designing dendrimers for ocular drug delivery

lable at ScienceDirect

European Journal of Medicinal Chemistry 45 (2010) 326–334

Contents lists avai

European Journal of Medicinal Chemistry

journal homepage: ht tp: / /www.elsevier .com/locate/e jmech

Original article

Designing dendrimers for ocular drug delivery

Gregory Spataro a,b, François Malecaze c,**, Cedric-Olivier Turrin a,b, Vincent Soler c, Carine Duhayon a,b,Pierre-Paul Elena d,**, Jean-Pierre Majoral a,b,**, Anne-Marie Caminade a,b,*

a CNRS, LCC (Laboratoire de Chimie de Coordination), 205 route de Narbonne, F-31077 Toulouse, Franceb Universite de Toulouse, UPS, INPT; LCC, F-31077 Toulouse, Francec Hopital Purpan, Service d’Ophthalmologie, Place Dr Baylac, 31059 Toulouse Cedex, Franced Iris Pharma, Les Nertieres, 06610 La Gaude, France

a r t i c l e i n f o

Article history:Received 22 July 2009Received in revised form7 September 2009Accepted 8 October 2009Available online 14 October 2009

Keywords:DendrimersDrug deliveryIn vivo assaysCarboxylic acidsAmmonium saltsCarteolol

* Corresponding author. CNRS, LCC (Laboratoire deroute de Narbonne, F-31077 Toulouse, France. Fax: þ** Corresponding authors.

E-mail addresses: [email protected] (F. Miris-pharma.com (P.-P Elena), [email protected] (A.-M. Caminade).

0223-5234/$ – see front matter � 2009 Elsevier Masdoi:10.1016/j.ejmech.2009.10.017

a b s t r a c t

New series of phosphorus-containing dendrimers, having one quaternary ammonium salt as core andcarboxylic acid terminal groups have been synthesized from generation 0 (3 carboxylic acid terminalgroups) to generation 2 (12 carboxylic acid terminal groups). These dendrimers react with the neutralform of carteolol (an ocular anti-hypertensive drug used to treat glaucoma) to afford ion pair (saline)species. The solubility in water of these charged dendrimers depends on the generation considered:generation 0 (3 carteolol) is well soluble, whereas generation 1 (6 carteolol) and generation 2 (12carteolol) are poorly soluble. These dendrimers have been tested in vivo, as vehicle for ocular drugdelivery of carteolol to rabbits.

� 2009 Elsevier Masson SAS. All rights reserved.

1. Introduction

Dendrimers constitute nowadays a ubiquitous type of preciselydefined polymers [1], potentially usable in numerous applications.Their branched layered architectures displaying a high number ofcontrolled terminal groups is in particular very promising forbiomedical applications [2–6], and more precisely as drug carriers[7–9]. Physical entrapment in or of the dendritic skeleton(depending on the respective size and ratio of the drug and thedendrimer), or chemical conjugation are studied to use dendrimersas drug delivery vehicles. However, most of these studies concern invitro delivery, and only very few in vivo drug delivery studies usingdendrimers are available to date [10]. The first important point tobe verified for such purpose concerns the water solubility andthe biocompatibility [11]. PAMAM (polyamidoamine) [12–14]dendrimers are certainly the most widely used type of dendrimersfor biological purposes, including drug delivery. However, there

Chimie de Coordination), 20533 5 61 55 30 03.

alecaze), pierre-paul.elena@(J.-P. Majoral), caminade@lcc-

son SAS. All rights reserved.

exists other types of biocompatible dendrimers [15–17], and wehave shown in several occasions that the phosphorus-containingdendrimers we synthesize [18,19] are useful for biologicalpurposes, in particular as transfection agents [20–22], anti-prion[23–25] (including in vivo) and anti HIV [26,27] agents, againstAlzheimer [28] diseases, as imaging agents (including in vivo)[29–31], and for the activation and multiplication of humanmonocytes and innate immune Natural Killer (NK) cells [32–36].

The biocompatibility of drug delivery systems is particularlyrelevant when ocular delivery is concerned. Indeed, eyes havea quasi impermeable corneal surface epithelium, which necessi-tates a long residence time to increase the efficiency and thebioavailability of the instilled drug, to deliver it in the inner eyestructure. The corneal surface is also susceptible to bacterial, fungal,and viral infections and inflammations, as well as to mechanicalinjuries. Furthermore, lachrymal drainage poses problems to obtainsufficiently high therapeutic drug concentration inside the eye,notably when treating disorders such as diabetic retinopathy orglaucoma, to name as a few [37]. The most common method forimproving the bioavailability of a drug consists in increasing theviscosity by adding water-soluble polymers to enhance thebioadhesion of the solutions instilled [38]. However, such galenicalformulations may induce a temporarily disturbed vision, particu-larly for people suffering of ‘‘dry eye’’ disorder [39]. Thus,

mailto:[email protected]






www.sciencedirect.com/science/journal/02235234

http://www.elsevier.com/locate/ejmech

Scheme 1. Synthesis of neutral Carteolol 2.

G. Spataro et al. / European Journal of Medicinal Chemistry 45 (2010) 326–334 327

penetrating the ocular surface still presents a challenge forchemotherapy, and it appears tempting to use dendrimers insteadof polymers in the formulation. Indeed, dendrimers have multipleextremities, which may increase the bioadhesion, but they havealso very distinct properties compared to polymers, in particulara very low intrinsic viscosity [40,41], and a perfectly definedstructure. However, to the best of our knowledge, only one paperhas previously reported the use of dendrimers (PAMAM) for thein vivo ocular delivery of a drug (pilocarpine nitrate) [42].

In this paper we report the synthesis of a new series of phos-phorus-containing dendrimers, and attempts to use them for theocular delivery of a drug to treat glaucoma [43] and ocular hyper-tension, which are among the most frequent and severe oculardiseases, susceptible to degenerate to blindness [44]. Thesediseases require a very constraining life treatment, with instilla-tions all the 3 or 4 h. Increase of the residence time of the drugcould decrease the number of daily takings. We choose to test thewell-known drug carteolol [45], which is a b-blocker and an ocularantihypertensive agent.

The structure of the dendrimers has been engineered in order tofulfill two criteria. The first one concerns the interaction withcarteolol: carboxylic acid terminal groups should be suitable for anelectrostatic association with the amino group of carteolol (viahydrogen transfer from carboxylic acid to amine). The secondcriterion concerns the limitation of chemical entities in theformulation: benzalkonium chloride [PhCH2NMe2R]þCl� is softenused as preservative [46]; we thought that having a quaternaryammonium group as core of dendrimers could replace benzalko-nium chloride.

Scheme 2. Synthesis of dendrimers having a quaternary ammonium as core.

2. Results and discussion

As indicated in the introduction, the interaction betweena dendrimer and the drug to be delivered can be either covalent ornot. The first case is frequently inefficient because the grafting ofthe drug is often chemically difficult, and may totally modify theproperties of the drug [47], unless a prodrug is grafted. On the otherhand, a purely physical entrapment might be not enough strong. Analternative option is to enhance the stability of the drug/vehiclesystem through the formation of the so-called catanionic systems[48]. The latter are mixtures of oppositely charged surfactants orhydrotrope entities, the resulting ion pair being stabilised bylipophilic interactions among others. Such mixtures generatea growing interest, particularly for pharmaceutical applications[49,50]. They are frequently obtained by mixing a positivelycharged entity such as an ammonium salt with a negativelycharged entity such as a carboxylic acid salt. However, in this casethe catanionic mixture is obtained together with a salt, frequentlysodium chloride. A more elegant method consists in using an amineand a carboxylic acid to generate the catanionic mixture withoutany by-product [51]. We have already applied successfully thisprocedure to carboxylic acid terminated phosphorus-containingdendrimers and long chained amino sugars, leading to the firstcatanionic dendrimers [26,27]. These results inspired us to adaptthis strategy to carteolol, which bears a free amine group, in orderto form ion pair (saline) dendritic systems. The location of counter-

ions in solutions of charged dendrimers is still a matter of debate,but recent simulations suggest that charged dendrimers havea high local counter-ions concentration, necessitated by chargeneutrality [52].

2.1. Syntheses

The presence of a secondary amine in carteolol makes itperfectly suitable for ion pair interactions with dendrimers endedby carboxylic acids. Neutral carteolol 2 is obtained by reacting thehydrochloride derivative 1 (the commercially available form of thedrug) with K2CO3 (Scheme 1).

As indicated in the introduction, in order to limit the number ofchemical entities in the formulation, we decided to design a newfamily of dendrimers, possessing an analogue of benzalkoniumchloride as core and carboxylic acid terminal groups for theinteraction with carteolol. We chose tris(2-chloroethyl)aminehydrochloride 3 as analogue of benzalkonium chloride, and suitableas core of the dendrimer. In order to be able to apply the method ofsynthesis of dendrimers that we classically use [18,19], we need tograft benzaldehyde functions to 3. The reaction of 3 equivalents ofhydroxybenzaldehyde 4 with NaOH in ethanol, followed by thereaction with 3 affords compound 5-G0, issued from boththe grafting of hydroxybenzaldehyde and the neutralization of theammonium salt. The next step for the classical synthesis of ourdendrimers [18,19] should be the condensation with the phos-phorhydrazide 7 (H2NNMeP(S)Cl2) but the presence of an amine atthe core of the dendrimer could induce unwanted side reactions,such as a partial oligomerization of compound 7. To avoid thisproblem, we decided to alkylate the tertiary amine, using methyltriflate to afford compound 6-G0 (Scheme 2). The alkylation isshown by the presence of a new singlet (methyl group) in 1H and13C NMR at 3.38 and 51.4 ppm, respectively.

Compound 6-G0 constitutes the real core of the new series ofdendrimers that we will now describe. The next step is thecondensation reaction with the phosphorhydrazide 7, affording thefirst generation dendrimer 8-G1, as shown by the total disappear-ance of the signals corresponding to the aldehydes by 1H and 13CNMR. Substitution reactions of chlorine atoms by hydrox-ybenzaldehyde 4 give the other first generation dendrimer 6-G1.This reaction is monitored by 31P NMR, which displays first theappearance of an intermediate singlet at d¼ 70 ppm,

Scheme 3. Synthesis of dendrimers terminated by tert-butyl benzoate.

G. Spataro et al. / European Journal of Medicinal Chemistry 45 (2010) 326–334328

corresponding to the substitution of one Cl on the P(S)Cl2 terminalgroups, together with the decrease of the signal at 66.9 ppm cor-responding to P(S)Cl2. When the reaction has gone to completion,both signals have totally disappeared, on behalf of a unique singletat 63.6 ppm. The last step of the synthesis is again the condensationreaction with the phosphorhydrazide 7, to afford the secondgeneration dendrimer 8-G2 (Scheme 2). As previously for 8-G1, thecompletion of the reaction giving dendrimer 8-G2 is shown by thedisappearance of the singlets corresponding to the aldehydes byboth 1H and 13C NMR.

In order to functionalize the surface of the dendrimer bycarboxylic acids, we tried first to transform the aldehydes ofcompound 6-G0 to cinnamic acids, using a Knoevenagel–Doebnerreaction with malonic acid, pyridine and piperidine, that we havesuccessfully used previously [53]. However, in the present case, wewere not able to isolate the expected compound, thus we decidedto try another strategy. The best way to functionalize our den-drimers is generally by reacting a phenol on the P(S)Cl2 terminalgroups. 4-hydroxybenzoic acid is not a suitable candidate, since thepresence of a carboxylic acid is incompatible with the P(S)Cl2functions, but the acid function can be temporarily protected. Wetried several ways to obtain tert-butyl-4-hydroxybenzoate from4-hydroxybenzoic acid. Despite a moderate yield (55%), the mostsuitable method (easier work-up) consists in coupling tert-butanolwith 4-hydroxybenzoic acid in the presence of DCC

Fig. 1. ORTEP drawing of compound 9.

(dicyclohexylcarbodiimide) and DBU (1,8-dia-zabicyclo[5.4.0]undec-7-ene) (Scheme 3). Single crystals suitablefor X-ray diffraction were obtained for compound 9; the ORTEPdrawing is shown in Fig. 1.

Compound 9 is reacted with tris(2-chloroethyl)amine hydro-chloride 3 in the presence of K2CO3 as base. Compound 10-G0 isissued from the reaction of 3 equivalents of 9 and the neutralizationof the ammonium salt. In this case, the reaction is monitored bythin layer chromatography. Compound 9 is then reacted with thefirst and second generations of dendrimers 8-Gn, to afforddendrimers 11-G1 and 11-G2, respectively (Scheme 3). In both casesthe synthesis is monitored by 31P NMR, as was the reaction with4-hydroxybenzaldehyde previously.

Obtaining carboxylic acid terminal groups necessitates thedeprotection of the tert-butyl benzoate in all cases, but an addi-tional reaction is needed for 10-G0 to alkylate the central nitrogen.For this purpose, compound 10-G0 is reacted with methyltriflate. Inthese conditions, the expected alkylation occurs, but it is alsoaccompanied by the deprotection of the tert-butyl benzoate, asshown by the disappearance of the signals corresponding to thetert-butyl groups by both 1H and 13C NMR. This unexpected reac-tion is presumably induced by catalytic amounts of triflic acid inmethyl triflate. The triflate anion inducing a poor solubility in waterand being not bio-compatible, it is replaced by chloride using theexchange ion resin Dowex 1X8, to afford the small dendrimer 12-G0

(Scheme 4). The absence of any signal by 19F NMR confirms the totalexchange.

In the case of the first and second generation dendrimers 11-Gn,a standard procedure with trifluoro acetic acid dichloromethanesolution (20%) is used for the deprotection of the tert-butyl groups[54], and after chromatography through the exchange ion resinDowex 1X8, dendrimers 12-G1 and 12-G2 are isolated (Scheme 4).In both cases, the absence of signals corresponding to the tert-butylgroup by 1H and 13C NMR, as well as the disappearance of any signalby 19F NMR confirm the obtaining of the expected compounds. Onthe other hand, all the expected signals are observed by 31P, 1H and13C NMR, confirming that the dendritic skeleton did not suffered,despite the very hard conditions used.

Finally, dendrimers 12-Gn are reacted with carteolol 2 to affordthe saline (catanionic) species 13-Gn (n¼ 0, 1, 2) (Scheme 5).Compound 13-G0 is relatively well soluble in water (its solubilitycorresponds to ca 1% in weight of carteolol in water), whereascompounds 13-G1 and 13-G2 are poorly soluble (ca 0.05% in weightof carteolol in water for both compounds). Furthermore, in these

Scheme 4. Deprotection of the terminal groups of dendrimers.


conditions only signals corresponding to the most exterior layer ofthe dendrimers and to carteolol can be detected by NMR. We havepreviously described such phenomenon for another type of water-soluble dendrimers, and shown that addition of a good solvent forthe inner layers allows to recover all the expected signals [55].Addition of acetonitrile to water allows to distinguish all the signalsfor 13-G0 and 13-G1, but not for 13-G2, which remains poorlysoluble even in these conditions. Creation of the saline species13-G0 induces dramatic changes in 1H and 13C NMR spectra for thecarteolol moieties (1H: DdCH2N¼þ0.3, and DdCHCH2N¼þ0.2 ppm; 13C: Dd(CH3)3C¼þ7.9, and Dd(CH3)3C¼�3.4 ppm) aswell as for the benzoic acid moieties (13C: DdCO2¼þ7.1,DdC0

4¼þ5.1) on going from carteolol 2 and 12-G0 to 13-G0 (seeFig. 3 for the numbering used). Analogous information is obtainedfor 13-G1 concerning the carteolol part, whereas the benzoic acidpart behaves slightly differently, because it is linked to a phos-phorus instead of a carbon atom for the generation zero (13C:DdC0

2¼þ4.0 ppm on going from carteolol 2 and 12-G1 to 13-G1). Inall cases, including for 13-G2, IR spectroscopy shows the disap-pearance of the strong absorption band located at 1688 cm�1 cor-responding to the benzoic acid and the appearance of two bands at

Scheme 5. Synthesis of the catanionic de

1550 and 1380 cm�1 for the carboxylate group (nCOO�

asym andnCOO�

sym respectively). The carteolol moieties are unambiguouslyidentified by the appearance of absorption bands at 1668 cm�1 and1336 cm�1, corresponding to the carbonyl and the lactam (C–N)vibrations, respectively. The collected spectroscopic informationare in good agreement with previous studies on formulations ofcarteolol with acid functionalized polyacrylates [56], and areindicative of the formation of dendrimer–carteolol complexesthrough acid–base proton exchange.

2.2. Drug delivery experiments

Saline dendrimers 13-Gn were synthesized to test their abilityfor improving the delivery of carteolol in eyes. For this purpose,solutions of these dendrimers are elaborated in milliQ water.The carteolol concentration in the solutions of carteolol and ofdendrimers 13-Gn is determined by HPLC–MS, and is given inTable 1 (initial concentration in carteolol). We intended to reacha concentration of about 10 g/L; such value is easily obtained withcarteolol alone and with dendrimer 13-G0, but not with dendrimers13-G1 and 13-G2, both being very poorly soluble in water. However,

ndrimers by reaction with carteolol.

Table 1Carteolol concentration in the initial solution and in the aqueous humour of rabbits’eyes, measured by HPLC–MS.

Entry Compound Initial[carteolol](g/L)a

Time(h)

Measured mean[carteolol] in aq.humour (ng/mL)a,b

Corrected mean[carteolol] in aq.humour (ng/mL)c

1 1 11.69 1 4352 37222 1 11.69 4 858 7743 1 11.75 6 239 2034 1 11.75 8 141 1205 13-G0 12.76 1 2922 22906 13-G0 12.76 4 898 7047 13-G0 13.14 6 760 5788 13-G0 13.14 8 102 789 13-G0 2� 13.14d 8 484 184e

10 13-G2 0.35 1 267 (7628)f

11 13-G2 0.35 4 18 (514)f

a Measured by HPLC–MS on the filtered solutions used for instillation.b This value is the mean value of four experiments.c Value corrected by considering 10 g/L as initial concentration in all cases (the

variability can be seen in Fig. 2).d Two drops were instilled in the eyes of this rabbit (the second drop was instilled

five minutes after the first one).e Corrected by considering that the initial concentration is 26.28 mg/mL.f Values to be used with caution, due to the low concentration of the initial

solution.

Fig. 3. Numbering used for NMR assignments.


we decided to try a few experiments with 13-G2, which possessestwice the number of carteolol compared to 13-G1. Solutions inmilliQ water of carteolol or of the catanionic dendrimers are filteredbefore instillation in the eyes of albino rabbits (one drop (50 mL) ineach eye). No trace of irritation is detected whatever the solutionused. 100 mL of aqueous humour are taken from the eyes ofanaesthetized rabbits after 1 or 4 h in the first series of experi-ments, or after 6 or 8 h for the other series of experiments; onlycarteolol and dendrimer 13-G0 have been tested in this last series ofexperiments. The quantity of carteolol in all the taken aqueoushumour samples is given in Table 1 (column 5). However, asindicated previously, the initial concentration in the drop instilledwas different; thus we decided to apply a standardized value of10 g/L for the initial concentration, and to modify the resultsaccordingly. This method allows a fair comparison of the resultsobtained with compounds 1 and 13-G0, as illustrated in Table 1 andFig. 2. On the other hand, the modified (excellent) results obtainedwith 13-G2, must be taken in consideration with caution, sincea concentration of 10 g/L is impossible to attain with thiscompound. One hour after the instillation, the quantity of carteololdetected in aqueous humour is higher in the case of carteolol alone(1) than in the case of its association with the dendrimer (13-G0),but after 4 h the results are equivalent for both series of experi-ments. In order to detect the influence of the quantity of carteololinitially instilled, we also added two drops of the solution of 13-G0

Fig. 2. Quantity of carteolol detected in aqueous humour (using the corrected data,and showing the uncertainty).

in each eye of the same rabbit, with an interval of 5 min. Anincrease of the quantity of carteolol measured in the aqueoushumour is observed (compare entries 8 and 9 in Table 1), largerthan expected even when taking into account the increasedquantity instilled.

3. Conclusions

We have synthesized several new dendritic compoundspossessing one ammonium salt as core, and having carboxylic acidterminal groups. These carboxylic acid groups are able to interactwith the amino function of carteolol, to afford catanionic (saline)species. These sophisticated compounds were elaborated with theaim of combining in a single entity an analogue of the preservative(benzalkonium chloride) classically used in commercial solutions ofcarteolol, and a potential drug delivery vehicle. The smallest salinespecies is soluble in water, but the largest ones are only poorlysoluble (in water, and even in organic solvents for the secondgeneration). On one side, this fact is deceptive, but on the otherside, it confirms that the expected association occurs, and is strongenough to remain in water. All these compounds (from generation0 to generation 2) in solution in milliQ water have been instilled inthe eyes of rabbits. No irritation is observed, whatever the cata-nionic dendrimer used and even after several hours. Measurementsof the quantity of carteolol having penetrated inside the aqueoushumour shows practically no difference between carteolol aloneand carteolol entrapped with the generation zero. Due to the verylow solubility of the second generation, the quantity of carteololinstilled is low, but the quantity of carteolol that penetrates insidethe eyes is larger than expected, when compared with carteololalone (2.5 times larger). Thus, even if the solubility is a realproblem, these pharmacodynamic observations highlight thebiocompatibility and the potential usefulness of our type ofapproach for drug delivery.

4. Experimental

4.1. Chemistry

All manipulations were carried out with standard high-vacuumand dry-argon techniques. Chemicals were purchased from Sigma–Aldrich and used without further purification; solvents were driedand distilled by routine procedures. 1H, 13C, and 31P NMR spectrawere recorded at 25 �C with Bruker AC 200, ARX 250, AV 300, DPX300, AMX 400 or AV500 spectrometers. References for NMRchemical shifts are 85% H3PO4 for 31P NMR and SiMe4 for 1H and 13CNMR. The attribution of 13C NMR signals has been done using Jmod,two-dimensional 1H–13C HSQC, 1H–13C HMBC, and 1H–31P HMQC,Broad Band or CW 31P decoupling experiments when necessary.Mass spectrometry was recorded on a Finniganmat TSQ 7000.FT–IR spectra were recorded on a Perkin Elmer PE GX 2000. Thenumbering used for NMR assignment is depicted on Fig. 3.


4.1.1. Synthesis of compound 2Aliquots of neutral carteolol 2 were obtained from carteolol

hydrochloride 1 by the following procedure: 0.50 g of carteololhydrochloride 1 was dissolved in a saturated solution of K2CO3 inwater. This mixture was extracted 3 times with 50 mL of chloro-form in a separation funnel. Then the organic phase was dried overMgSO4, filtered and evaporated to dryness. Carteolol 2 was isolatedas a white powder in 95% yield (0.42 g). 1H NMR (CD3CN/D2O,200 MHz): d¼ 1.67 (s, 9H, C(CH3)3), 2.92 (t, 3JHH¼ 6.0 Hz, 2H, CH2–CH2–C]O), 3.14 (m, 2H, NH–CH2), 3.32 (t, 3JHH¼ 6.0 Hz, 2H, CH2–C]O), 4.38 (br s, 3H, O–CH2, CHOH), 6.94 (d, 2JHH¼ 8.0 Hz, 1H, CfH),7.09 (d, 2JHH¼ 8.0 Hz, 1H, CdH), 7.56 (t, 3JHH¼ 8.0 Hz, 1H, CeH) ppm.13C–{1H} NMR (CD3CN/D2O, 63 MHz): d¼ 19.90 (s, CH2–CH2–C]O),29.10 (s, C(CH3)3), 32.90 (s, CH2–C]O), 44.80 (s, N–CH2), 50.20 (s,C(CH3)3), 67.21 (s, CHOH), 72.75 (s, O–CH2), 106.20 (s, Cf), 107.50 (s,Cd), 107.80 (s, Cc), 129.20 (s, Ce), 140.00 (s, Cb), 155.95 (s, Ca), 171.70(s, CONH) ppm. Anal. [C16H24NO6] C, H, N. Calcd: C, 58.88; H, 7.41; N,4.29. Found: C, 58.75; H, 7.38; N, 4.25.

4.1.2. Synthesis of dendrimer 5-G0

We modified the procedure for the synthesis and purification ofthis compound [57]. Hydroxybenzaldehyde 4 (1.83 g, 15.00 mmol)and NaOH (0.80 g, 20.00 mmol) were vigorously stirred for 1 h inabsolute ethanol (250 mL). Then tris(2-chloroethyl)amine hydro-chloride 3 (1.21 g, 5.00 mmol) was added and the mixture washeated to reflux for 4 h. After drying under vacuum, the residuewas dissolved in dichloromethane (250 mL). The resulting solutionwas washed twice with a saturated solution of potassium carbonatein water. The organic phase was dried over MgSO4, filtered andevaporated to dryness. The resulting solid was crystallized twice in10 mL of ethanol to afford dendrimer 5-G0 as a white powder in 79%yield (1.85 g). 1H NMR (CDCl3, 200 MHz): d¼ 3.20 (t, 3JHH¼ 4.0 Hz,6H, CH2–O), 4.17 (t, 3JHH¼ 4.0 Hz, 6H, N–CH2), 6.95 (d, 2JHH¼ 8.8 Hz,6H, C0

2H), 7.89 (d, 2JHH¼ 8.8 Hz, 6H, C03H), 9.85 (s, 3H, CHO) ppm.

13C–{1H} NMR (CDCl3, 50 MHz): d¼ 54.32 (s, CH2–O), 67.33 (s,N–CH2), 114.73 (s, C0

2), 130.03 (s, C04), 131.97 (s, C0

3), 163.64 (s, C01),

190.74 (s, CHO) ppm. Anal. [C27H27NO6] C, H, N. Calcd: C, 70.27; H,5.90; N, 3.04. Found: C, 70.34; H, 5.95; N, 2.97.


Methyltriflate (104 mL, 0.95 mmol) was added at 0 �C to den-drimer 5-G0 (200 mg, 0.43 mmol) dissolved in 20 mL of dichloro-methane. The resulting mixture was stirred for 18 h at roomtemperature then the solution was evaporated to dryness. Theresidue was washed twice with ether. In order to eliminate thesmall amount of protonated amine issued from the reaction withtriflic acid (contaminant of methyltriflate), the following procedurewas used. Dichloromethane (5 mL) and a few mg of Cs2CO3 wereadded, and the resulting mixture was vigorously stirred for a fewminutes. This mixture was filtered and the precipitate was recov-ered. Hot acetonitrile (5 mL) was added to this precipitate, and theresulting mixture was filtered. The solution was kept then evapo-rated to dryness, to afford dendrimer 6-G0 as a pale yellow powderin 95% yield (225 mg). 1H NMR (CD3CN, 200 MHz): d¼ 3.38 (s, 3H,Nþ–CH3), 4.06 (br s, 6H, CH2–O), 4.65 (br s, 6H, N–CH2), 7.18 (d,2JHH¼ 8.8 Hz, 6H, C0

2H), 7.93 (d, 2JHH¼ 8.8 Hz, 6H, C03H), 9.93 (s, 3H,

CHO) ppm. 13C–{1H} NMR (CD3CN, 50 MHz): d¼ 51.45 (s, Nþ–CH3),62.79 (s, CH2–O), 63.39 (s, N–CH2), 115.95 (s, C0

2), 131.87 (s, C04),

132.62 (s, C03), 162.95 (s, C0

1), 191.87 (s, CHO) ppm. 19F–{1H} NMR(CD3CN, 188 MHz): d¼�3.27 (s) ppm. Anal. [C29H30F3NO9S] C, H, N.Calcd: C, 55.68; H, 4.83; N, 2.24. Found: C, 55.79; H, 4.87; N, 2.18.


A 0.18 M solution of H2NNMeP(S)Cl2 7 in CHCl3 (6.9 mL,1.23 mmol) was added to a solution of dendrimer 6-G0 (200 mg,

0.32 mmol) in 5 mL of acetonitrile. The resulting mixture wasstirred for 20 min at room temperature then concentrated underreduced pressure. 15 mL of ether then 5 mL of pentane were addedto precipitate the dendrimer. The powder was recovered by filtra-tion through canula then dissolved in 3 mL of THF, and precipitatedwith 20 mL of ether/pentane (50:50). This procedure was repeatedtwice, to afford dendrimer 8-G1 as a white powder in 92% yield(327 mg). 31P-{1H} NMR (CD3CN, 80 MHz): d¼ 66.92 (s) ppm. 1HNMR (CD3CN, 200 MHz): d¼ 3.37 (s, 3H, Nþ–CH3), 3.49 (d,3JHP¼ 14.7 Hz, 9H, P–N–CH3), 4.04 (br s, 6H, N–CH2), 4.59 (br s, 6H,CH2–O), 7.09 (d, 3JHH¼ 8.8 Hz, 6H, C0

2H), 7.88 (d, 3JHH¼ 8.8 Hz, 6H,C0

3H), 7.92 (d, 4JHP¼ 2.8 Hz, 3H, CH]N) ppm. 13C–{1H} NMR (CD3CN,50 MHz): d¼ 32.61 (d, 2JCP¼ 13.5 Hz, P–N–CH3), 51.43 (s, Nþ–CH3),62.58 (s, CH2–O), 63.48 (s, N–CH2), 115.99 (s, C0

2), 129.10 (s, C04),

129.80 (s, C03), 144.10 (d, 3JCP¼ 18.9 Hz, CH]N), 159.81 (s, C0

1) ppm.19F–{1H} NMR (CD3CN, 188 MHz): d¼�3.21 (s) ppm. Anal.[C32H39Cl6F3N7O6P3S4] C, H, N. Calcd: C, 34.67; H, 3.55; N, 8.84.Found: C, 34.81; H, 3.61; N, 8.77.


4-hydroxybenzaldehyde 4 (191 mg, 1.57 mmol) was added toa solution of dendrimer 8-G1 (300 mg, 0.26 mmol) in THF (5 mL) inthe presence of Cs2CO3 (1.05 g, 3.12 mmol). The resulting mixturewas stirred overnight at room temperature, then diluted with THF(10 mL), filtered, centrifuged and evaporated to dryness. Theresidue was dissolved in 3 mL of THF then precipitated with 20 mLof ether/pentane (50:50). This procedure was repeated three times,then the product was purified by column chromatography oversilica (AcOEt/Pentane, 50/50, Rf¼ 0.3). Dendrimer 6-G1 was iso-lated after evaporation to dryness as a white powder in 61% yield(270 mg). 31P-{1H} NMR (CDCl3, 80 MHz): d¼ 63.58 (s) ppm. 1HNMR (CDCl3, 200 MHz): d¼ 3.37 (d, 3JHP¼ 10.9 Hz, 9H, P–NCH3),3.53 (s, 3H, Nþ–CH3), 4.22 (br s, 6H, CH2–O), 4.62 (br s, 6H, N–CH2),6.93 (d, 3JHH¼ 8.8 Hz, 6H, C0

2H), 7.37 (d, 3JHH¼ 8.8 Hz, 12H, C12H),

7.58 (d, 3JHH¼ 8.8 Hz, 6H, C03H), 7.60 (br s, 3H, CH]N), 7.85 (d,

3JHH¼ 8.8 Hz, 12H, C13H), 9.93 (s, 6H, CHO) ppm. 13C–{1H} NMR

(CD3CN, 50 MHz): d¼ 33.39 (d, 2JCP¼ 13.0 Hz, P–NCH3), 51.40 (s,Nþ–CH3), 62.56 (s, CH2–O), 63.48 (s, N–CH2), 115.84 (s, C0

2), 122.75(s, C1

2), 129.42 (s, C03), 129.52 (s, C1

4), 132.26 (s, C04), 134.79 (s, C1

3),142.35 (d, 3JCP¼ 11.4 Hz, CH]N), 155.90 (s, C0

1), 159.42 (br s, C11),

192.36 (s, CHO) ppm. 19F–{1H} NMR (CDCl3, 188 MHz): d¼�2.38 (s)ppm. Anal. [C74H69F3N7O18P3S4] C, H, N. Calcd: C, 54.78; H, 4.29; N,6.04. Found: C, 54.84; H, 4.33; N, 5.98.


A 0.18 M solution of H2NNMeP(S)Cl2 7 in CHCl3 (6.7 mL,1.20 mmol) was added to a solution of dendrimer 6-G1 (250 mg,0.155 mmol) in 5 mL of THF. The resulting mixture was stirred for30 min at room temperature then concentrated under reducedpressure. 15 mL of ether then 5 mL of pentane were added toprecipitate the dendrimer. The powder was recovered by filtrationthrough canula then dissolved in 3 mL of THF, and precipitated with20 mL of ether/pentane (50:50). This procedure was repeatedtwice, to afford dendrimer 8-G2 as a white powder in 82% yield(322 mg). 31P-{1H} NMR (CD3CN, 80 MHz): d¼ 65.36 (s, P1), 66.16 (s,P2) ppm. 1H NMR (CD3CN, 200 MHz): d¼ 3.18 (s, 3H, Nþ–CH3), 3.19(d, 3JHP¼ 13.5 Hz, 18H, P2–NCH3), 3.37 (d, 3JHP¼ 13.5 Hz, 9H, P1–NCH3), 3.75 (br s, 6H, CH2–O), 4.28 (br s, 6H, N–CH2), 6.75 (d,3JHH¼ 8.3 Hz, 6H, C0

2H), 7.01 (d, 3JHH¼ 8.2 Hz, 12H, C12H), 7.41 (d,

3JHH¼ 8.3 Hz, 6H, C03H), 7.48 (d, 3JHH¼ 8.2 Hz, 12H, C1

3H), 7.60 (br s,9H, CH]N) ppm. 13C–{1H} NMR (CD3CN, 50 MHz): d¼ 31.61 (d,2JCP¼ 13.0 Hz, P2–NCH3), 32.58 (d, 2JCP¼ 12.0 Hz, P1–N–CH3), 50.41(s, Nþ–CH3), 61.46 (s, CH2–O), 62.51 (s, N–CH2), 114.73 (s, C0

2), 121.50(s, C1

2), 128.41 (s, C03), 128.69 (s, C1

4), 128.75 (s, C04), 131.47 (s, C1

3),141.93 (d, 3JCP¼ 11.8 Hz, CH]N2), 142.47 (d, 3JCP¼ 11.8 Hz, CH]N1),


151.68 (s, C01), 158.09 (s, C1

1) ppm. 19F–{1H} NMR (CD3CN, 188 MHz):d¼�3.20 (s) ppm. Anal. [C80H87Cl12F3N19O12P9S10] C, H, N. Calcd: C,37.12; H, 3.39; N, 10.28. Found: C, 37.26; H, 3.45; N, 10.19.

4.1.7. Synthesis of compound 9Hydroxybenzoıc acid (3.00 g, 22.00 mmol), tert-butanol (34 mL,

35.50 mmol), DBU (0.37 mL, 2.4 mmol), and DCC (5.00 g, 24 mmol)were mixed in dichloromethane (80 mL) and vigorously stirred for18 h. After evaporation to dryness, 100 mL of dichloromethanewere added to the residue, and the resulting heterogeneous solu-tion was filtered. The solution was washed twice with a saturatedsolution of potassium carbonate in water, then with a saturatedsolution of NaCl in water. The organic phase was dried over MgSO4,filtered then evaporated to dryness. The residue was purified bycolumn chromatography over silica (CH2Cl2/AcOEt, 98:2, Rf¼ 0.3).Single crystals suitable for X ray diffraction were grown from thissolution after one night at room temperature. Then the solutionwas evaporated to dryness to afford compound 9 as a white powderin 55% yield (2.20 g). 1H NMR (CDCl3, 200 MHz): d¼ 1.58 (s, 9H,C–(CH3)3), 5.98 (br s, 1H, OH), 6.83 (d, 3JHH¼ 8.8 Hz, 2H, C2H), 7.91(d, 3JHH¼ 8.8 Hz, 2H, C3H) ppm. 13C–{1H} NMR (CDCl3, 50 MHz):d¼ 28.23 (s, C–(CH3)3), 81.17 (s, C–(CH3)3), 115.11 (s, C2), 123.81 (s,C4), 131.71 (s, C3), 160.16 (s, C1), 166.49 (s, COO) ppm. MS (CI, NH3):m/z¼ 195 [MH]þ. Anal. [C11H14O3] C, H. Calcd: C, 68.02; H, 7.26.Found: C, 68.05; H, 7.28.


Compound 9 (388 mg, 2.0 mmol) and potassium carbonate(415 mg, 3.0 mmol) were vigorously stirred for 1 h in 15 mL ofanhydrous DMF. Then tris(2-chloroethyl)amine hydrochloride 3(120.5 mg, 0.5 mmol) was added and the mixture was heated at85 �C for 4 h. The mixture was diluted with 100 mL of ethyl acetate,and filtered. The solution was evaporated to dryness then theresidue was dissolved in toluene and evaporated to dryness again.This sequence was repeated several times. The residue was thendissolved in 5 mL of dichloromethane. The resulting solution waswashed twice with a saturated solution of potassium carbonate inwater, then with a saturated solution of NaCl in water. The organicphase was dried over MgSO4, filtered and evaporated to dryness.The residue was purified by column chromatography over silica(CHCl3/CH3CN, 85:15, Rf¼ 0.85). Dendrimer 10-G0 was isolated asvery viscous translucent oil in 88% yield (300 mg). 1H NMR (CDCl3,250 MHz): d¼ 1.58 (s, 27H, C–(CH3)3), 3.18 (t, 3JHH¼ 5.5 Hz, 6H,CH2–O), 4.14 (t, 3JHH¼ 5.5 Hz, 6H, N–CH2), 6.86 (d, 2JHH¼ 8.8 Hz, 6H,C0

2H), 7.91 (d, 2JHH¼ 8.8 Hz, 6H, C03H) ppm. 13C–{1H} NMR (CDCl3,

50 MHz): d¼ 28.24 (s, C–(CH3)3), 54.37 (s, CH2–O), 66.99 (s, CH2–N),80.55 (s, C–(CH3)3), 113.86 (s, C0

2), 124.60 (s, C04), 131.37 (s, C0

3), 162.03(s, C0

1), 165.53 (s, COO). MS (CI, NH3): m/z¼ 678 [MþH]þ. Anal.[C39H51NO9] C, H, N. Calcd: C, 69.11; H, 7.58; N, 2.07. Found: C,69.20; H, 7.62; N, 2.01.


Compound 9 (153 mg, 0.73 mmol) was added to a solution ofdendrimer 8-G1 (150 mg, 0.13 mmol) in 5 mL of THF in the presenceof Cs2CO3 (503 mg, 1.42 mmol). The resulting mixture was stirredovernight at room temperature. The solution was diluted with THF(15 mL) filtered, centrifuged then evaporated to dryness. Theresidue was dissolved in 3 mL of THF then precipitated with 20 mLof ether/pentane (50:50). This procedure was repeated three times,then the product was purified by column chromatography oversilica (CHCl3/CH3CN, 75:25, Rf¼ 0.2). Dendrimer 11-G1 was isolatedafter evaporation to dryness as a white powder in 86% yield(228 mg). 31P-{1H} NMR (CD3CN, 80 MHz): d¼ 65.16 (s) ppm. 1HNMR (CD3CN, 200 MHz): d¼ 1.54 (s, 54H, C–(CH3)3), 3.35 (s, 3H,Nþ–CH3), 3.37 (d, 3JHP¼ 11.0 Hz, 9H, P–NCH3), 4.01 (br s, 6H, CH2–

O), 4.55 (br s, 6H, N–CH2), 7.03 (d, 3JHH¼ 8.8 Hz, 6H, C02H), 7.29

(d, 3JHH¼ 8.8 Hz, 12H, C12H), 7.65 (d, 3JHH¼ 8.8 Hz, 6H, C0

3H), 7.80 (brs, 3H, CH]N), 7.96 (d, 3JHH¼ 8.8 Hz, 12H, C1

3H) ppm. 13C–{1H} NMR(CD3CN, 50 MHz): d¼ 28.19 (s, C–(CH3)3), 32.45 (d, 2JCP¼ 13.5 Hz,P–N–CH3), 51.39 (s, Nþ–CH3), 62.57 (s, CH2–O), 63.51 (s, N–CH2),81.95 (s, C–(CH3)3), 115.85 (s, C0

2), 121.94 (s, C12), 129.37 (s, C0

3), 129.66(s, C1

4), 130.18 (s, C04), 131.90 (s, C1

3), 141.86 (d, 3JCP¼ 13.8 Hz, CH]N),154.81 (s, C1

1), 159.33 (br s, C01), 165.41 (s, COO) ppm. 19F–{1H} NMR

(CD3CN,188 MHz): d¼�3.22 (s) ppm. Anal. [C98H117F3N7O24P3S4] C,H, N. Calcd: C, 57.27; H, 5.74; N, 4.77. Found: C, 57.38; H, 5.80; N,4.71.


Compound 9 (138 mg, 0.71 mmol) was added to a solution ofdendrimer 8-G2 (150 mg, 0.06 mmol) in 5 mL of THF in the pres-ence of Cs2CO3 (503 mg, 1.42 mmol). The resulting mixture wasstirred overnight at room temperature. The solution was dilutedwith THF (15 mL) filtered, centrifuged then evaporated to dryness.The residue was dissolved in 3 mL of THF then precipitated with20 mL of ether/pentane (50:50). This procedure was repeated threetimes, then the product was purified by column chromatographyover silica (CHCl3/CH3CN, 2:1, Rf¼ 0.2). Dendrimer 11-G2 wasisolated after evaporation to dryness as a white powder in 76% yield(195 mg). 31P-{1H} NMR (CDCl3, 80 MHz): d¼ 64.86 (s, P2), 65.61 (s,P1) ppm. 1H NMR (CDCl3, 200 MHz): d¼ 1.54 (s, 108H, C–(CH3)3),3.33 (d, 3JHP¼ 10.6 Hz, 18H, P2–NCH3), 3.36 (d, 3JHP¼ 10.6 Hz, 9H,P1–NCH3), 3,57 (br s, 3H, Nþ–CH3), 4.31 (br s, 6H, CH2–O), 4.63 (br s,6H, N–CH2), 6.95 (d, 3JHH¼ 8.2 Hz, 6H, C0

2H), 7.22 (d, 3JHH¼ 8.3 Hz,24H, C2

2H), 7.26 (d, 3JHH¼ 7.8 Hz, 12H, C12H), 7.65 (m, 27H, C0

3H, C13H,

CH]N), 7.94 (d, 3JHH¼ 8.3 Hz, 24H, C23H) ppm. 13C–{1H} NMR

(CDCl3, 50 MHz): d¼ 28.20 (s, C–(CH3)3), 32.50 (d, 2JCP¼ 13.0 Hz,P2–NCH3, P1–NCH3), 49.20 (s, Nþ–CH3), 61.50 (s, CH2–O), 63.10(s, N–CH2), 80.70 (s, C–(CH3)3), 114.70 (s, C0

2), 121.20 (s, C22), 122.10

(s, C12), 128.50 (s, C1

3), 128.70 (s, C03), 129.40 (s, C1

4), 129.60 (s, C04),

131.20 (s, C23), 132.30 (s, C2

4), 139.30 (br d, 3JCP¼ 11.5 Hz, CH]N),151.60 (s, C1

1), 153.90 (br s, C21) 157.90 (br s, C0

1), 165.10 (s, COO) ppm.19F–{1H} NMR (CDCl3, 188 MHz): d¼�2.37 (s) ppm. Anal.[C212H243F3N19O48P9S10] C, H, N. Calcd: C, 56.82; H, 5.46; N, 5.94.Found: C, 56.89; H, 5.50; N, 5.89.


Methyl triflate (72 mL, 0.66 mmol) was added at 0 �C to a solu-tion of dendrimer 10-G0 (200 mg, 0.3 mmol) in dichloromethane(20 mL). The resulting mixture was stirred for 24 h at roomtemperature then evaporated to dryness. The residue was washedtwice with Et2O/pentane (50:50) then dissolved in 10 mL of CH3CN/H2O (80:20). This solution was passed through an ion exchangeresin DOWEX-1X8 beforehand charged in Cl� using the followingprocedure: 3 mL of DOWEX 1X8 (4.20 mmol in OH�) were placedon a frit and 30 mL of acid chloride 2.0 M were passed through; theresin was washed with distilled water up to pH 7, and was used as itwas. The solution of dendrimer passed through this resin wasevaporated under vacuum then lyophilized. Dendrimer 12-G0 wasisolated as a white powder in 92% yield (154 mg). 1H NMR (DMSOD6, 250 MHz): d¼ 3.39 (s, 3H, Nþ–CH3), 4.12 (br s, 6H, CH2–O), 4.64(br s, 6H, N–CH2), 7.06 (d, 3JHH¼ 8.5 Hz, 6H, C0

2H), 7.92 (d,3JHH¼ 8.5 Hz, 6H, C0

3H) ppm. 13C–{1H} NMR (DMSO D6, 63 MHz):d¼ 49.90 (s, Nþ–CH3), 61.40 (s, CH2–O), 61.43 (s, N–CH2), 114.07 (s,C0

2), 126.92 (s, C04), 130.98 (s, C0

3), 159.93 (s, C01), 167.49 (s, COOH)

ppm. IR (KBr): 1694 cm�1 (nC]O). Anal. [C28H30ClNO9] C, H, N.Calcd: C, 60.06; H, 5.40; N, 2.50. Found: C, 60.10; H, 5.42; N, 2.47.


20 mL of a solution of trifluoroacetic acid (4 mL) at 20% indichloromethane (16 mL) was added to powdered dendrimer 11-G1


(205 mg, 0.10 mmol). The resulting solution was stirred at roomtemperature for 1.5 h. A co-evaporation was carried out in thepresence of ethyl acetate under reduced pressure. This procedurewith ethyl acetate was repeated with the residue to afford a yellowpowder, which was dissolved in 15 mL of CH3CN/H2O (80:20). Thissolution was passed through an ion exchange resin DOWEX-1X8beforehand charged in Cl� (see above this procedure for 12-G0). Thesolution of dendrimer passed through this resin was evaporatedunder vacuum then lyophilized. Dendrimer 12-G1 was isolated asa white powder in 80% yield (137 mg). 31P-{1H} NMR (CD3CN/CD3COOD, 80 MHz): d¼ 64.51 (s) ppm. 1H NMR (CD3CN/CD3COOD,200 MHz): d¼ 3.34 (d, 3JHP¼ 10.6 Hz, 9H, P–NCH3), 3.43 (s, 3H, Nþ–CH3), 4.11 (br s, 6H, CH2–O), 4.60 (br s, 6H, N–CH2), 6.97(d, 3JHH¼ 8.8 Hz, 6H, C0

2H), 7.28 (d, 3JHH¼ 8.8 Hz, 12H, C12H), 7.60 (d,

3JHH¼ 8.8 Hz, 6H, C03H), 7.73 (br s, 3H, CH]N), 7.99 (d, 3JHH¼ 8.8 Hz,

12H, C13H) ppm. 13C–{1H} NMR (CD3CN/CD3COOD, 63 MHz):

d¼ 33.56 (d, 2JCP¼ 12.6 Hz, P–N–CH3), 51.32 (s, Nþ–CH3), 62.93 (s,CH2–O), 63.93 (s, N–CH2), 116.03 (s, C0

2), 122.43 (s, C12), 127.70 (s, C0

3),129.68 (s, C1

4), 130.00 (s, C04), 132.95 (s, C1

3), 141.86 (d, 3JCP¼ 13.8 Hz,CH]N), 155.92 (s, C1

1), 159.62 (br s, C01), 170.63 (s, COOH) ppm. IR

(KBr): 1698 cm�1 (nC]O). Anal. [C73H69ClN7O21P3S3] C, H, N. Calcd:C, 54.63; H, 4.33; N, 6.11. Found: C, 54.69; H, 4.36; N, 6.07.


20 mL of a solution of trifluoroacetic acid (4 mL) at 20% indichloromethane (16 mL) was added to powdered dendrimer 11-G2

(150 mg, 0.033 mmol). The resulting solution was stirred at roomtemperature for 1.5 h. A co-evaporation was carried out in thepresence of ethyl acetate under reduced pressure. This procedurewas repeated with the residue to afford a yellow powder, whichwas dissolved in CH3CN/H2O (80:20) (15 mL). This solution waspassed through an ion exchange resin DOWEX-1X8 beforehandcharged in Cl� (see above this procedure for 12-G0). The solution ofdendrimer passed through this resin was evaporated under vacuumthen lyophilized. Dendrimer 12-G2 was isolated as a white powderin 85% yield (106 mg). 31P-{1H} NMR (CD3COOD/CD3CN, 80 MHz):d¼ 64.69 (s, P2), 65.60 (s, P1) ppm. 1H NMR (CD3COOD/CD3CN,200 MHz): d¼ 3.41 (m, 27H, P2–NCH3, P1–NCH3), 3.55 (br s, 3H, Nþ–CH3), 4.25 (br s, 6H, CH2–O), 4.70 (br s, 6H, N–CH2), 7.05 (br s, 6H,C0

2H), 7.34 (br s, 36H, C22H, C1

2H), 7.69 (m, 27H, C03H, C1

3H, CH]N),8.06 (br s, 24H, C2

3H) ppm. 13C–{1H} NMR (CD3COOD/CD3CN,63 MHz): d¼ 33.15 (d, 2JCP¼ 13.0 Hz, P2–NCH3, P1–NCH3), 51.32 (s,Nþ–CH3), 62.20 (s, CH2–O), 63.80 (s, N–CH2), 116.11 (s, C0

2), 122.73 (s,C2

2), 122.93 (s, C12), 128.21 (s, C1

3), 128.88 (s, C03), 129.68 (s, C2

4), 130.10(s, C1

4), 130.20 (s, C04), 133.72 (s, C2

3), 141.12 (m, CH]N), 153.23 (s, C11),

156.45 (br s, C21), 159.25 (br s, C0

1), 170.62 (s, COO) ppm. IR (KBr):1698 cm�1 (nC]O). Anal. [C163H147ClN19O45P9S9] C, H, N. Calcd: C,52.99; H, 4.01; N, 7.20. Found: C, 53.05; H, 4.06; N, 7.17.


10 mL of water were added to dendrimer 12-G0 (60 mg,0.11 mmol) and carteolol 2 (106 mg, 0.36 mmol). The resultingmixture was stirred for 24 h to afford a clear solution, which wasthen lyophilized. The resulting powder was washed with ether(2�10 mL) to eliminate the slight excess of 2. Dendrimer 13-G0 wasisolated as a white powder in 96% yield (160 mg). 1H NMR (CD3CN/D2O, 200 MHz): d¼ 1.57 (s, 27H, C(CH3)3), 2.92 (t, 3JHH¼ 6.0 Hz, 6H,CH2–CH2–C]O), 3.32 (t, 3JHH¼ 6.0 Hz, 6H, CH2–C]O), 3.35–3.55(m, 6H, NH2–CH2), 3.65 (s, 3H, Nþ–CH3), 4.34 (br s, 12H, CH2CH2O,Me–Nþ–CH2), 4.58 (m, 3H, CHOH), 4.87 (br s, 6H, CH–CH2–O), 6.86(d, 2JHH¼ 8.1 Hz, 3H, CfH), 6.98 (d, 2JHH¼ 8.1 Hz, 3H, CdH), 7.26 (d,2JHH¼ 8.5 Hz, 6H, C0

2H), 7.56 (t, 3JHH¼ 8.0 Hz, 3H, CeH), 8.15 (d,2JHH¼ 8.5 Hz, 6H, C0

3H) ppm. 13C–{1H} NMR (CD3CN/D2O, 63 MHz):d¼ 18.80 (s, CH2–CH2–C]O), 25.70 (s, C(CH3)3), 30.15 (s, CH2–C]O), 44.90 (s, N–CH2), 52.40 (s, Nþ–CH3), 58.00 (s, C(CH3)3), 62.40

(s, Me–Nþ–CH2), 63.45 (s, CH2–CH2–O), 66.85 (s, CHOH), 70.75(s, O–CH2CH), 108.25 (s, Cf), 110.11 (s, Cd), 113.10 (s, Cc), 114.75 (s, C0

2),129.10 (s, Ce), 131.30 (s, C0

3), 132.00 (s, C04), 140.30 (s, Cb), 156.25 (s,

Ca), 159.05 (s, C01), 174.60 (br s, CO2, CONH) ppm. IR (KBr): 1672 (nCO

carteolol), 1603 (nC]C aromatic), 1549 (nCOO�

asym), 1480 (nC]C aromatic),1381 cm�1 (nCOO

�sym), 1335 (nCN lactam). Anal. [C76H102ClN7O18] C, H,

N. Calcd: C, 63.52; H, 7.15; N, 6.82. Found: C, 63.50; H, 7.17; N, 6.79.


10 mL of water were added to dendrimer 12-G1 (60 mg,0.036 mmol) and carteolol 2 (80 mg, 0.24 mmol). The resultingmixture was stirred for 24 h to afford a clear solution, which wasthen lyophilized. The resulting powder was washed with ether(2�10 mL) to remove the slight excess of 2. Dendrimer 13-G1 wasisolated as a white powder in 97% yield (136 mg). 31P-{1H} NMR(CD3CN/CD3COOD, 80 MHz): d¼ 66.40 (s) ppm. 1H NMR (CD3CN/D2O, 200 MHz): d¼ 1.65 (s, 54H, C(CH3)3), 2.89 (t, 3JHH¼ 6.0 Hz,12H, CH2–CH2–C]O), 3.27 (t, 3JHH¼ 6.0 Hz, 12H, CH2–C]O), 3.35–3.85 (m, 33H, NH2–CH2, Nþ–CH3, N–Me), 4.39 (br s, 18H, O–CH2CH,Me–Nþ–CH2), 4.56 (m, 6H, CHOH), 4.92 (br s, 6H, CH2CH2–O), 6.92(d, 2JHH¼ 8.1 Hz, 6H, CfH), 7.03 (d, 2JHH¼ 8.1 Hz, 6H, CdH), 7.38 (d,2JHH¼ 8.5 Hz, 6H, C0

2H), 7.52 (m, 18H, CeH, C12H), 8.05 (m, 6H, C0

3H),8.25 (m, 18H, C1

3H, CH]N) ppm. 13C–{1H} NMR (CD3CN/D2O,63 MHz): d¼ 18.54 (s, CH2–CH2–C]O), 25.83 (s, C(CH3)3),30.15 (s, CH2–C]O), 44.93 (s, N–CH2), 52.32 (s, Nþ–CH3), 57.88(s, C(CH3)3), 62.42 (s, Me–Nþ–CH2), 63.49 (s, CH2–CH2–O), 66.98(s, CHOH), 70.80 (s, O–CH2CH), 108.21 (s, Cf), 110.11 (s, Cd),113.16 (s, Cc), 115.91 (s, C0

2), 121.38 (s, C12), 127.21 (s, C0

3), 129.15 (s, Ce),129.50 (br s, C0

4, C14), 131.85 (s, C1

3), 140.25 (br s, CH]N, Cb), 153.80 (s,Ca), 153.1 (br s, C1

1), 156.30 (s, Ca), 174.60 (br s, CO2, CONH) ppm. C01

could not be detected. IR (KBr): 1668 (nCO carteolol), 1601 (nC]C

aromatic), 1558 (nCOO�

asym), 1480 (nC]C aromatic), 1382 cm�1 (nCOO�

sym),1336 (nCN lactam). Anal. [C169H213ClN19O39P3S3] C, H, N. Calcd: C,60.43; H, 6.39; N, 7.92. Found: C, 60.48; H, 6.42; N, 7.85.


10 mL of water were added to dendrimer 12-G2 (60 mg,0.016 mmol) and carteolol 2 (65 mg, 0.22 mmol). The resultingmixture was stirred for 24 h to afford a clear solution, which was thenlyophilized. The resulting powder was washed with ether (2�10 mL)to eliminate the slight excess of 2. Dendrimer 13-G2 was isolated asa white powder in 85% yield (99 mg). This compound is not enoughsoluble to perform NMR experiments. IR (KBr): 1666 (nCO carteolol),1600 (nC]C aromatic), 1555 (nCOO

�asym), 1482 (nC]C aromatic), 1390 cm�1

(nCOO�

sym), 1336 (nCN lactam). Anal. [C355H435ClN43O81P9S9] C, H, N.Calcd: C, 59.19; H, 6.09; N, 8.36. Found: C, 59.25; H, 6.11; N, 8.32.

4.2. Pharmacology

4.2.1. Materials and animalsAll the rabbits used in this study were cared for and treated in

accordance with the ARVO statement for the use of animals inophthalmic and vision research. Female New Zealand albino rabbits2.5 to 3 kg in weight and 10–12 weeks of age were used. The car-teolol titration in the initial solutions and in the aqueous humoursamples was performed by HPLC–MS method (Agilent Technolo-gies), using Timolol as internal standard.

4.2.2. Methods for drug delivery experimentsCarteolol or dendrimers 13-Gn (n¼ 0, 1, 2) were dissolved in

milliQ water 1 min before instillation, and one drop (50 mL) of thecarteolol/dendrimer solution was administered in each eye. After 1,4, 6 or 8 h and before taking the aqueous humour samples, therabbits were sedated with 2 mg/kg Nidazolam (Hypnovel ROCHE,Neuilly sur Seine, France), administered intramuscularly 20 min


before sampling. The eyes were examined to detect any corneal orconjunctival irritation (none was seen), and subsequently wereanesthetized topically with one drop (50 mL) of oxybuprocaıne(Chlorhydrate d’oxybuprocaıne, Novartis Pharma SAS, France).Then after one minute, the eyes were washed with physiologicalserum and 100 mL of aqueous humour were taken using a syringeand an operating microscope.

Acknowledgements

Thanks are due to the Fonds Social Europeen (PhD grant to G.Spataro) and to the CNRS for financial support.

Appendix. Supplementary Data

Supplementary data associated with this article can be found inthe online version, at doi:10.1016/j.ejmech.2009.10.017.

References

[1] G.R. Newkome, C.D. Shreiner, Polymer 49 (2008) 1–173.[2] U. Boas, P.M.H. Heegaard, Chem. Soc. Rev. 33 (2004) 43–63.[3] C.C. Lee, J.A. MacKay, J.M.J. Frechet, F.C.Szoka,Nat.Biotechnol. 23 (2005) 1517–1526.[4] S. Svenson, D.A. Tomalia, Adv. Drug Delivery Rev. 57 (2005) 2106–2129.[5] M. Najlah, A. D’Emanuele, Curr. Opin. Pharmacol. 6 (2006) 522–527.[6] A.M. Caminade, C.O. Turrin, J.P. Majoral, Chem. Eur. J. 14 (2008) 7422–7432.[7] E.R. Gillies, J.M.J. Frechet, Drug Discovery Today 10 (2005) 35–43.[8] H.L. Crampton, E.E. Simanek, Polym. Int. 56 (2007) 489–496.[9] A. Myc, I.J. Majoros, T.P. Thomas, J.R. Baker Jr., Biomacromolecules 8 (2007) 13–18.

[10] U. Gupta, H.B. Agashe, A. Asthana, N.K. Jain, Nanomedicine 2 (2006) 66–73.[11] R. Duncan, L. Izzo, Adv. Drug Delivery Rev. 57 (2005) 2215–2237.[12] D.A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos, S. Martin, J. Roeck, J. Ryder,

P. Smith, Polym. J. 17 (1985) 117–132.[13] R. Esfand, D.A. Tomalia, Drug Discovery Today 6 (2001) 427–436.[14] D.A. Tomalia, L.A. Reyna, S. Svenson, Biochem. Soc. Trans. 35 (2007) 61–67.[15] P. Ortega, J.F. Bermejo, L. Chonco, E. de Jesus, F.J. de la Mata, G. Fernandez,

J.C. Flores, R. Gomez, M.J. Serramıa, M.A. Munoz-Fernandez, Eur. J. Inorg.Chem. (2006) 1388–1396.

[16] M. Wyszogrodzka, R. Haag, Chem. Eur. J. 14 (2008) 9202–9214.[17] H.T. Chen, M.F. Neerman, A.R. Parrish, E.E. Simanek, J. Am. Chem. Soc. 126

(2004) 10044–10048.[18] N. Launay, A.M. Caminade, R. Lahana, J.P. Majoral, Angew. Chem., Int. Ed. Engl.

33 (1994) 1589–1592.[19] N. Launay, A.M. Caminade, J.P. Majoral, J. Organomet. Chem. 529 (1997) 51–58.[20] C. Loup, M.A. Zanta, A.M. Caminade, J.P. Majoral, B. Meunier, Chem. Eur. J. 5

(1999) 3644–3650.[21] M. Maszewska, J. Leclaire, M. Cieslak, B. Nawrot, A. Okruszek, A.M. Caminade,

J.P. Majoral, Oligonucleotides 13 (2003) 193–207.[22] C. Padie, M. Maszewska, K. Majchrzak, B. Nawrot, A.M. Caminade, J.P. Majoral,

New J. Chem. 33 (2009) 318–326.[23] J. Solassol, C. Crozet, V. Perrier, J. Leclaire, F. Beranger, A.M. Caminade, B. Meunier,

D. Dormon, J.P. Majoral, S. Lehmann, J. Gen. Virol. 85 (2004) 1791–1799.[24] B. Klajnert, M. Cortijo-Arellano, J. Cladera, J.P. Majoral, A.M. Caminade,

M. Bryszewska, Biochem. Biophys. Res. Commun. 364 (2007) 20–25.[25] B. Klajnert, M. Cangiotti, S. Calici, M. Ionov, J.P. Majoral, A.M. Caminade,

J. Cladera, M. Bryszewska, M.F. Ottaviani, New J. Chem. 33 (2009) 1087–1093.

[26] M. Blanzat, C.O. Turrin, E. Perez, I. Rico-Lattes, A.M. Caminade, J.P. Majoral,Chem. Commun. (2002) 1864–1865.

[27] M. Blanzat, C.O. Turrin, A.M. Aubertin, C. Vidal, A.M. Caminade, J.P. Majoral,I. Rico-Lattes, A. Lattes, ChemBioChem. 6 (2005) 2207–2213.

[28] B. Klajnert, M. Cangiotti, S. Calici, J.P. Majoral, A.M. Caminade, J. Cladera,M. Bryszewska, M.F. Ottaviani, Macromol. Biosci. 7 (2007) 1065–1074.

[29] O. Mongin, T.R. Krishna, M.H.V. Werts, A.M. Caminade, J.P. Majoral,M. Blanchard-Desce, Chem. Commun. (2006) 915–917.

[30] T.R. Krishna, M. Parent, M.H.V. Werts, L. Moreaux, S. Gmouh, S. Charpak,A.M. Caminade, J.P. Majoral, M. Blanchard-Desce, Angew. Chem. Int. Ed. 45(2006) 4645–4648.

[31] O. Mongin, A. Pla-Quintana, F. Terenziani, D. Drouin, C. Le Droumaguet,A.M. Caminade, J.P. Majoral, M. Blanchard-Desce, New J. Chem. 31 (2007)1354–1367.

[32] M. Poupot, L. Griffe, P. Marchand, A. Maraval, O. Rolland, L. Martinet,F.E. L’Faqihi-Olive, C.O. Turrin, A.M. Caminade, J.J. Fournie, J.P. Majoral,R. Poupot, FASEB J. 20 (2006) 2339–2351.

[33] L. Griffe, M. Poupot, P. Marchand, A. Maraval, C.O. Turrin, O. Rolland,P. Metivier, G. Bacquet, J.J. Fournie, A.M. Caminade, R. Poupot, J.P. Majoral,Angew. Chem. Int. Ed. 46 (2007) 2523–2526.

[34] O. Rolland, L. Griffe, M. Poupot, A. Maraval, A. Ouali, Y. Coppel, J.J. Fournie,G. Bacquet, C.O. Turrin, A.M. Caminade, J.P. Majoral, R. Poupot, Chem. Eur. J. 14(2008) 4836–4850.

[35] S. Fruchon, M. Poupot, L. Martinet, C.O. Turrin, J.P. Majoral, J.J. Fournie,A.M. Caminade, R. Poupot, J. Leukocyte Biol. 85 (2009) 553–562.

[36] P. Marchand, L. Griffe, M. Poupot, C.O. Turrin, G. Bacquet, J.J. Fournie,J.P. Majoral, R. Poupot, A.M. Caminade, Bioorg. Med. Chem. Lett. 19 (2009)3963–3966.

[37] H.H. Sigurdsson, F. Konradsdottir, T. Loftsson, E. Stefansson, Acta Ophthalmol.Scand. 85 (2007) 598–602.

[38] S.K. Sahoo, F. Dilnawaz, S. Krishnakumar, Drug Discovery Today 13 (2008)144–151.

[39] M.E. Stern, R.W. Beuerman, R.I. Fox, J.P. Gao, A.K. Mircheff, S.C. Pflugfelder,Cornea 17 (1998) 584–589.

[40] S. Uppuluri, S.E. Keinath, D.A. Tomalia, P.R. Dvornic, Macromolecules 31 (1998)4498–4510.

[41] S. Merino, L. Brauge, A.M. Caminade, J.P. Majoral, D. Taton, Y. Gnanou, Chem.Eur. J. 7 (2001) 3095–3105.

[42] T.F. Vandamme, L. Brobeck, J. Controlled Release 102 (2005) 23–38.[43] D.F. Woodward, J. Chen, Expert Opin. Emerg. Drugs 12 (2007) 313–327.[44] A.F. Clark, T. Yorio, Nat. Rev. Drug Discov. 2 (2003) 448–459.[45] S. Henness, T.S. Harrison, G.M. Keating, Drugs Aging 24 (2007) 509–528.[46] C. Baudouin, C. de Lunardo, Br. J. Ophthalmol. 82 (1998) 39–42.[47] R. Goller, J.P. Vors, A.M. Caminade, J.P. Majoral, Tetrahedron Lett. 42 (2001)

3587–3590.[48] P. Jokela, B. Jonsson, A. Khan, J. Phys. Chem. 91 (1987) 3291–3298.[49] T. Bramer, N. Dew, K. Edsman, J. Pharm. Pharmacol. 59 (2007) 1319–1334.[50] S. Consola, M. Blanzat, E. Perez, J.C. Garrigues, P. Bordat, I. Rico-Lattes, Chem.

Eur. J. 13 (2007) 3039–3047.[51] M. Blanzat, E. Perez, I. Rico-Lates, J.C. Prome, A. Lattes, Langmuir 15 (1999)

6163–6169.[52] G. Guipponi, D.M.A. Buzza, D.B. Adolf, Macromolecules 40 (2007) 5959–5965.[53] M.K. Boggiano, G.J.A.A. Soler-Illia, L. Rozes, C. Sanchez, C.O. Turrin,

A.M. Caminade, J.P. Majoral, Angew. Chem. Int. Ed. 39 (2000) 4249–4254.[54] T.W. Green, P.G.M. Wuts, Protective Groups in Organic Synthesis. Wiley

Interscience, New-York, 1999.[55] J. Leclaire, Y. Coppel, A.M. Caminade, J.P. Majoral, J. Am. Chem. Soc. 126 (2004)

2304–2305.[56] M.A. Holgado, M. Fernandez-Arevalo, J. Alvarez-Fuentes, I. Carballo, J.M. Llera,

A.M. Rabasco, Int. J. Pharm. 114 (1995) 13–21.[57] K.G. Ragunathan, R. Shulkla, S. Mishra, P.K. Bharadwaj, Tetrahedron Lett. 35

(1993) 5631–5634.

http://dx.doi.org/doi:10.1016/j.ejmech.2009.10.017

designing dendrimers for ocular drug delivery

Documents