Bioconjugate Chem. 19Q2, 3, 49-57
`Synthesis, Conformation, Biodistribution, and in Vitro Cytotoxicity of
`Daunomycin-Branched Polypeptide Conjugates?
`
`49
`
`F. Hudecz,*p* J. A. Clegg,g J. Kajtiir,ll M. J. Embleton,s M. Szekerke,t and R. W. Baldwine
`Research Group for Peptide Chemistry, Hungarian Academy of Science, and Institute of Organic Chemistry,
`Eotvos L. University, Budapest 112 POB 32, H-1518 Hungary, and Cancer Research Campaign Laboratories,
`University of Nottingham, University Park, Nottingham NG7 2RD, United Kingdom. Received August 2, 1991
`
`Daunomycin has been attached to various structurally related synthetic branched polypeptides with
`a polylysine backbone, using its acid-labile cis-aconityl derivative (CAD). Due to the importance of the
`side-chain structure in a-helix formation and immunological and pharmacological properties of branched
`polypeptides, we have investigated the conformation, biodistribution, and in vitro cytotoxicity of CAD-
`carrier conjugates with polypeptides containing amino acid residues of different identity and/or con-
`figuration at the side-chain end (XAK type) or at the position next to the polylysine backbone (AXK
`type), where X = Leu, D-Leu, Pro, Glu, or D-G~u. According to CD studies, polycationic conjugates with
`hydrophobic Leu in the side chains could assume a highly ordered conformation, while amphoteric
`conjugates containing Glu proved to be unordered in PBS. The reduction of in vitro cytotoxic activity
`of CAD by conjugation to carriers and the biodistribution profile of the conjugates were found to be
`dependent predominantly on the charge properties and on the side-chain sequence of the carrier poly-
`peptide. It was demonstrated that by proper combination of structural elements of the carrier molecule,
`it is feasible to construct a CAD-branched polypeptide conjugate with significantly prolonged blood
`survival and with no reduction in in vitro cytotoxicity of the drug.
`
`INTRODUCTION
`Daunomycin (Dau)l is one of the clinically important
`anthracycline antitumor antibiotics. However, its efficacy
`is restricted by a range of undesirable side effects such as
`cumulative myocardial toxicity. Considerable efforts have
`been made to improve the therapeutic index, including
`attempts at site-specific targeting or providing controlled
`release of Dau by using natural macromolecules [e.g.,
`hormones ( I ) , enzymes (21, poly- and monoclonal anti-
`bodies (3, 41, or DNA (511.
`Daunomycin has also been linked to various synthetic
`polymers such as poly(Lys) (6-8), poly(Asp) (9), SPDP-
`modified poly(G1u) (101, poly[Glu(NsHs)l (10, poly(di-
`vinyl ether-co-maleic anhydride) (12), N-(2-hydroxypropyl)-
`methacrylamide (HPMA) (13), poly[acryloyl-2-amido-2-
`(hydroxymethyl)-l,3-propanediol] (14), and dextran (15).
`Some of these daunomycin conjugates, prepared with dex-
`
`* To whom all correspondence and reprint requests should be
`addressed.
`+ Presented in part at the 31st Annual Meeting of the British
`Association of Cancer Research, Brighton, UK, March 19-22,
`1990, and at the 21st European Peptide Symposium, Barcelona,
`Spain, September 2-7, 1990.
`Hungarian Academy of Science, Eotvos L. University.
`5 Cancer Research Campaign Laboratories, University of Not-
`tingham.
`11 Institute of Organic Chemistry, Eotvos L. University.
`1 Abbreviations used (in order of appearance in the text): Dau,
`daunomycin; SPDP, N-succinimidyl 3-(2-pyridyldithio)propi-
`onate; HPMA,N-(2-hydroxypropyl)methacrylamide; XAK, poly-
`[Lys-(Xi-~~-Ala,)] where X = Glu, D-G~u, Leu, D-Leu, or Pro;
`AXK, poly[Lys-(~~-Ala,-Xi)] where X = Glu or Leu; CAD, cis-
`aconityldaunomycin; CD, circular dichroism; HSA, human serum
`albumin; CMC, N-cyclohexyl-N’-(2-morpholinoethyl)carbodi-
`imide methyl p-toluenesulfonate; DS, average degree of substi-
`tution; XAK-CAD, poly[Lys-(cADj-Xi-~~-Ala~)l where X = Glu,
`D-G~u, Leu, D-Leu, or Pro; AXK-CAD, poly[Lys-(cADj-DL-Ala,-
`Xi)] where X = Glu or Leu; E,, number average of the degree
`of polymerization; AK, poly[Lys-(~~-Ala,)]; MEM, minimum
`essential medium; NBCS, newborn calf serum.
`
`tran (161, HPMA (In, or derivatives of poly(G1u) (10, II),
`have been used for coupling to cell-specific poly- or mon-
`oclonal antibodies.
`The methods for conjugation of daunomycin to synthetic
`carriers have involved (a) the a-amino group of dauno-
`samine using a cross-linking agent [e.g., carbodiimide (IO)],
`anhydride (121, or active ester (13) of the polymer’s car-
`boxylic groups, (b) cleavage of the bond between C-3 and
`C-4 of the amino sugar (151, (c) the methyl ketone side
`chain of the aglycon by nucleophilic substitution of its
`14-bromo derivative (8, 9, 11). Conjugates have been
`synthesized by the incorporation of leucyl or aspartyl (14),
`maleyl or cis-aconityl (61, and succinyl or several other
`diacidic (7) spacers.
`Although considerable work has been performed with
`preparation and analysis of daunomycin-polymer conju-
`gates (6-15), few empirical studies have been reported to
`establish the structure-activity relationships of carrier
`molecules in terms of their biological properties required
`for optimized daunomycin delivery and/or targeting.
`In the past decade new groups of branched polypep-
`tides with the general formula poly[Lys-(Xi-~~-Ala,)]
`(AXK), where i < 1 and
`(XAK) or poly[Lys-(~~-Ala~-Xi)]
`(20-24) by our laboratories. Potential advantages provided
`by this water-soluble, biodegradable group of polymeric
`polypeptides include (a) beneficial chemical characteris-
`tics, like a large number of a-amino groups available for
`simple and efficient coupling of daunomycin derivative
`without polymer cross-linking, (b) reliable determination
`of conjugate composition and conformation, and (c)
`comparative analysis of the functional properties (e.g. cy-
`totoxicity, immunogenicity, biodistribution) of carriers
`with respect to their size, primary structure, charge (poly-
`cationic vs amphoteric compounds), and secondary struc-
`ture (ordered vs unordered). The aim of present work is
`to provide a logical basis for selection of synthetic mac-
`romolecular carrier for constructing conjugates with dauno-
`mycin. In order to achieve this, Dau has been attached
`to various structurally related polymeric polypeptides with
`
`m - 3, have been developed (18-21) and characterized
`
`0 1992 American Chemical Society
`
`SANOFI-AVENTIS Exhibit 1040 - Page 49
`
`IPR for Patent No. 8,951,962
`
`

`
`--
`
`50 Bioconjugate Chem., Vol. 3, No. 1, 1992
`
`a
`
`b
`
`o l i g o ( D L - A l a )
`
`A
`
`o l i g o ( D L - A l a )
`
`Hudecz et ai.
`composition of the side chains (33). The size of these
`compounds was analyzed by sedimentation analysis and
`gel chromatography (18).
`Methods. Synthesis of Branched Polypeptide-CAD
`Conjugates. Coupling of cis-aconityldaunomycin to the
`branched polypeptides was performed using the method
`previously described for the conjugation of CAD to mon-
`oclonal antibodies (4), with some modifications. Briefly,
`5.0-9.5 pmol (3.4-6.5 mg) of CAD in 144-210 pL of PBS
`was diluted with PBS (pH 5) to a final concentration of
`4 mg/mL. The solution was stirred and treated with 7.5-
`14.25 pmol(3.1-6.0 mg) of CMC dissolved in 400-600 pL
`of distilled water (Le., a 1.5 times molar excess of carbo-
`diimide reagent) for 30 min at room temperature. This
`mixture was added dropwise to 20 mg (0.1-0.5 pmol) of
`polypeptide in 10 mL of 0.05 M carbonate buffer (pH 9.0).
`After stirring for 2 h, the reaction was allowed to proceed
`for 16 h at room temperature. The final concentration of
`polypeptide was 1.6 mg/mL. The input molar ratio of
`CAD to polypeptide was between 20:l and 9O:l. After
`conjugation, the reaction mixture was applied to a Sepha-
`dex G-25 (medium grade) column equilibrated with PBS.
`The appropriate macromolecular peak as defined by UV
`detection at 275 nm was collected. The CAD was deter-
`mined spectrophotometrically as described previously (4).
`It was assumed that the CAD coupled to the a-amino group
`of terminal amino acid has the same molar extinction
`coefficient at X = 476 nm as CAD itself (e = 8000).
`Analytical data on the conjugates are presented in Table
`I.
`
`HPLC Analysis of Branched Polypeptide-CAD
`Conjugates. Instrumentation. The HPLC system
`consisted of one Model 302 and one Model 303 liquid-
`delivery module, a Model HM/HPLC Holochrome UV-
`visible detector, a Model 811 dynamic mixer, a Model 802
`manometric module, a Model N1 recorder (all from Gil-
`son France S.A.), a Model 7125 injector valve, and 100-pL
`and 200-pL sample loops (Rheodyne, Inc., Cotati, CAI. A
`25 cm X 4.6 mm column and a 3 cm X 4.6 mm Aquapore
`RP-300 guard column (Brownlee Labs., Inc., Santa Clara,
`CA) were used. Both columns contained spherical 7-pm
`silica (300-A pore size) with a hydrophobic bonded phase.
`Conditions. The mobile phase consisted of 0.2 M
`ammonium carbonate/methanol(50:50, v/v). Ammonium
`carbonate solution was prepared daily using high-purity
`(Analar) salt and distilled, deionized water and was mixed
`with methanol 20 min before use. The injection volume
`was 100 pL containing 0.3 pg (0.53 nmol) of daunomycin
`and/or 0.3 pg (0.44 nmol) of CAD dissolved in eluent or 200
`pL containing 1.1-5.7 pg (1.5-10.1 nmol) of CAD coupled
`to branched polypeptide fn PBS. Daunomycin hydro-
`chloride and CAD were used separately as standards and
`their retention times were determined. UV absorbance
`was monitored at a wavelength of 280 nm. All analyses
`were carried out at room temperature with a flow rate of
`1.0 mL/min.
`Spectroscopic Measurements. Absorption spectra
`were recorded on an ultraviolet-visible spectrophotom-
`eter (Unicam-Pye) in a cell of optical path 1.0 cm at room
`temperature. CAD and conjugates were dissolved in PBS
`at pH 7.3. Conformations of the conjugates were studied
`by CD spectroscopy. CD spectra were recorded using a
`Roussel-Jouan (Jobin-Yvon) Model I11 dichrograph in
`quartz cells of optical paths 1.0, 0.2, and 0.1 cm at room
`temperature under constant nitrogen flush. The dichro-
`graph was calibrated with epiandrosterone at 304 nm and
`D-(-)-pantoy1 lactone at 220 nm (34). The samples were
`dissolved in PBS at pH 7.3. The concentration of
`
`Poly ( L Y S ) \ ,/ P O l Y ( L Y S )
`C A D
`Figure 1. Schematic representation of branched polypeptide-
`cis-aconityldaunomycin (CAD) conjugates: (a) poly[Lys-(cADj-
`( W - C A D ) , (b) pOly[Lys-(CADj-DL-~-x,)] (m-
`&-DL-Al&)]
`CAD).
`different chemical (20, 21 1, immunological (25-28), and
`pharmacokinetic (29) properties.
`In this paper we describe the synthesis of daunomycin-
`branched polypeptide conjugates (Figure l), in which an
`acid-labile, cis-aconityl derivative of daunomycin (CAD)
`was applied (6). To understand the possible role of the
`secondary structure, the solution conformation of these
`conjugates has been analyzed by circular dichroism (CD)
`spectroscopy. Comparative studies have been performed
`on CAD conjugates to determine predominant structural
`features of the carrier polypeptide influencing the con-
`jugate’s biodistribution (blood-clearance profile, whole-
`body survival, tissue distribution) and cytotoxicity. Con-
`jugates have been tested and their in vitro cytotoxicity
`compared with that of free drug and of other conjugates
`like CAD-HSA or CAD-monoclonal antibody (791T/36) on
`osteogenic sarcoma cell line 791T.
`
`EXPERIMENTAL PROCEDURES
`Abbreviations used in this paper follow the rules of the
`IUPAC-IUB Commission of Biochemical Nomenclature
`(30) in accord with the recommended nomenclature of
`graft polymers (31).
`Materials. Daunomycin hydrochloride was obtained
`from Wyeth; cis-aconitic anhydride and N-cyclohexy1-N’-
`(2-morpholinoethy1)carbodiimide methyl p-toluenesul-
`fonate (CMC) were from BDH (UK). cisdconityldauno-
`mycin (CAD) was prepared as reported elsewhere (4).
`Branched polypeptides used in these studies were syn-
`thesized in our laboratory as described previously (20,21).
`Briefly, poly(Lys) was prepared by the polymerization of
`Na-carboxy-W-(benzyloxycarbony1)lysine anhydride un-
`der conditions that allowed an average degree of polym-
`
`erization (m,) of approximately either 80-120 or 400-
`
`500. After cleavage of protecting groups from poly[Lys(Z)] ,
`either poly[Lys-(~~-Ala,l (AK) was prepared by grafting
`of short oligomeric DL-Ala side chains onto the e-amino
`groups of poly(Lys) or protected amino acid was coupled
`by the active ester method. Poly[Lys-(Xi-o~-Ala,)I (XAK)
`was synthesized by reacting a suitably protected amino
`acid pentachlorophenyl ester to the a-amino groups of
`AK. Blocking groups were removed completely with HBr
`in glacial acetic acid, as confirmed by UV spectroscopy at
`254 nm. Poly[Lys-(~~-Ala,-Xj)] (AXK) was prepared by
`the introduction of DL-Ala oligomers to the previously de-
`protected a- and e-amino groups of p~ly[Ly~-(Xi)] by the
`aid of N-carboxy-DL-Ala anhydride. The primary structure
`of polypeptides was studied by amino acid analysis, by
`the identification of the branch-terminating amino acid
`residue (321, and by the determination of the enantiomer
`
`SANOFI-AVENTIS Exhibit 1040 - Page 50
`
`IPR for Patent No. 8,951,962
`
`

`
`Daunomycln-Branched Polypeptide Conjugates
`
`Bloconjugare Chem, Vol. 3. No. 1, 1992
`
`51
`
`Table 1. Characteristics of Branched Polypeptide—¢.-is-Aconityldaunomycin Conjugates
`
`molar ratio
`% free OF
`
`conjugate
`abbreviation‘
`polypeptide
`cAD
`% DSD
`M.‘ I 5%
`cAD/Dau
`poly[Lys-(cAD,--oi.-Ala2.s4)l
`alr—cAD
`1
`5.5
`5.5
`37700
`0.7
`poly[Lys-(cAD,'-DL-Ala_g_1)]
`AK-cAD
`1
`24.3
`5.4
`172300
`0.8
`p0ly[LyS-(CADj-DL-Alfl2_9'Lel10.78)I
`alk-CAD
`1
`4.8
`4.8
`42100
`0.7
`p0ly[LyS-(CAD,'-L&uo,g1-DL-Al82.94)I
`la.k—cAD
`1
`4.5
`4.5
`37000
`0.9
`p0ly[Ly8-(CAD;-Leuogg-DL'Al8&1)l
`LAK-cAD
`1
`24.8
`5.5
`223100
`0.8
`p0ly[Lys-(CAD;-D-Leuo,a-DL-Al82.D4)]
`D-lak—cAD
`1
`4.8
`4.3
`37200
`<0.5
`p0ly[Lys-(CADj-D-Leuo,93~DL-Al&g_1)]
`D—LAK—cAD
`1
`20.8
`4.6
`219900
`<0.5
`p0ly[LyS-(CADj-Pl'0o_g4-DL-Ai82.94)]
`palr—cAD
`1
`16.0
`16.0
`56100
`<0.5
`poly[Lys-(cAD,~-Proogs-m.-Alas]ll
`PAK-cAD
`1
`70.0
`15.5
`245300
`<0.5
`poly[Lys-(cAD,-or-Ala2.e7~Glu1.o)l
`aek-cAD
`1
`5.3
`5.3
`48140
`<0.5
`poly[Lys-(cADj—Gluo_9g-DL-Alaz.94)l
`eak—cAD
`1
`12.5
`12.5
`53800
`<0.5
`poly[Lye-(cAD;-Gluose-m.-Alas.1)l
`EAK-CAD
`1
`61.9
`13.8
`253200
`<0.5
`poly[Lys-(cAD;-D-Gluo_g5-Dr.-Alas.1)]
`D-EAK—cAD
`1
`63.1
`14.0
`254000
`<0.5
`
`‘' Based on one-letter symbols of amino acids and abbreviation of cis-aconityldaumonycin (cAD). Capital or small letters denote the size
`of the conjugates. 5 Average degree of substitution, expressed as percentage of side chains modified with cAD. ° Calculated from the average
`degree of polymerization of poly(L-Lys) and of the side chain composition as described in the Experimental Procedures. ‘ Determined by
`reversed-phase HPLC as described in the Experimental Procedures.
`
`solutions was approximately 0.5 mg/mL. CD band in-
`tensities are expressed as Ac values, representing the
`difference of molar absorptivity of the left and right
`circularly polarized components. As values were related
`either to one lysine residue in the main chain including
`a whole side chain (in the range of the spectrum between
`190 and 250 nm) or to the cAD content of the conjugate
`(in the 250-600-nm absorption range). Interpretation of
`CD spectra was based on the secondary structure analysis
`of the well characterized poly(Lys) (20, 22, 23).
`Cytotoxicity Assay. Osteogenic sarcoma cell line 791T
`was grown in monolayer culture in Eag1e’s minimum
`essential medium supplemented with 10% newborn calf
`serum (MEM + NBCS).
`It was harvested for routine
`passage and for cytotoxicity assays with a mixture of 0.25 %
`trypsin and 0.1% ethylenediaminetetraacetic acid in
`phosphate-buffered saline, adjusted to pH 7.2. Cells
`suspended in MEM + NBCS were adjusted to a concen-
`tration of 5 X 10‘ per mL and plated at 100 uL (5 X 103
`cells) per well in 96-well flat-bottom tissue culture grade
`microtiter plates (Falcon 3072). They were incubated at
`37 °C for 4 h until fully adherent. Drugs (Dau and cAD)
`and conjugates in PBS were diluted in MEM + NBCS to
`twice the desired final concentration and a range of 10-
`fold dilutions prepared. Each dilution was added in 100
`;¢L to quadruplicate wells, the concentrations stated in
`the Results being the final concentrations in the wells at
`this point. Control wells were treated with 100 uL ofMEM
`+ NBCS alone. The cells were incubated for 48 h and
`then labeled with [75Se] selenomethionine (Amersham
`International plc, Amersham, UK) (0.1 uCi in 50 uL per
`well) for 16 h. They were gently washed three times with
`0.9 % NaCl solution and dried down, and the dry contents
`of the wells were sealed in with Nobecutane spray (Astra
`Pharmaceuticals Ltd). The wells were separated with a
`band saw and their radioactivity was measured as counts
`per minute (cpm) in a 7-spectrometer. Percent cytotox-
`icity was calculated by the formula
`
`mean cpm in controls — mean cpm in treated wells X 100
`mean cpm in controls
`
`Radioiodination of Branched Polypeptides-cAD
`Conjugates. Branched polypeptides—cAD conjugates
`were labeled with N-succinimidyl 3-(4-hydroxy-5-[125I]-
`iodophenyl)propionate (Amersham International plc, Am-
`ersham, UK) using the Bolton and Hunter procedure (35).
`Reagent solution (10-20 ;.LL) was added to plastic microfuge
`tubes and evaporated to dryness under a stream of
`
`nitrogen. Then 500 uL of conjugate solution at 1 mg/mL
`in 0.1 M borate buffer (pH 8.6) was reacted with iodinated
`ester (2.0 mol of ester/mol of conjugate). The reaction
`was allowed to proceed for 20 min at 0 °C and terminated
`by adding 500 ;:L of 0.2 M glycine in the same buffer for
`5 min at 0 °C. The 1251-labeled conjugate was purified on
`a G-25 Sephadex gel column using 0.066 M phosphate
`buffer (pH 7.6) containing 0.25 % gelatin as eluent. Elec-
`trophoresis on native polyacrylamide gel with a continuous
`8-25 % gradient (PhastGel gradient 8-25 Pharmacia-LKB,
`Uppsala, Sweden) was applied to assess the low molecular
`weight labeled product content of the preparation.
`Blood-Clearance and Tissue-Distribution Studies.
`Balb/t: mice (Bantin and Kingman, Hull, UK) were used
`throughout these studies. Drinking water was supple-
`mented with 0.1 % w/v sodium iodide. Groups of mice (n
`= 3) received a single injection (0.2 mL) of 1251-labeled
`conjugate via a tail vein. Serial blood samples (10 uL)
`were taken from the tail tip up to 24 h after injection. At
`this time the mice were killed and dissected. The blood
`samples, visceral organs, and residual carcasses were
`weighed and assayed for radioactivity. Results of the
`blood-clearance study were expressed as a percentage of
`the zero-time count rate assuming the blood volume of
`the mouse (mL) to be 11.2% of the body weight (g) (36).
`Area under the blood concentration-time curve up to 6 h
`following injection was calculated by the trapezoidal rule
`(37). Results of the tissue-distribution analyses were
`expressed as (i) a percentage of the injected dose of
`radioactivity per gram of tissue or blood and as (ii) ratios
`of radioactivity per gram of tissue to radioactivity per gram
`of blood (tissue to blood). Levels of statistical difference
`between groups of animals were assessed by Student's t-
`test.
`
`RESULTS
`
`Synthesis and Chemical Characterization of
`Branched Polypeptide-cAD Conjugates. The coupling
`of cAD to branched polypeptides was achieved by a car-
`bodiimide method in which one carboxyl group of the
`molecule was linked to the a-amino group of the side-
`chain terminal amino acid to provide covalent oz-amide
`bonding. First, the carboxyl groups of cAD were activated
`by water-soluble carbodiimide under conditions (i.e., 1.5
`times molar excess of carbodiimide, at pH 5.0) described
`in detail elsewhere (4). In the second step, the carboxyl-
`activated derivative was added to the amino component
`(polypeptide) and the coupling reaction allowed to proceed
`
`SANOFI-AVENTIS Exhibit 1040 - Page 51
`IPR for Patent No. 8,951,962
`
`

`
`I C '
`
`-
`
`-
`
`l
`
`
`
`5 -
`
`i
`/I
`
`
`
`1
`
`5
`
`I T
`
`B
`
`I
`
`Hudecz et el.
`
`52 Bioconjugste Chem., Vol. 3, No. 1, 1992
`in alkaline solution (pH 9.0). It should be noted that no
`precipitate was detected during the synthesis of conjugates
`under these conditions. The conjugates composed of CAD
`and XAK or AXK type polypeptide were purified by gel
`filtration and characterized byreversed-phase HPLC, CAD
`content, and iUw determination. These data are summa-
`rized in Table I. The results presented in this table indicate
`that the amount of CAD incorporated into the polypep-
`tide conjugate depends on the size of the polymer and on
`the identity of the terminal amino acid residue of the side
`chain.
`The average molar substitution ratio was in the range
`of 4.5-16.0 cis-aconityldaunomycin per carrier molecule
`in the case of small relative molar mass polypeptides
`(E, = 100) or in the range of 20.8-70.0 in the case of large
`= 450). Thus the total
`relative molar mass polymers (&,
`Figure 2. CD spectra of branched polypeptide-CAD conjugates
`and CAD in PBS of pH 7.3: (a) ak-CAD, (b) eak-CAD, (c) lak-
`number of CAD molecules in conjugates was found to
`correlate with the size of the carrier polypeptide (e.g., the
`CAD, (d) CAD.
`number of CAD per polypeptide molecule is 24.3 for poly-
`poly[Lys-(cAD;-~~-Ala2.9-Leu0.7g)l (alk-CAD) (data not
`[Lys-(cADj-~~-Alas.l)l (AK-CAD) and 5.5 for poly[Lys-
`shown). The CD signals in this range originated from the
`(cADj-~~-Ala2.94)1 (ak-CAD) or 61.9 for poly[Lys-(cADj-
`optical activity of amide bonds and from the contribution
`
`G ~ u o . ~ ~ - D L - A ~ ~ ~ J ) ] (EAK-CAD) and 12.5 for poly[Lys-
`of the CAD moiety. In order to consider this, the ellip-
`( C A D ; - G ~ U ~ . ~ ~ - D L - A ~ ~ ~ . ~ ~ ) ]
`(eak-CAD). In order to compare
`ticity values (A€) are related to one lysine residue in the
`the coupling efficacy in conjugate groups of different size,
`the average degree of CAD substitution (E) has been also
`main chain including a whole CAD substituted side chain.
`Due to the relative low CAD content of the conjugates
`calculated and expressed as percent of modified side chains
`(4.5-14.0%), the CD contribution of CAD was found to be
`in the conjugates. No significant differences have been
`negligible and therefore the pattern of the curves reflects
`observed - between the respective DS values of high
`predominantly the secondary structure of conjugates. It
`(DP, = 450) and low (E, = 100) relative molar mass
`should be noted that the CD spectra of the unconjugated
`polypeptides [DS for p0ly[Lys-(cADj-Le~o.~i-~~-Ala2.g4)]
`polypeptides lak and alk correspond to a highly ordered
`helical conformation (231, while ak (22) and eak (23) proved
`(lak-CAD) = 4.5 % and for pOly[LyS-(CADj-Leuo.gs-DL-
`to be completely unordered under identical circumstances.
`Ala.l)] (LAK-CAD) = 5.5 % , or for poly[Lys-(cADj-Proo.g~-
`These conjugates also have optical activity in the 250-
`
`~~-Ala2.94)] (pak-CAD) = 16.0 % and for poly[Lys-(CAD;-
`360-nm absorption range, which is distinct from the amide
`(PAK-CAD) = 15.5% I.
`P ~ O O . ~ ~ - D L - A ~ ~ ~ J ) ]
`chromophores and corresponds only to the structure of
`The choice of carrier polypeptide could also influence
`CAD. Consequently, this region of the CD spectra could
`the composition of the conjugates. The highest coupling
`be used to monitor the presence of the drug and local
`efficacy was obtained with polypeptides containing Pro
`chromophore interaction in the conjugates. CD spectra
`-
`-
`or Glu/~-Glu at the terminal positions - of the side chains
`in the range of 250-360 nm demonstrated the presence of
`(DSpAK = 15.596, DSEAK = 13.8%, DSD-EAK = 14.0%).
`daunomycin moieties in conjugates (Figure 2B). In the
`Under identical conditions CMC-mediated synthesis of
`case of ak-CAD and lak-CAD, the shape and the charac-
`conjugates made of branched polypeptides with N-terminal
`teristic values of the CD curves were similar to that of free
`DL-Ala or Leu/D-Leu led to much lower - DS values (e.g.,
`CAD, indicating no interaction between coupled cis-ac-
`-
`DSLAK = 5.5%, Sa,, = 5.3%, or DSAK = 5.4%). The
`onityldaunomycin groups. In contrast, a significant red-
`shift could be demonstrated in the CD spectrum of eak-
`absolute configuration of the side chain terminal amino
`CAD conjugate as compared to free CAD. The minimum
`residue had no significant influence - on CAD incorporation -
`at 380 nm and the maximum at 450 nm in the CAD
`into the - conjugates studied - (DS,,,
`= 5.5%, DSD.LAK =
`spectrum shifted to 420 nm and to 480 nm, respectively.
`4.6% or DSEAK = 13.8%, DSD-EAK = 14.0%).
`These observations are in agreement with changes detected
`Following purification by gel filtration, HPLC was used
`in the UV spectra of CAD and eak-CAD conjugate (data
`to detect free anthracycline derivatives in the samples of
`not shown).
`polypeptide-CAD conjugates. Quantitative analysis, per-
`Cytotoxicity against Tumor Cells in Vitro. Mod-
`formed on a reversed-phase column (pore size of 300 A,
`ification of daunomycin by substitution of a cis-aconityl
`isocratic elution), indicated no detectable amount of free
`group significantly reduced its cytotoxicity against 791T
`osteogenic sarcoma cells (P < 0.001) compared with that
`daunomycin and the presence of less than 1% of CAD
`could be demonstrated in the conjugates using peak-area
`of the free drug, as shown in Figure 3. Comparison of the
`measurement calibrated with appropriate standards.
`cytotoxicity curves around the 50 5% inhibitory concen-
`Conformation of Branched Polypeptide-CAD
`trations showed that this reduction was about &fold.
`Conjugates. The chiroptical properties of branched poly-
`Conjugation of cis-aconityldaunomycin to a branched poly-
`peptide-CAD conjugates in water solution was studied by
`peptide carrier caused a further reduction in cytotoxicity.
`CD spectroscopy in the wavelength region 190-360 nm.
`This reduction was not significantly influenced by dif-
`The CD curves of ak-CAD,
`lak-CAD, and eak-CAD
`ferences in the size of the branched polypeptide-cAD
`conjugates in PBS at pH 7.3 are shown in Figure 2.
`molecule. Thus, the example in Figure 3 shows a reduction
`of CAD cytotoxicity by approximately 1 order of magnitude
`In the 190-250-nm wavelength region, the circular di-
`following conjugation to ak (conjugate Mw-37700) and a
`chroic spectra indicate the formation of a helical secondary
`similar reduction in the case of AK-CAD (M, 172300) (P
`structure for lak-CAD, but only marginally ordered
`< 0,001). This pattern was consistent using other poly-
`conformation for ak-CAD and eak-CAD conjugates (Figure
`peptides where only the molecular size was changed,
`2A). Highly ordered conformation was also observed with
`
`SANOFI-AVENTIS Exhibit 1040 - Page 52
`
`IPR for Patent No. 8,951,962
`
`

`
`-
`
`80 -
`.$ 60 -
`.-
`5 40-
`0
`0"
`B 20 -
`
`20; 0
`
`0 -
`
`Daunomycin-Branched Polypeptide Conjugates
`
`Bloconjugate Chem., Vol. 3, No. 1, 1992 53
`
`1 0 2
`c
`(pmollml)
`Figure 5. Influence of amino acid configuration of the terminal
`side chain on cytotoxicity of polypeptide-cis-aconityldaunomy-
`D-LAK-CAD, (0) LAK-CAD, (a) D-EAK-
`cin conjugates: (.)
`CAD, (0) EAK-CAD.
`
`' " " ' " I " "".11
`
`1 o 4
`
`1 o 5
`
`-2o-c
`1 o0
`
`' " " " ' I
`
`IO'
`
`' '
`
`"""I
`
`'
`
`' " " " I
`
`1 o3
`
`-204
`1 0 0
`
`'
`
`"
`
`'
`
`"
`
`'
`
`' 1
`
`1 0 '
`
`'
`
`" " " ' r
`
`' " ' " " I
`
`'
`
`' " ' " ' ?
`
`1 o4
`
`' """.I
`
`1 o5
`
`1 o3
`1 o2
`c
`( p m o l h l )
`Figure 3. Molecular size of the polypeptide carrier molecule in
`relation to daunomycin cytotoxicity and comparison with free
`drug: ( 0 ) free daunomycin, (.)
`cis-aconityldaunomycin (CAD),
`(A) &-CAD, (A) AK-CAD.
`
`1
`
`80
`loo
`
`-204
`1 o0
`
`'
`
`' """,
`IO'
`
`'
`
`'
`
`""'.I
`
`'
`
`' " " " <
`
`'
`
`' " " " I
`
`1 o3
`
`1 o 4
`
`'
`
`' ""'I
`
`1 o 5
`
`1 o2
`c
`(pmollml)
`Figure 4. Influence of different side-chain terminal amino acids
`on cytotoxicity of polypeptide-cis-aconityldaunomycin conju-
`gates and comparison with a protein carrier: (w) CAD, (0) eak-
`CAD, (A) ak-cAD, (A) lak-CAD, (X) HSA-CAD.
`Table 11. Cytotoxicity of Daunomycin (Dau),
`cis-Aconityldaunomycin (CAD), and Branched
`PolypeptidecAD Conjugates against 791T Osteogenic
`Sarcoma Cells
`
`compound
`daunomycin
`CAD
`
`compound
`pak-cAD
`PAK-CAD
`aek-CAD
`eak-CAD
`EAK-CAD
`D-EAK-CAD
`
`HSA-CAD
`791T/36-~AD
`
`ICmP
`pmol/mL
`3.3 x 103
`3.0 x 103
`7.2 x 103
`1.2 x 103
`1.2 x 103
`8.1 x 103
`4.0 x 103
`9.6 X lo3
`
`n 0 , a
`pmol/mL
`9.3 X 10'
`7.1 X 10'
`5.3 X lo3
`ak-CAD
`9.6 x 103
`AK-CAD
`1.0 x 104
`alk-CAD
`>6.0 x 103
`lak-CAD
`4.3 x 103
`LAK-CAD
`D-lak-CAD
`>5.0 X IO3
`3.8 x 103
`JFLAK-CAD
`Concentration is expressed in terms of molar daunomycin or CAD
`content of conjugates.
`although the extent of the reduction differed according to
`the amino acid residue in the side chain (Table 11).
`The identity of the terminal amino acid in the side chains
`had a marked influence on CAD cytotoxicity (Figure 4).
`The addition of leucine to produce lak-CAD resulted in
`a similar reduction to that seen with ak-CAD, but the
`addition of glutamic acid to make the conjugate ampho-
`teric (eak-CAD) resulted in a more cytotoxic conjugate.
`The cytotoxicity of eak-CAD wm not significantly different
`from that of CAD alone, while both ak-CAD and lak-CAD
`were significantly less cytotoxic (P < 0.001). A control
`conjugate of CAD prepared with human serum albumin
`(HSA) as a carrier showed activity similar to that of ak-
`CAD and lak-CAD (Figure 4).
`The configuration of the terminal amino acid residue
`had different effects on cytotoxicity, depending on the
`amino acid involved. Replacement of L-leucine with D-
`
`100 1
`
`- 2 0 4
`1 0 0
`
`'
`
`' " " " I
`
`1 0 '
`
`'
`
`' " " " I
`
`'
`
`' " " " I
`
`'
`
`" " ' " 1
`
`1 o 4
`
`' ' " " T
`
`1 o 5
`
`1 o3
`1 o 2
`c
`(pmollml)
`Figure 6. Influence of amino acid sequence of the side chain on
`cytotoxicity of polypeptide-cis-aconityldaunomycin conjugates:
`(A) lak-cAD, (A) alk-CAD, (0) eak-CAD, (a) aek-CAD.
`
`leucine to make pOly[Lys-(cADj-D-Leuo,98-D~-Ala~.~)]
`(D-
`LAK-CAD) had no effect on cytotoxicity compared with
`that of LAK-CAD (Figure 5), but in the case of EAK-CAD
`a similar replacement had a profound effect. EAK-CAD
`was significantly more cytotoxic than LAK-CAD (P < 0.05
`around lo3 pmol/mL), as also observed with the smaller
`relative molar mass variants in Figure 4, but poly[Lys-
`(cADj-~-Gluo.sa-~~-Alaal)l
`(D-EAK-CAD) was much less
`active (Figure 5). Cytotoxicity of D-EAK-CAD was about
`7 times less than that observed with D-LAK-CAD (P <
`0.01 at concentrations above lo3 pmol/mL).
`The importance of the sequence of the amino acids in
`the side chain was also tested (Figure 6). In the case of
`conjugates containing a polycationic polypeptide, lak-cAD
`was compared with alk-CAD. Both compounds had similar
`cytotoxic activity. In contrast, cytotoxic activity of the
`amphoteric CAD-polypeptide conjugates eak or aek was
`found to be significantly different. Eak-CAD was about
`6 times more active than poly[Lys-(cAD~~~-Ala~.~~-Glu~.~)l
`(aek-CAD) (Figure 6). It was concluded that the side-
`chain sequence could have little (in polycationic conju-
`gates) or marked (in amphoteric conjugates) influence on
`cytotoxicity of CAD, depending on the identity of the amino
`acid X involved (Table 11).
`Biodistribution of Branched Polypeptide-CAD
`Conjugates, The blood survival of iodinated alanylated
`polylysine-CAD conjugates following iv administration is
`shown in Figure 7. There was no significant difference in
`the area under the blood-clearance curve (AUCH~) for
`conjugates prepared with either the higher (AK-CAD) or
`lower (ak-CAD) relative molecular mass branched poly-
`peptide. Twenty-four hours following iv injection the
`whole-body survivals for AK-CAD and ak-CAD were 7.5
`f 1.6% injected dose and 7.6 f 0.7% injected dose,
`respectively. The tissue-distribution profiles of these two
`
`SANOFI-AVENTIS Exhibit 1040 - Page 53
`
`IPR for Patent No. 8,951,962
`
`

`
`54 Blocon/agate Chem. Vol. 3. No. 1. 1992
`
`Hudecz et al.
`
`t=0cpm
`%calculated
`
`time ([1)
`
`Figure 7. Blood-clearance profiles of ‘£51-branched polypeptide-
`cAD and HSA-cAD conjugates following iv administration to
`Balb/c mice. AUCH;, calculated from these data are given in
`Table III. Results are expressed as mean :h standard deviation
`for groups of three animals: (0) ak—cAD, (O) AK—cAD, (E1) lak-
`cAD, (I) eak—cAD, (A) HSA—cAD.
`
`100
`
`.2 O
`
`
`
`°/ocalculatedt=0cpm
`
`.1
`
`0
`
`1
`
`2
`
`4
`
`3
`time (h)
`
`5
`
`6
`
`Figure 9. A comparison of the blood clearance of ml-labeled
`EAK—cAD and D-EAK—cAD and of ""51-labeled lak—cAD and D-
`lak—cAD in Balb/c mice following iv administration. AUCo.a.
`calculated from these data are given in Table III. Results are
`expressed as mean :t standard deviation for groups