing columnar (A, C and D) and squamous epithelium (B). (A-D) P1 (red) showed exclusively nuclear localization in the wing epithelium. Cell membrane was marked by CD8-GFP (green) (A). Microtubule web was marked by anti-Tubulin (green) (B). (C) P1 (red) and Omb (green) showed a complementary nuclear staining pattern. (D) P1 (red) and DAPI (blue) staining was essentially coincident. Single channels were separated in (D') and (D").

Fig. 8 Gel-electrophoresis of free DNA and the mixture of DNA and P1.

Note: P1 treated DNA fragments migrated at same rate as free

DNA, indicating that P1 did not interact with DNA.

Fig. 9 (A). Diagram to illustrate the Drosophila wing epithelium.

(B) Variety of possible arrangements of fibres network in ECM. (C) Disaccharide building blocks of the sulfated GAGs chondroitin sulfate (CS), dermatan sulfate (DS), heparan sulfate (HS) and keratan sulfate (KS).

Fig. 10 Structures of the PDI derivatives containing multiple amine groups (P4-P6) and toluidine blue (TB).

Fig. 11 Dot blot assay for P4-protein interaction.

P4 did not accumulate on the dot of collagen (A, under VIS light; A', under UV light). P4 accumulated on the dot of aggrecan (B, under VIS light; B', under UV light).

Fig. 12 Normalized emission spectra of aggrecan (dot), P4 (dash) and mixture of P4 and aggrecan in water.

Fig. 13 PDI labelled polymers specificcally stained the ECM in Drosophila wing imaginal discs. (A) A VIS light picture in X-Z plane showed the outline of the epithelia by Toluidine blue staining. (B) A electron microscope picture in X-Z plane which was positioned by a box in (A) showed the ECM surrounding the epithela. (C) A fluorescence image in X-Z plane showed the ECM by P4 (red) staining and the cell membranes of the epithelia by GFP (green) antibody staining. Fluorescence images of ECM network stained by PDI labelled polymers P4 (D), P5 (E) and P6 (F) in X-Y planes.

Fig. 14 H-NMR spectrum of macroinitiators 5 (250 Mhz, CD ₂ CI ₂ , 298 K).

Fig. 15 GPC curves of the initiator 4 and star polymers 14a, 14b, 14c.

Fig. 16 (A) The absorption of P4 (solid), P5 (dash) and P6 (dot) in water; (B) Normalized emission spectra of P4 (solid), P5 (dash) and P6 (dot) in water.

Fig. 17 VIS images of P5 (A) and P6 (B) accumulated on the dot of Aggrecan.

EXAMPLES

EXAMPLE 1 : Negatively Charged Macromolecules

1.1 Synthesis and characterization of core-shell macromolecules P1 and P2 carrying negatively charged groups and comparative macromolecule P3

The dendritic core-shell macromolecules P1 and P2 consist of a central perylenediimide (PDI) chromophore, a polyphenylene dendrimer scaffold and a polymer shell with multiple functional groups. As discussed previously, the high tendency of the PDI chromophore to form aggregates in aqueous media usually leads to a significant reduction of fluorescence quantum yields. Therefore a rigid polyphenylene dendrimer shell was introduced in the bay region of the PDI dye in order to suppress the aggregation of the central PDI chromophore (Scheme 1).

Subsequently, flexible polymer chains containing polar functionalities were attached to the polyphenylene dendrimer shell in order to build up an outer, polar shell. In this way, fluorescent and water-soluble macromolecules were achieved thus facilitating their study in a biological environment. The preparation of core-shell macromolecules consisting of precisely eight or sixteen arms by the core-first method required the synthesis of an initiator with eight or sixteen initiator sites.

Scheme 1. Synthesis of initiators 4, PDI-Gi-(OCOC(CH ₃ ) ₂ Br) ₈₎ 5, PDI-G ₂ - (OCOC(CH ₃ ) ₂ Br) ₁₆ . (a) 1.5 eq. 3, o-xylene, 17O ⁰ C, microwave 300 W ₁ 92 %. (b) 2 eq. 3, o-xylene, 17O ⁰ C ₁ microwave 300 W ₁ 90 %.

The detailed synthesis of macroinitiators possessing eight or sixteen 2- bromo-2-methylpropionic ester initiating sites around a first generation or a second generation of polyphenylene dendrimer core was as described below (4 and 5, Scheme 1 ). ¹²

A synthetic strategy towards the introduction of a PDI chromophore into the centre of a polyphenylene dendrimer was reported before and it was based on a PDI chromophore carrying four ethynyl groups (1, Scheme 1). ^w According to the present invention, free ethynyl groups of the PDI chromophore were reacted with four tetraphenylcyclopentadienones 3 each carrying two 2-bromo-2-methylpropionic ester groups which are suited as initiators for atom transfer radical polymerization (ATRP) for the build up of the hydrophilic polymer shell. The reaction proceeded in a microwave reactor at 170 ⁰ C for 2-3 hours to give the first-generation polyphenylene

dendrimer 4 with eight 2-bromo-2-methylpropionic ester groups. ¹¹ " The functionalized cyclopentadienone 3 survives the reaction conditions required for the Diels-Alder cycloaddition since decomposition reactions occurred only above 250 °C. ^11c A second-generation polyphenylene dendrimer was achieved by reacting 1 with tri/sopropyisilylcyclopentadienone (Cp-Tips ₂ ), subsequent deprotection of the tri/sopropylsilyl groups in the presence of fluoride ions to give a first-generation polyphenylene dendrimer 2 with eight ethynyl groups. ⁸¹ Then, 2 reacted with 3 thus yielding the multifunctional initiator 5 with sixteen 2-bromo-2-methylpropionic ester groups at the periphery.

The subsequent preparation of core-shell macromolecules 6 or 7, having precisely eight or sixteen arms, was achieved via an atom transfer reaction polymerization (ATRP) of tert-butyl acrylate (tBA) by using 4 or 5 as macroinitiators (Scheme 2). ¹¹ In order to avoid intermolecular combination of the growing polymer chains, the polymerization conversion was controlled at conversions below 20 %, ¹³ monitored by ¹ H-NMR (Table 1 ).

Table 1. Molecular weights of 6, and 7, prepared by ATRP of te/t-butyl acrylate (tBA) at 70 ⁰ C (M:Br-l:CuBr:DTB-bipy = 400:1:1 :1).

Finally, well-defined core-shell poly(tBA) (6 and 7, Scheme 2) were obtained. The ATRP data of the products is given in Table 1. The molecular weights obtained by GPC are lower than the theoretical molecular weights (Mtheor), calculated by conversion. Such systematic errors have already been reported in previous research work ¹³ ' ^14b and were attributed to the difference in shape between the star polymers and the linear standards used in the GPC measurements. The GPC curves showed mono-modal and narrow distributions (polydispersities (PD) of 6 and 7 are 1.15 and 1.25,

respectively) indicating that all initiating sites participated in the initiation of the polymerization. ^14b Compound 6 is based on a first generation polyphenylene dendrimer with eight arms and compound 7 consists of a second generation polyphenylene dendrimer with sixteen arms. Both macromolecules are soluble in organic solvents such as dichloromethane and tetrahydrofurane.

i 7: R = C(CH ₃ J ₃ ii P2: R = H

Scheme 2. Synthesis strategy towards PDI labeled star polymer P1 and P2. i: te/t-butyl acrylate, CuBr 100 ⁰ C. ii: CF ₃ COOH, CH ₂ CI ₂ , RT.

Scheme 3. Synthesis of PDI labeled star polymer containing -OH groups (P3).

Subsequently, the removal of the fe/f-butyl protective groups under acidic conditions produced the final products P1 and P2 (Scheme 2). ¹⁵ Due to the attached functional polymer shell containing multiple carboxylic acid groups, these materials showed good water solubility (>40 g/L in water). The hydrodynamic radii of P1 and P2 were measured by Fluorescence Correlation Spectroscopy studies (FCS). As expected, the hydrodynamic radius (R _h ) of the macromolecule increased with the generation of the dendritic shell. A hydrodynamic radius of 4 nm was found for P1 based on a first-generation polyphenylene dendrimer with eight arms, and a R _h of 9 nm for P2 based on a second-generation polyphenylene dendrimer with 16 arms (Table 2).

Table 2. fluorescence maxima fluorescence quantum yields (φ _f ) ^[8a] and the Rh (nm) in water for PI and P2.

The UV/Vis absorption and emission spectra of P1 and P2 were measured in water (Fig. 2, Table 2). The fluorescence quantum yields (φ _f ) of P1 and P2 were obtained by using the quantum yield of the negatively charged PDI derivative 8 in water as reference. ⁸³ Interestingly, P2 had higher extinction coefficients (ε) and φ _f in water than P1. In addition, P2 showed a slight bathochromic shift of the longest wavelength absorption maximum in the UV spectrum (4 nm) in comparison to P1 (Fig. 2A). However, no differences were observed in the emission spectra with both P1 and P2 revealing an emission maximum at 619 nm (Fig. 2B).

Emission maxima above 600 nm are generally considered to be advantageous for biological imaging experiments since the background coming from the autofluorescence of living cells is negligible in this region. ^ω The intensities of the UV/Vis absorptions and the fluorescence bands of P1 and P2 in water remained almost unchanged after irradiation under UV light (350 nm, 40 W) for two days or exposure to sun light for two weeks, pointing to a good photostability of these novel fluorescent dyes.

1.2 Cytochemistry and histochemistry application

It was hypothesized that PDI labeled compounds containing multiple mono- functional carboxylic acid groups in the side chains might interact with histones by mechanisms similar to the DNA-histone interactions. In order to test this assumption, the binding of P1 to histones was tested in a nitrocellulose dot blot assay using bovine serum albumin (BSA) as a control protein. As shown in Fig. 4, histones, but not BSA, led to P1 accumulation on the protein dot. P2 showed similar binding as P1 in this assay. The higher emission intensity of histones stained with P2 versus P1 is in agreement with its higher φ _f and does not necessarily indicate stronger binding. Thus, the dot blotting experiments revealed a strong interaction between the side chains of PDI labelled compounds and histones.

In order to further characterize the interaction between PDI labelled

compounds and histories, the emission spectra of their mixtures was recorded. In contrast to the emission intensities of the isolated macromolecules, the mixtures containing histones and P1 or P2 revealed significantly increased emission intensities (φ _f = 54.0% for P2/histones and 41.7% for P1/histones, Supporting information), indicating an interaction between histones and P1 (Fig. 5A) or P2 (Fig. 5B) thus supporting the dot- blotting results. In addition, the mixtures of histones and P1 or P2 showed a hypsochromic shift (10 nm) of the emission maxima in comparison to P1 or P2 alone.

Since P1 and P2 bound to isolated histones in vitro, we determined whether they could be used to stain histones in the natural context of chromatin. Therefore, different Drosophila larval tissues were used for histochemical binding studies. Dissected third instar Drosophila larvae were fixed in formaldehyde solution and then treated with Triton X-100 solution to increase the permeability of the cell membrane. The dissected larvae were stained by an aqueous solution of P1 (0.01 g/L) for 1 h. Afterwards, the dissected larvae were washed extensively with buffer (for details see supporting information). P1 staining was found exlusively in the cell nucleus in all the tested tissues as demonstrated by single optical sections obtained by confocal microscopy (Fig. 6A-E).

In high resolution confocal microscopic sections of double stained preparations, this nuclear localization was confirmed and detailed (Fig. 7). Co-staining of P1 with the cell membrane marker CD8-GFP (Fig. 7A) and the cytoskeletal marker α-tubulin (Fig. 7B) showed no overlap. DNA is not evenly distributed in nucleus. The interchromatin compartment and various nuclear bodies contain little or no DNA ^{15θ f} . This is also reflected in the distribution of histones ¹⁵⁹ . Transcription occurs at the interface between condensed chromatin and the interchromatin compartment ^15β .To test in which part of the nucleus P1 staining was localized, a commercial DNA dye 4',6-diamidino-2-phenylindole (DAPI), and an antibody directed against the nuclear transcription factor Optomotor-blind (anti-Omb) ¹⁶ , were selected for

double staining with P1. In agreement with current views of nuclear architecture ¹⁵ ', P1 and Omb showed a complementary nuclear staining pattern (Fig. 7C). On the other hand, DAPI and P1 staining was essentially coincident (Fig. 7D ₁ D ^' , D ^" ). Therefore, it was concluded that P1 specifically stains those compartments of the nucleus where positively charged histones are enriched along with the genomic DNA. Agarose gel electrophoresis showed that P1 did not interact with free DNA (Fig.8). Together with the results of dot-blotting and emission spectra of P1/histones this leads to the conclusion that P1 nuclear staining is likely due to interaction with the positively charged histones.

In case of the larger macromolecule P2 (Rh = 9 nm), weak staining of the cell membrane was observed (asterisk) in addition to the nuclear staining (Fig. 6F). It is conceivable that P2 was trapped, due to its increased size, within the extracellular space between the adjacent cells. In order to assess the importance of the -COOH groups, the multiple -OH groups core-shell compound P3 (Fig. 1 and Scheme 3) was used to stain larval tissue under the same conditions as described above. P3 did not show any staining (Fig. 6G) and no binding to histones was observed in the dot blot assay (data not shown), indicating that the -COOH groups play a crucial role for the binding. The multiple -COOH groups in the side chains of P1 may mimic to some extent the DNA sugar-phosphate backbone allowing binding to histones by forming hydrogen bonds, salt links as well as electrostatic interactions.

P1 had a similar sensitivity and specificity as the conventional nuclear fluorescent stain DAPI (FigJD and Fig. 6H). However, DAPI has a broad fluorescence emission spectrum (380-700nm), its quantum yield is 0.043 and it is easily bleached during scanning on fluorescence microscopy. ¹⁷ The emission of P1 is much more stable than that of DAPI and the narrower emission spectrum of P1 provides the opportunity for greater choice for combinations with other fluorescent dyes when performing double or triple staining. Compared with antibody staining, the P1 staining method has the advantage of being much simpler and faster.

In summary, although there exist several fluorescent dyes for staining the cell nucleus by interaction with DNA, no histone-specific fluorescent dyes have been available so far. Carboxylated water-soluble PDI core-shell macromolecules with attractive optical properties were synthesized. They specifically stain the cell nucleus by interaction with positively charged chromatin histones. We attribute the interaction to the presence of multiple side chain -COOH groups, which could interact with histones by electrostatic interaction as well as by hydrogen bonds and salt links. Due to their ease of synthesis and their high fluorescence quantum yields, these compounds offer an attractive alternative to conventional expensive antibodies or chromophores displaying broad emission spectra such as DAPI.

1.3 Experimental

Materials

CuBr (Aldrich, 99.999%), 4,4'-di-tert-butyl-2 _I 2 ^I -bipyridine (DTB-bipy) (Aldrich, 98%), 2-butanone (ARCOS, 99%) and 1-propanol (Aldrich, 99.7%) were used as obtained. PDI-labeled first-generation dendrimer bearing eight 2- bromo-2-methylpropionic ester groups (4) and PDI-labeled second- generation dendrimer bearing sixteen 2-bromo-2-methylpropionic ester groups (5) were synthesized as reported. 1 2- tert-Butylacylate (ARCOS, 99%) was rinsed with 5% NaOH, followed with water and dried over CaCI ₂ . It was then distilled under reduced pressure from CaCI ₂ and stored under nitrogen at -20 ⁰ C.

2-Hydroxyethyl methacrylate (HEMA) (Aldrich, 97%) was purchased and purified by distillation under vacuum prior to polymerization.

ATRP oftert-Butylacrylate (tBA)

All polymerizations were performed in a Schlenk apparatus. The reaction

mixtures (tertbutylacrylate: Br in PDI derivative: copper bromide: DTB-bipy = 400:1 :1 :1, molar ratio) in 2-butanone (1 g tBA/1ml solvent) were degassed by three freeze-pump-thaw cycles and placed in a thermostated oil bath maintained at 70 ⁰ C prior to the polymerization. After a specific polymerization time (4 h for 4, 6 h for 5) the reaction mixtures were cooled down to room temperature, and the content was diluted with THF and passed through a column of neutral alumina to remove the copper salts. The polymers were precipitated from an excess of methanol, filtered, and dried under vacuum to give products 6 (Conversion: 12.5%, Mn.GPC = 48800 g/mol, PD=1.15) and 7 (Conversion: 20%, Mn ₁ GPC = 153000 g/mol, PD = 1.25) as red powders.

Hydrolysis of PtBA in core-shell macromolecules

The PtBA polymer (6 or 7, 100 mg) was placed in a Schlenk flask equipped with a magnetic stirring bar. Then, the flask was evacuated and backfilled with argon three times. Dichloromethane (20 ml) was added to dissolve the polymer. Then, trifluoroacetic acid (10 ml) was added to the solution, and the mixture was stirred at room temperature for 20 h. The solvent was removed from the resulting heterogeneous mixture, and the residual solid was washed with dichloromethane (10 ml, 3 times) followed by drying under vacuum at room temperature for 10 h to give P1 or P2 as a red solids.

ATRP of HEMA

The PDI derivative 4 bearing eight 2-bromo-2-methylpropionic ester groups was used as macroinitiator for ATRP of HEMA. The reaction mixtures (HEMA: Br in PDI derivative: copper bromide: DTB-bipy = 400:1 :1:1 , molar ratio) in a solvent mixture of 2-butanone and propanol (70:30, v/v, 1 g HEMA/4 ml solvent) was degassed by three freeze-pump-thaw cycles and placed in a thermostated oil bath maintained at 40 ⁰ C prior to the polymerization. After 30 min the reaction mixture was cooled down to room temperature, the content was diluted with methanol and passed through a

column of neutral alumina to remove the copper salts. The polymer was precipitated from an excess of diethyl ether, filtered, and dried under vacuum to give product P3 as a red powder (Conversion: 12.5%, Mn ₁ GPC = 50800 g/mol, PD=1.28).

Drosophila stocks

All fly stocks are described at http://flybase.bio.indiana.edu. The cell membrane marker CD8-GFP was expressed by the GaW-UAS system[2] using the Gal4-driver C765-Gal4. Larvae were reared at 25°C.

Histochemical staining protocol

1. Dissect Drosophila third instar larvae in cold PBT solution (PBT: PBS (phosphate-buffered saline) + 0.1% BSA + 0,1% TritonXIOO + 0.02% sodium azide). Transfer dissected larvae to a 0.5 ml reaction tube on ice containing PBT solution.

2. Fix for 30 - 40 min. rotating at room temperature in fixation solution (375 ml PBT + 20 ml of 37% formaldehyde + 2 ml of 10% Triton X-100). Rinse: 4 x in PBT Wash for 1 h rotating at room temperature.

3. Incubate dissected larvae in the primary antiserum rabbit anti-GFP

(1 :2000) (Clontech), rabbit anti-Omb (1 :1000), or α-Tubulin (1 :1000) (Sigma) overnight at 4 ⁰ C. Rinse: 4 x in PBT Wash for 0.5 h rotating at room temperature.

4. Incubate dissected larvae in Donkey anti-mouse or anti-rabbit IgG-FITC

(1 :100 dilution) (Jackson lmmuno Reserch) for 1h rotating at room temperature.

Rinse: 4 x in PBT Wash for 0.5 h rotating at room temperature. 5. Incubate dissected larvae in 0.1% P1 solution for 1 h rotating at room temperature.

Rinse: 4 x in PBT Wash for 1 h rotating at room temperature. 6. Mount in 50% glycerol, optionally containing 0.5 μg/ml DAPI.

Images were taken on a Leica Laser Scanning Confocal Microscope.

Steps 3 and 4 were omitted when no immunostaining was required.

Agarose gel electrophoresis

A 0.8% agarose (w/v) gel in 1X TBE buffer was performed for 2 h at 4 V/cm. 5μl 1kb DNAIadder (PeqLab) was incubated with 5 μl 0.2% core-sheH macromolecules for 10 min before loading. The gel was stained with 1 μg/ml ethidium bromide in IxTBE for 30 min after electrophoresis. The photograph was taken under UV light.

Dot Blotting

10 μl of 1% histone solution (Sigma, H7755) were spotted onto a nitrocellulose membrane. 10 μl of 1% BSA were spotted as control. 2 μl 0.2% core-shell compound solution were applied to the spot. After 2 min the membrane was washed with water for 5 min. After drying for 2 min the blot was photographed under UV light.

Fluorescence Correlation Spectroscopy ¹¹²¹

A commercial FCS setup manufactured by Carl Zeiss (Jena, Germany) consisting of the module ConfoCor 2 and an inverted microscope, model Axiovert 200 was used. For our experiments we employed a Zeiss C- Apochromat 40x/1.2 W water immersion objective. The fluorophores were excited by a HeNe laser at 543 nm and the emission was collected after filtering with a LP560 long pass filter. For detection, avalanche photodiodes enabling singlephoton counting were used. The eight-well, polystyrene chambered cover-glass (Lab-Tek, Nalge Nunc International) was used as sample cells for the water solutions of the dendrimers. These cells had bottom slides with optical quality surfaces and thickness of 0.17 mm. For each solution, a series of 10 measurements with a total duration of 5 min were performed. The obtained autocorrelation curves were fitted with the so-

called biophysical model function in order to determine the characteristic diffusion coefficients and the hydrodynamic radii of the compounds. The calibration of the confocal observation volume was performed using a reference dye with known diffusion coeficient i.e. Rh6G.

Fluorescence quantum yields of the mixtures ofhistones and P1 or P2

The relative fluorescence quantum yields of the mixtures of histones and P1 or P2 were calculated using the following equation:

I _R OD n _&

where Q is the quantum yield, I is the integrated emission intensity, n is the refractive index, and OD is the optical density. The subscript R refers to the reference fluorophore of known quantum yield. Herein, the negatively charged PDI derivative 8 (Fig. 3) in water (the standard value is 0.58) was used as reference. ^[8a]

EXAMPLE 2: Positively Charged Macromolecules

2.1 Synthesis and characterization of core-shell macromolecules carrying positively charged groups

The fluorescent star polymers P4 and P5 consist of a central PDI chromophore, a stiff polyphenylene dendrimer scaffold, and a polymer shell with multiple amine groups. P4 is based on a first-generation polyphenylene dendrimer with eight arms, P5 is based on a second-generation polyphenylene dendrimer with 16 arms. In order to determine how molecular shape affects the staining properties, the linear polymer P6 which is based on the perylenemonoimide (PMI) chromophore was tested as well. The photochemical properties of P4, P5, and P6 are summarized in Table 3 and

figure 16. The emission maxima of all fluorescent polymers are above 600 nm.

Table 3: Absorption (A^abs), fluorescence maxima Ow,fl _U ), fluorescence quantum yields (φ _f ) [a] and the R _h (nm) in water for P4, P5 and P6.

φ _f was measured at room temperature using the negatively charged PDI derivative 8 in water (the standard value is 0.58) as reference. ⁸³

As a first test, the interaction of P4 with characteristic ECM proteins was tested by dot blot analysis. Commercial bovine collagen and aggrecan were selected for a dot-blotting experiment. Aggrecan is a major structural proteoglycan of cartilage extracellular matrix with a 210-250 kDa core protein to which 100-150 CS, DS, and KS chains are attached (Sigma product description) (Fig. 9C). Tissue collagen occurs as big fibers composed of trimeric largely uncharged α-chains which are stabilized by cystine bonds (Sigma product description). The primary sequence of the collagen α-chains consists of a repeating tri peptide made up of glycine, proline and hydroxyproline such that collagen has little net charge. As shown in Fig. 11 A, collagen attached to a positively charged Nylon membrane did not bind P4. Filter-bound aggrecan accumulated P4 when incubated under the same conditions. The accumulation of P4 could be visualized under both VIS and UV light (Fig. 11 B). P5 and P6 showed similar binding property as P4 in this non-quantitative assay (Fig. 17). Thus the dot blot assay revealed an interaction between the amine derivatives of PDI and PMI and the highly negatively charged aggrecan.

To test for interaction between P4 and aggrecan in solution, the emission spectra of their mixtures was compared to the individual solutions. The peak emission intensity of a mixture of P4 and aggrecan was twice as large that of P4 and was associated with a slight bathochromic shift (3 nm) (Fig. 12).

Since the fluorescent polymers interacted in vitro with aggrecan which is a characteristic ECM proteoglycan, we determined whether the compounds could be used to stain ECM of animal cells. Epithelial tissues were selected for histochemical analysis because these present the main forms of cellular organization in animal tissue. The Drosophila wing imaginal disc was selected as an example of epithelial tissue.The wing imaginal discs consists of two layers of cells (squamous and columnar epithelia) shown in cross- section of a toluidin-blue stained preparation in Fig. 13A. The two epithelial layers enclose a lumenal space. The basal aspects of the imaginal disc cells are oriented toward the exterior and secrete the basal lamina. It is not selectively stained by the small cationic dye toluidin blue. Even in electron micrographs the wing disc basal lamina is a thin inconspicuous layer which outlines the imaginal disc (Fig.13 B). In contrast, P4 (and P5, P6) allowed the selective visualization of this cellular structure.

To test the specificity of the PDI amine derivatives, larval wing discs of transgenic flies which express the membrane marker CD8-GFP were stained. Dissected larvae were incubated in P4 aqueous solutions (0.1%) for 1 hour, and then subjected to GFP antibody staining. The double fluorescence staining showed the GFP immunoreactivity restricted to the cell membranes and P4 marked the ECM surrounding the epithelia (Fig.13C). In a confocal section in the plane of the ECM it was apparent that P4 could visualize the ECM fibrous network at high resolution (Fig.13D). P5 and P6 showed the same staining property as P4 (Fig.13E and 13F and Fig. 18). The common characteristic of the three polymers is the high density of mono-functional -NH ₂ groups which will be largely protonated in the neutral- buffered solution. It is likely that they are involved in electrostatic interaction and hydrogen bonds as well as salt links with the negatively charged GAG

chain sulfate groups. Despite of their different size and shape, P4, P5 and P6 with similar specificity visualized the ECM network.

In summary, a series of new fluorescent dyes were applied for the staining of the ECM and could efficiently visualize its structure at high resolution in animal tissue. These dyes can be combined with other fluorescent probes as demonstrated by the double fluorescence staining shown in Fig. 13. Due to their ease of synthesis and their narrow emission spectra, these fluorescent polymers offer an attractive tool for labeling the micro-network of ECM in life science research.

2.2 Synthesis of dendritic macroinitiators

A synthetic strategy towards the introduction of a PDI chromophore into the centre of a polyphenylene dendrimer was reported before and it was based on a PDI chromophore carrying four ethynyl groups (1, Scheme 1 ). Here, free ethynyl groups of the PDI chromophore were reacted with four tetraphenylcyclopentadienones 3 each carrying two 2-bromo-2- methylpropionic ester groups which are suited as initiators for atom transfer radical polymerization (ATRP) for the build up of the hydrophilic polymer shell. The reaction proceeded in a microwave reactor at 170 ⁰ C for 2-3 hours to give the first-generation polyphenylene dendrimer 4 with eight 2-bromo-2- methylpropionic ester groups. It was described before that the functionalized cyclopentadienone 3 survives the reaction conditions required for the Diels- Alder cycloaddition since decomposition reactions occurred only above 250

⁰ C.

A second-generation polyphenylene dendrimer was achieved by reacting 1 with tri/sopropylsilylcyclopentadienone (Cp-Tips ₂ ), subsequent deprotection of the tri/sopropylsilyl groups in the presence of fluoride ions to give a first- generation polyphenylene dendrimer 2 with eight ethynyl groups. Then, 2 reacted with 3 thus yielding the multifunctional initiator 5 with sixteen 2- bromo-2-methylpropionic ester groups at the periphery.

In order to understand the impact of the shape of a molecule on its ability to cross biological membranes or to induce cellular toxicity, a molecular initiator based on a perylenemonoimide (PMI) chromophore was synthesized in a multi-step reaction (Scheme 4). First, a Suzuki coupling reaction of compound 9 and 10 in toluene at 100 ⁰ C gave PMI 11 carrying a phenoxy group at position 9. Subsequent ether cleavage was achieved with boron tribromide to give the hydroxy-substituted PMI 12 which underwent a condensation reaction with 2-bromo-2-methylpropionyl bromide under caustic condition to give the PMI-initiator 13.

Scheme 4. Synthesis of the PMI initiator 13.

All initiators 4, 5 and 13 were characterized by NMR spectroscopy, elemental analysis and MALDI TOF mass spectrometry. MALDI-TOF mass spectra (dithranol matrix, Ag ⁺ ) of the initiators gave broad peaks (see supplementary material) most probably due to fragmentation processes which occurred during the measurement.

Therefore, the number of 2-bromo-2-methylpropionic ester groups per molecule were calculated from the ¹ H-NMR spectrum according to the relative intensities of aromatic protons at 6.4-7.4 ppm or isopropyl protons at 2.56-2.66 ppm and methyl protons at 1.95-1.96 ppm which are the main peaks corresponding to the 2-bromo-2-methylpropionic ester protons (Figure 14). It is worth mentioning that all synthesized initiators are highly soluble in

most of the common organic solvents.

2.3 Synthesis of outer polymer shell

The synthesis of the outer, hydrophilic polymer shell was achieved via ATRP of 2-tertbutoxycarbonylaminoethyl methacrylate (Boc-AEMA) with the multifunctional initiators 4, 5 as well as the monofunctional PMI initiator 13 at 70 ⁰ C. The general synthetic strategy is depicted in Scheme 5 with the first- generation macroinitiator 4 as an example 1.

Scheme 5. Synthesis of PDI labelled G1 star polymers containing functional amino groups.

In order to avoid intermolecular combination of the growing polymer chains, the polymerization conversion was controlled at conversions below 20 %, monitored by ¹ H-NMR spectroscopy. The molecular weights of the prepared core-shell macromolecules were determined using GPC and ¹ H-NMR spectroscopy. In accordance with previous observations, the theoretical molecular weights, calculated by the conversion of polymerization and estimated from the relative intensities of the signals obtained by ¹ H NMR

spectroscopy usually point towards higher molecular weights of the macromolecules in comparison to the results from GPC experiments. Such systematic errors were already reported in previous work and were attributed to the difference in shape between the star polymers and the linear standards used in the GPC measurements. Nevertheless, monomodal distributions were found in all the GPC measurements except for the star polymers with 80 repeat units of the polymers shell.

Exemplary GPC curves of the initiator 4 and the star polymers 14a, 14b, 14c with various length of polymer chain to 15, 30 and 50 units were given in

Figure 15. For initiator 4 and star polymers 14a, 14b, 14c, relatively low polydispersities were obtained with 1.06, 1.12, 1.13, 1.15, respectively. In addition, all core-shell macromolecules reveal higher molecular weights and slightly broader distributions according to GPC measurements in contrast to their respective initiators, which provided further evidence of the formation of the core-shell structures.

The preparation of water-soluble polymers was achieved via removal of the BOC- protective groups by treatment with phenol and chlorotrimethyl silane. The final products were purified by repetitive precipitation from methanol solution. The successful removal of the cytotoxic reagents phenol and chlorotrimethyl silane was proven by ¹ H-NMR and UV spectroscopy. Figure 16 gives an overview over all dendritic core-shell structures as well as the linear reference polymers that were synthesized within this study. All core- shell macromolecules P4, 15b, 15c, 15d, P5 and linear polymer P6 showed suitable water solubility (above 40 g /L in water), which is the basic requirement for cell staining experiments.

2. Photochemical properties of P4, P5 and P6

The UV/Vis absorption and emission spectra of P4, P5 and P6 were measured in water. The fluorescence quantum yields (φ _f ) of P4, P5 and P6 were obtained by using the quantum yield of the negatively charged PDI

derivative 5 in water as reference. Interestingly, both P4 and P5 revealing an absorption maximum at 580 nm (Fig. 16A). However, P5 shows a bathochromic shift of the largest wavelength emission maximum in the UV spectrum (5 nm) in comparison to P4 (Fig. 16B). In addition, P5 has higher extinction coefficients (ε) and φ _f in water than PA. In comparison with the core-shell macromolecules, the absorption and emission spectra of linear polymer P6 are much smaller.

2.3 Experimental

Materials

CuBr (Aldrich, 99.999 %), 4,4 ^I -di-teAf-butyl-2,2'-bipyridine (DTB-bipy) (Aldrich, 98%), N-(2, 6-Diisopropylphenyl)-9-bromo-perylene-3, 4-dicarboxy- imide (9, BASF >90%), 4-methoxyphenylboronic acid (10, ARCOS, 97%), phenol (ARCOS, 99%), chlorotrimethyl silane (1M in CH ₂ CI ₂ ), 2-butanone (ARCOS, 99%), trifluoroacetic acid (Roth) were used as obtained. 1,6,7,12- Tetrakis(4-ethynylphenoxy)-N _I N'-(2,6-diisopropylphenyl)perylene-3,4 _I 9,10- tetracarboxdiimide (1) and PDI-labeled first-generation dendrimer bearing 8 ethynyl groups (2) were synthesized as reported. 2-Bromo-2-methylpropionic acid 4-{5-[4-(2-bromo-2-methylpropionyloxy)-phenyl]-3-oxo-2,4-dip henyl- cyclopenta-1,4-dienyl}-phenyl ester (3) and 2-[(terf-butoxycarbonyl)- amino]ethyl methacrylate (BOC-AEMA) were synthesized according to the previous report.

Instruments

Gel permeation chromatography (GPC) analyses were performed in (i) THF with a Waters pump model 515, detectors Ri 101 ERC and UV-vis S- 3702 Soma (255 nm), temperature: 30 ⁰ C ₁ standards: PSt.

¹ H NMR spectra were recorded on a "Bruker-Spectrospin" (250 MHz) at room temperature.

¹³ C NMR spectroscopy was performed on a "Bruker AMX 300" spectrometer at room temperature.

MALDI-TOF mass spectrometry measurements were performed on VG ZAB2-SE-FPD Spectrofield, Bruker Reflex I (MALDI-ToF) and Bruker Reflex Il (MALDI-TOF) mass spectrometers.

Synthesis Procedures

Synthesis of first-generation of dendrimer with eight 2-bromo-2- methylpropionic ester groups (4).

The Diels-Alder cycloaddition of 1 ,6,7,12-tetrakis(4-ethynylphenoxy)-N,N'- (2 _I 6-diisopropylphenyl)perylene-3,4 _I 9,10-tetracarboxdiimide (1) (140 mg, 0.119 mmol) and 2-bromo-2-methyl-propionic acid 4-{5-[4-(2-bromo-2- methylpropionyloxy)-phenyl]-3-oxo-2,4-diphenylcyclopenta-1 ,4-dienyl}- phenyl ester (3) (510 mg, 0.714 mmol) was performed in 5 ml o-xylene in a microwave reactor at 170 ⁰ C (max. power 300 W, max. pressure 11 bar). The process was monitored by mass spectrometry till completion. The desired product 4 (429 mg, 92.0%) was purified by column chromatography (silica gel, CH ₂ CI ₂ /petroleum ether (2:1)) as a red powder.

¹ H-NMR (300 MHz ₁ CD ₂ CI ₂ , 300 K): δ ppm; 8.06 (s , 4H), 7.52 (s, 4H), 7.45 (t, J=7.7 Hz ₁ 2H), 7.42 (d, J=7.9 Hz, 4 H), 7.40-6.60 (m, 88H), 2.68 (m, 4 H,

CH), 1.91 (S, 24H, CH ₃ ), 1.90 (s, 24H ₁ CH ₃ ), 1.05 (d, J=6.6 Hz, 24H, CH ₃ );

¹³ C-NMR (75 MHz ₁ CD ₂ CI ₂ , 300 K): δ ppm; 170.3 (OCOC(CH ₃ ) ₂ Br), 163.6

(C=O), 156.1, 154.5, 149.3, 149.1, 146.6, 141.8, 141.6, 141.1, 140.6, 140.1,

139.9, 138.8, 138.6, 138.4, 138.3, 133.3, 132.9, 132.8, 131.9, 130.3, 128.1 , 127.6, 126.9, 126.4, 124.4, 123.4, 121.3, 121.1, 120.8, 120.1, 119.8, 56.1

(OCOC(CH ₃ ) ₂ Br), 30.8 (OCOC(CH ₃ ) ₂ Br, 29.5 (CH isopropyl), 24.1 (CH ₃ isopropyl).

IR: ^ (cm ¹ ): =3110, 3030, 2970, 2870, 1750, 1600, 1500, 1400, 1260, 1200,

1160, 1130, 1100, 700, 631 cm ¹ . UVλ/is (CH ₂ CI ₂ ): a™* (ε)= 456, 540, 580 nm; photoluminescence spectrum (H ₂ CI ₂ ): ;w = 611 nm (excitation 456 nm). (MS (MALDI-TOF): m/z (%) = 3922 ([M+H] ⁺ ), calcd. for C ₂₂₄ H ₁₇₈ Br ₈ N ₂ O ₂₄ , m/z = 3921.

Elemental analysis (%) calcd for C ₂₂₄ Hi ₇₈ Br ₈ N ₂ O ₂₄ : C 68.62, H 4.58, N 0.71 ; found: C 68.57, H 4.50, N 0.66.

Synthesis of second-generation of dendrimer with sixteen 2-bromo-2- methylpropionic ester groups (5).

PDI-labeled first-generation dendrimer bearing 8 ethynyl groups (2) (100 mg, 0.0308 mmol) and 2-bromo-2-methylpropionic acid 4-{5-[4-(2-bromo-2- methylpropionyloxy)-phenyl]-3-oxo-2,4-diphenylcyclopenta-1 ,4-dienyl}- phenyl ester (3) (351.3 mg, 0.492 mmol) were dissolved in 5 ml_ o-xylene and sealed in a 8-mL glass tube. The Diels-Alder cycloaddition was performed in a microwave reactor at 170 ⁰ C. The process was monitored by mass spectrometry. Column chromatography (silica gel, CH ₂ CI ₂ ) afforded the desired product 5 as a red powder (267 mg, 90.0%).

¹ H-NMR (300 MHz, CD ₂ CI ₂ ): δ ppm; 8.06 (s, 4H), 7.52 (m, 12H) ₁ 7.42 (d, J=7.9 Hz ₁ 4 H), 7.20-6.60 (m, 220H), 6.50-6.35 (m, 14H), 2.65 (m, 4 H), 1.91 (s, 48H), 1.90 (s, 48H) ₁ 1.04 (d, J=6.0 Hz, 24H);

¹³ C-NMR (75 MHz, CD ₂ CI ₂ ): δ ppm; 170.3 (OCOC(CH ₃ J ₂ Br), 163.6(CO), 156.1, 154.4, 149.2, 149.0, 146.5, 142.0, 141.6, 141.4, 141.3, 141.2, 141.0, 140.4, 140.2, 139.8, 139.7, 138.8, 138.7, 138.6, 138.4, 132.9, 131.9, 130.3, 129.0, 128.8, 127.5, 126.7, 124.8, 123.4, 121.3, 120.8, 120.1, 119.4, 56.2 (OCOC(CHa) ₂ Br), 30.8 (OCOC(CH ₃ ) ₂ Br, 29.5 (CH isopropyl), 24.1 (CH ₃ isopropyl). JR: v (cm- ⁱ ): =3050, 3030, 2970, 2930, 1750, 1710, 1670, 1590, 1500, 1260, 1200, 1160, 1140, 1100, 1020, 1010,700, 758, 698, 577 cm ¹ . UV/Vis (CH ₂ CI ₂ ): w (ε) = 456, 540, 580 nm; photoluminescence spectrum (CH ₂ CI ₂ ): ;w = 613 nm (excitation 456 nm); MS (MALDI-TOF): m/z (%) = 8392 (broad, [M+Ag] ⁺ ), calcd. for

C496H ₃ 78Bri6N ₂ O40, m/z = 8285, Elemental analysis (%) calcd for C ₄₉₆ H ₃₇₈ Br ₁₆ N ₂ O ₄₀ : C 71.91 , H 4.60, N 0.34; found: C 71.81 , H 4.53, N 0.31.

Synthesis of the single molecular initiator (13)

N-(2, 6-Diisopropylphenyl)-9-bromo-perylene-3, 4-dicarboxy-imide (9) (2.2 g, 3.9 mmol) and 4-methoxyphenylboronic acid (10) (500 mg, 3.3 mmol) were dissolved in 200 ml_ toluene under argon atmosphere, then 30 mL of 1 M aqueous potassium carbonate solution and tetrakis(triphenylphosphine)palladium(0) (Pd(PPh ₃ J ₄ ) (380 mg, 0.33 mmol) added. The resultant mixture was heated under argon for 24 h at 100 ⁰ C. After cooling, the organic phase washed with water three times, and then dried over magnesium sulfate. After evaporation of the solvent, the residue was purified by column chromatography (silica gel, CH ₂ CI ₂ ) to afford N-(2,6- diisopropylphenyl)-9-(4-methoxyphenyl)-perylene-3,4-dicarbox y-imide (11 , 1.38 g, 60.3%) as a purple powder.

¹ H-NMR (250 MHz, CD ₂ CI ₂ ): δ ppm; 8.53 (d, J=8.2 Hz ₁ 2H), 8.41-8.33 (m, 4H), 7.94 (d, J=8.2 Hz, 1 H), 7.54-7.39 (m, 5H), 7.25 (d, J=8.8 Hz, 2H), 7.00 (d, J=8.8 Hz, 2H), 3.83 (s, 3H), 2.68 (m, 2H), 1.05 (d, J=5.6 Hz, 12H).

¹³ C-NMR (75 MHz, CD ₂ CI ₂ ): δ ppm; 164.5 (CO), 160.1, 146.5, 143.7, 138.2, 138.0, 133.0, 133.4, 132.2, 132.0, 131.6, 130.9, 129.9, 129.6, 128.6, 128.4, 127.3, 124.4, 124.1 , 121.2, 121.0, 120.4, 114.4, 55.8 (OCH ₃ ), 29.5 (CH isopropyl), 24.1 (CH ₃ isopropyl). MS (FD, 8 kV): m/z (%): 588 (100) [M ⁺ ] (calcd for C ₄ IH ₃₃ NO ₃ : 587.7).

BBr ₃ (1 M in CH ₂ CI ₂ , 10 mL) was dropped slowly into the solution of compound 11 (1.30 g, 2.21 mmol) in CH ₂ CI ₂ (20 mL) under argon at 0 ⁰ C. After stirring for 24 h, the solvent was removed under reduced pressure and the residue purified by column chromatography (silica gel, CH ₂ CI ₂ ) to afford N-(2,6-diisopropylphenyl)-9-(4-hydroxy-phenyl)-perylene-3,4- dicarboxy-imide (12, 1.20 g, 95%) as a purple powder.

¹ H-NMR (250 MHz, DMSO): δ ppm; 9.60 (s, 1 H, OH), 8.59-8.20 (m, 6H), 7.86 (d, J=8.2 Hz, 1 H), 7.50-7.26 (m, 7H) ₁ 6.89 (d, J=8.2 Hz, 2H), 2.58 (m,

2H), 0.99 (d, J=3.8 Hz, 12H).

¹³ C-NMR (75 MHz, DMSO): δ ppm; 162.9 (CO), 157.1 , 145.2, 142.8, 137.0,

136.8, 131.6, 131.1 , 131.0, 130.5, 129.5, 129.4, 128.7, 128.6, 128.2, 127.7,

127.2, 126.8, 126.7, 124.2, 124.0, 123.1 , 120.5, 120.2, 119.7, 119.4, 115.2, 28.1 (CH isopropyl), 23.1 (CH ₃ isopropyl).

MS (FD ₁ 8 kV): m/z (%): 574 (100) [M ⁺ ] (calcd for C ₄₀ H ₃ INO ₃ : 583.7).

2-Bromo-2-methylpropionyl bromide (5 ml_) was added dropwise into the solution of compound 12 (200 mg, 0.35 mmol) in dry tetrahydrofuran (20 ml_) and triethylamine (5 ml_) at 0 ^C C. After stirring for 12 hours, the reaction was quenched with methanol. Flash chromatography (silica gel, CH ₂ CI ₂ ) afforded the desired product N-(2,6-diisopropylphenyl)-9-(4-(2-bromo-2- methylpropionic acid phenyl ester)-yl)-perylene-3,4-dicarboxy-imide (13) (739 mg, 88%) as a red powder.

¹ H-NMR (250 MHz, CD ₂ CI ₂ ): δ ppm; 8.54 (d, J=8.2 Hz, 2H), 8.41-8.38 (m, 4H), 7.93 (d, J=8.2 Hz, 1 H) ₁ 7.56-7.39 (m, 5H), 7.29-7.23 (m, 4H), 2.68 (m, 2H), 2.04 (s, 6H), 1.05 (d, J=6.6 Hz, 12H).

¹³ C-NMR (75 MHz, CD ₂ CI ₂ ): δ ppm; 170.7 (OCOC(CH ₃ ) ₂ Br), 164.4 (CO), 151.1 , 146.5, 142.6, 138.2, 138.0, 137.8, 133.9, 132.2, 132.0, 131.6, 130.9,

129.7, 129.6, 129.0, 128.8, 128.7, 127.6, 127.2, 124.5, 124.4, 121.7, 121.4,

121.3, 120.9, 120.6, 56.1 (OCOC(CH ₃ ) ₂ Br), 29.5 (CH isopropyl), 24.1 (CH ₃ isopropyl).

MS (FD, 8 kV): m/z (%): 695 (100) [M ⁺ ] (calcd for C ₄₂ H ₃₂ NO ₄ : 694.6). Elemental analysis (%) calcd for C ₄₂ H ₃₂ NO ₄ : C 73.13, H 5.02, N 1.94; found: C 72.83, H 5.15, N 1.76.

General procedure for radical polymerization by ATRP

The monomer was placed in a Schlenk flask and dissolved in the 2- butanone. CuBr were then added and the mixture was degassed by three freeze-pump-thaw cycles. The ligand 4,4'-di-tert-butyl-2 _l 2'-bipyridine (DTB- bipy) was added under argon. And then the solution was stirred at 25 ⁰ C for

10 min. Finally the initiator was added and the flask was sealed and placed in a thermostated oil bath at 7O ⁰ C to start the reaction. The polymerizations were stopped by cooling with liquid nitrogen after a defined reaction time. The monomer conversion was determined by ¹ H NMR spectroscopy. The reaction mixture was diluted and eluted through a column filled with neutral alumina to remove the copper complex. The solvent was removed under vacuum and the polymer was isolated by precipitation into methanol and drying under vacuum to constant weight.

Hydrolysis of the tert-butoxycarbonyl (BOC) in core-shell macromolecules

Core-shell star polymer of 2-[(teff-butoxycarbonyl)amino]ethyl methacrylate (BOC-AEMA) (100 mg) was dissolved in CH ₂ CI ₂ (30 mL) in a 100-mL round- bottomed flask, and then phenol (3 M in CH ₂ CI ₂ , 5 mL) and chlorotrimethyl silane (1 M in CH ₂ CI ₂ , 5 mL) were added and the mixture was stirred for 2 h at room temperature. The solution was evaporated to dryness and the residue dissolved in methanol (10 mL). The product was precipitated upon addition of this solution to a mixture of diethyl ether and dried under vacuum.

Dot Blotting

5 μl of 1% collagen (Sigma) or aggrecan (Sigma) were spotted on a positively charged Nylon membrane (Sigma). Let the membrane dry in the air. 2. Incubate the membrane in 1 %BSA for 30min.

3. Incubate the membrane in dye/ PBT solution (1 :2 (v/v)) for 30min.

4. Washing the membrane in PBT for 5 minutes.

5. Then take image under VIS/UV light.

Drosophila stocks

All the stocks are described at http: flybase.bio.indiana.edu. Cell membrane

marker CD8-GFP is expressed by GaW-UAS system. Virgins of UAS-CD8- GFP cross to males of C765-Gal4. The larvae are reared at 25C°.

Histochemistry

1. Dissect Drosophila third instar larvae in cold PBT solution (PBT: 500 ml PBS + 0.1% (0,5 g) BSA + 0,1% (5 ml of 10%) Triton + 0.02% (0.5 ml of 20%) NaAz). Transfer dissected larvae to a 0.5 ml reaction tube on ice containing PBT solution. 2. Fix for 30 - 40 min. rotating at room temperature in fixation solution (375 ml PBT + 20 ml of 37% formaldehyde + 2 ml 10% Triton X-100).

Rinse: 4 x in PBT

Wash for 1 h rotating at room temperature.

3. Incubate dissected larvae in 0.1% P1 aqueous solution for 1 h rotating at room temperature.

Rinse: 4 x in PBT

Wash for 1h rotating at room temperature.

4. Incubate dissected larvae in mouse GFP antibody (1:1000 dilution) overnight at 4°C. Rinse: 4 x in PBT

Wash for 0.5h rotating at room temperature.

5. Incubate dissected larvae in Donkey anti mouse FITC (1 :100 dilution) (Jackson lmmuno Reserch) for 1h rotating at room temperature.

Rinse: 4 x in PBT Wash for 0.5h rotating at room temperature.

6. Mount in 50% Glycerol.

7. Take images on a confocal microscopy.

References

[1] a) Y. Nagao, T. Misono, Dyes Pigm. 1984, 5, 171-188; b) H. Zollinger,

Color Chemistry, VCH, Weinheim, 1987; c) R. M. Christie, Polym. Int. 1994, 34, 351-361.

[2] a) M. Sadrai, L. Hadel, R. R. Saϋrs, S. Husain, K. Krogh-Jespersen, J. D.

Westbrook, G. R. Bird, J. Phys. Chem. B 1992, 96, 7988-7996; b) R. Gvishi,

R. Reisfeld, Z. Brushtein, Chem. Phys. Lett. 1993, 213, 338-344.

[3] a) L. Schmidt-Mende, A. Fechtenkόtter, K. Mullen, E. Moons, R. H. Friend, J. D. MacKenzie, Science 2001 , 293, 1119-1122; b) A. J. Breeze, A.

Salomon, D. S. Ginley, B. A. Gregg, H. Tillmann, H.-H. Hδrhold, Appl. Phys.

Lett. 2002, 81, 3085-3087.

[4] a) J. Kalinowski, P. D. Marco, V. Fattori, L. Giuletti, M. Cocchi, J. Appl.

Phys. 1998, 83, 4242-4248. b) M. A. Angadi, D. Gosztola, M. R. Wasielewski, Mater. ScL Eng. B 1999, 63, 191-194. c) P. Ranke, I. Bleyl, J.

Simmerer, D. Haarer, A. Bacher, H. W. Schmidt, Appl. Phys. Lett. 1997, 71,

1332-1334.

[5] a) F. Wϋrthner, A. Sautter, Org. Biomol. Chem. 2003, 1, 240-243; b) T.

Weil, E. Reuther, K. Mullen, Angew. Chem. 2002, 114, 1980-1804; Angew. Chem. Int. Ed. 2002, 41, 1900-1904; c) R. Gronheid, J. Hofkens, F. Kόhn, T.

Weil, E. Reuther, K. Mullen, F. C. De Schryver, J. Am. Chem. Soc. 2002,

124, 2418-2419.

[6] a) M. J. Fuller, M. R. Wasielewski, J. Phys. Chem. B 2001 , 105, 7216-

7219; b) M. J. Fuller, C. J. Walsh, Y. Zhao, M. R. Wasielewski, Chem. Mater. 2002, 14, 952-953.

[7] B. A. Gregg, R. A. Cormier, J. Am. Chem. Soc. 2001, 123, 7959-7960.

[8] a) J. Qu, C. Kohl, M. Pottek, K. Mullen, Angew. Chem. 2004, 116, 1554-

1557; Angew. Chem., Int. Ed. 2004, 43, 1528-1531 ; b) F. Yukruk, A. L.

Dogan, H. Canpinar, D. Guc, E. U. Akkaya, Org. Lett. 2005, 7, 2885-2887; c) C. Kohl, T. Weil, J. Qu, K. Mullen, Chem. Eur. J. 2004, 10, 5297-5310. d) C.

Jung, B. K. Mϋller, D. C. Lamb, F. Nolde, K. Mullen, C. Brauchle, J. Am.

Chem. Soc. 2006, 128, 5283-5291 ; e) A. Margineanu, J. Hofkens, M. Cotlet,

S. Habuchi, A. Stefan, J. Qu, C. Kohl, K. Mullen, J. Vercammen, Y.

Engelborghs, T. Gensch, F. C. De Schryver, J. Phys. Chem. B 2004, 108,

12242-12251. 8f). J. Qu ₁ N. G. Pschirer, D. Liu, A. Stefan, F. C. De Schryver,

K. Muellen, Chemistry - A European Journal 2004, 10, 528-537.

[9] a) G. Schnuφfeil, J. Stark, D. Wόhrle, Dyes Pigm. 1995, 27, 339-350; b) H. Langhals, German Patent DE-3703513, 1987; H. Langhals, W. Jona, F.

Einsiedl, S. Wohnlich, Adv. Mater. 1998, 10, 1022-1024.

[10] K. Harmanton, J. R. Hernandez, S. Becker, K. Mullen, J. Polym. Sci. A

2001, 39, 1572-1582.

[11] K. Matyjaszewski, J. Xia, Chem. Rev. 2001 , 101, 2921-2990. [12] M. Yin, K. Sorokina, C. Kuhlmann, C. Li ₁ G. Mihov, K. Koynov, H.

Luhmann, M. Klapper, K. Mullen, T. Weil, submitted to Chemistry & Biology,

2007.

[13] S. Angot, K. S. Murthy, D. Taton, Y. Gnanou, Macromolecules 1998, 31,

7218-7225. [14] a). V. Atanasov, V. Sinigersky, M. Klapper, K. Mullen, Macromolecules

2005, 38, 1672-1683 b) M. Yin, R. Bauer, M. Klapper, K. Mullen, Submitted to Macrom. Chem. Phys. 2007. c) M. Yin, K. Ding, J. Shen, R. Berger, K.

Mullen, Manuscript in preparation 2007.

[15] a) J. J. Hayes, J. C. Hansen, Curr Opin Genet Dev 2001 , 11, 124-129; b) P. J. Horn, C. L. Peterson, Science 2002, 297, 1824-1827; c) J. L.

Peterson, E. H. McConkey, J Biol Chem. 1976, 251, 548-554; d) K. Luger, A.

W. Mader, R. K. Richmond, D. F. Sargent, T. J. Richmond, Nature 1997,

389, 251-260; e) P. J. Verschure, I. Van Der Kraan, J. M. Enserink, M. J.

Mone, E. M. Manders, R. Van Driel, J Histochem Cytochem. 2002, 50, 1303- 1312; f) M. Dundr, T. Misteli, Biochem J. 2001 , 356, 297-310; g) H. Kimura,

P. R. Cook, J Cell Biol. 2001, 153, 1341-1353.

[16] S. Grimm, G. O. Pflugfelder, Science, 1996, 271, 1601-1603.

[17] a). H. Du, R.A. Fuh, J. Li, A. Corkan, J.S. Lindsey, Photochem.

Photobiol. 1998, 68, 141-142; b). T. Hard, P. Fan, and D. R. Kearns, Photochem. Photobiol. 1990, 51, 77-86.

[18]. A. H. Brand, N. Perrimon, Development 1993, 118, 401.