March 1966
CHELATING COMPOUNDS AS POTEXTIAL BONESEEKERS
stability of the aziridine ring is well d ~ c u n i e n t e d , ' ~ there was some doubt as to whether the relatively high activity observed with compounds IIc and IId was due to the intact molecule or to one of the hydrolysis products. A study of the stability of these two compounds under in vitro testing conditions in phosphate buffer (pH 7.4), containing not more than 4% p-dioxane, indicated that the isopropyl derivative mas stable whereas the phenethyl derivative underwent slight hydrolysis as detected by thin layer chroniatography using ethyl acetate-benzene (1:3j as solvent. Experiments were designed to determine whether the AIAO-inhibitory action of 1-(2-phenethy1)-2-phenylaziridine was due to the intact molecule or a hydrolysis product. Both 2-(2-phenethyljamino-l-phenylethanol and 2-(phenethyl)amin0-2-phenylethanol were tested along with 1-(2-phenethy1)-2-phenylaziridine for in vitro 3IAO inhibition. Although the results indicated that 2-(2-phenethyl)aniino-l-phenylethanol has considerable activity a t 5 X 111 concentration, its ID5, was found to be 9.25 X lop5 M . This indicated that the above amino alcohol, in the aniourits formed during in vitro testing could not be solely responsible for the M A 0 inhibition observed with compound IId. Thin layer chromatograms with a solution of I d (9.25 X 10-6 111)and of IId which gave approxiniately corresponding in vitro activity showed that, whereas the amino alcohol was easily detectable on the plates, the parent aziridine showed only a very faint spot of hydrolysis product after subjecting the two to the in vitro testing conditions. (15) J. E. Early, C. E. O'Rourke, L. B. Clapp, J. 0. Edwards, and B. C. Lawes, J. B m . Chem. Soc., 80, 3458 (1958).
197
It is possible that the inductive eflect of an alkyl substituent would serve to increase the availability of the unshared pair of electrons on the aziridine nitrogen, making it favored as a center of high nucleophilic reactivity. l 6 This does not, however, explain why compound IId is considerably more active than compound IIb. The inductive effect should be about the same from both substituents, which leaves a bulk effect as a possible explanation. The compounds which are most active (IIc and IId) have in conimon a t the 1-position, nonpolar, hydrocarbon moieties which have considerable bulk. Compound I11 which contains a polar group also shows Mho-inhibition activity. However, the observation that IV and T' showed very little activity as compared to I11 leads to the conclusion that even in the series of polar substituents in the 1-positioii, greatest activity is shown by the conipounds that have a bulky substituent. Although 2-amino-1-phenylethanol is a good substrate of i\IAO,l7 no phenethylaniine derivatives containing groups larger than methyl or dimethyl on the nitrogen have been found to be good substrates of IfAO. One may speculate that due to the presence of the phenethyl unit, compounds IC and Id are able to establish their presence near the active site of M A 0 and thus may partially prevent the enzymatic oxidation of the substrate. Here again it would seem that the large hydrocarbon substituent of the I-position is responsible for inhibitory activity. (16) B. Belleau and J. Moran, J . M e d . Pharm. Chem., 6, 215 (1962). (17) N.Weiner, Arch. Bzochem., 91, 182 (1960).
Synthesis of Chelating Compounds t o Be Used as Potential Bone Seekers1>2 G A D s H T A C H E R 3 BND WILLIA3I
TAUB
Department of Pharmaceutical Chemistry, Weizmann Institute of Science, Rehovoth, Israel Received July IO, 1965 A number of new derivatives of imiitodiacetic acid, of the general formula RN+H(CH&OOH)CH2C00- have been prepared for evaluation as potential bone seekers. Radical R stands for a chain containing, inter alia, an oxygen atom belonging to a hydroxy, carbonyl, amide, carboxyl, or ether function. In this way, tridentate or quadridentate chelating agents are formed. These compounds were prepared mainly by introducing the iminodiacetic acid group in toto into different molecules by the action of various alkyl halides on iminodiacetic acid dimethyl ester in nonpolar solvents, followed by saponification and precipitation of the imino acids a t their isoelectric points. In order to study their biological behavior, the imino acids were labeled with one of the radioThe labeling methods were based on exchange reactions, direct halogenation, or recoil halogens F18,BrB2,or labeling. Acid dissociation constants of the imino acids and the chelate stability constants of the corresponding anions with divalent metal ions were determined potentiometrically. With calcium ions, the major metallic constituent of bone mineral, each of the amino acids forms a single stable 1:1 metal chelate, its stability ranging from log KBl = 3 to log Ksl = 5 ( t = 30", p = 0.100). Biological studies have proved that the affinity of the different synthetic imino acids for bone can be correlated with their chelating ability with calcium ions.
Bone seekers as a group comprise a heterogeneous list which by broad definition includes any substance localizing in the ~ k e l e t o n . ~Most of the research (1) This investigation was supported by t h e Israel Atomic Energy Cornmission, Contract 49-04. (2) P a r t of a thesis submitted b y G. Shtacher t o the Senate of the Hebrew University, Jerusalem, 1965, in partial fulfillment of the requirements for the PI1.D. degree. (3) T h e Hospital for Special Surgery, New York, N. Y. (4) P. S. Chen, Jr., A. R. Terepka. and H. C. Hodge, Ann. Rev. Pharmuc o l . . 1, 369 (1961).
work on bone seekers has been concerned with radioactive elements, because of the radiation hazard arising from the deposition of such elements in the skeleton, and because they can be used to study the physiological activities of bone, qualitatively and quantitatively. Bone-seeking have found extensive use in the study of both localized lesions in bone (e.g., fractures and primary or metastatic neoplasms) and metabolic disorders of the skeleton (e.g., Paget's
disciisc) In contrast l o iiiiiieid bone beelier., relatively little is known about orgaiiic boiie seehcrs. Organic compounds which havc, so far, been s1iow-n to deposit in bone iiiiiieral arc (a) aliLuiiii (I) atid rclated anthraquinone dyes nliich staiii IIPW holies by forming calciuiii laltes (the early studies on bone growth using alizarin mere based 011 the red rolor of the slieleton laid down (luring feeding of the dye"), (b) tetracyclines (11) n-hich complex calciuiii in solutioii aiid havc been showii to localize iii bone iiiirieral by ultraviolet fluorescerice t e c h n i q ~ e s ,arid ~ (0) N-(2-carbo~S-~-iodo~)liciiyljiiiiinodiacelic acid (111) ~vliichacciiniulate. i i i The pheiiyl ring ivas used as a carrier for the rndiogrowth areas of bone aiitl n a 5 UlllCCl to a c t ;L- :t x t i v e halogens l>I*, BrX2, or which served as tracers. tc%radmtxte chelating agent? The aromatic curbon-halogen bond, due to it, rela~ ' the live stability, is not readily cleaved in ~ i u o . ~ Of \ :wioii> methods for p r e p ~ ~ i i iiiiiiiodiacetic g acid clcrivativeh," the iiiost frequently used rvas the alky1~CONH, tioii of iiiiiiiodiacetic wid diiiiethyl ester (IT.') l 2 wit I1 cliff creii t alkyl halide- (niostly bromides), followed by 0 HO 0 OH 0 .apoiiificatioii of the, resulting twtiary iriiiiio e I 11 :Itid isolation of thc free iinirio acitih. Thib inethotl w:ii m~~o~cooH)2 fouiid most coiivtmieiit for t lie following re:isoiiL. .rJ
q$yH gqg I
1:s
111
+ L'NH(CH~COOXle)2---+ IT'
The exact niechariisni of localization of coiiipouiid5 1-111 lias not been elucidated completely, but since the feature coiiiino11 to the t h e e orgaiiic bone seekers iiieritioried abovc is their chelating property, it 1121' been assuiiicd that their depo-itiori in bone is the result, of chelate formation with calcium 1011s at the surfacc of bone crystal;. Stability c~o1 rits for thebe chelate, with calcium :uid iiiagiic.iiuii ions I V ~ eI rccordcd a, ~liowvlibelow iti Table I.
RS(CH2C;OOLIe), 1. o v 2 . 11
a
+ KH( CH2COOAIe)L H i
C;H ,CO() -
KT;"' 'CILCOOH
rtiiig iiiaterials, both tlic v:irioub dhy1 Iinlitles arid tlie iiiiiiiodi:icc4c :ic.itl cliiiiethyl estclr, : i r ~ easily obtaiii:iblc 111 :my amount :itid it1 good yidtl(b) l'hc alkylation ofdiniiriodislcetic acid dinictliyl r proceed^ -iiioothly in yield\ of SO-lOO% :+cc.or(liiig t o tlic rextivity of the halogeii. (v) 'L'Ii(1 pi'ogre-. oi tlie alhylatioti ieacw)n c:in btl f d l o n c d eabily 1,- wigliitig thcl Iiydroliulicle qnlt of t lie iiiiiiiodiacetic a('ic1 (11iiidhyl e-ter, ITX . SH(CH2C'OOAIe)3,n hich sep~trat from the rcactio11 I111.) s i ~ t ~ ~ ~ : ~ ttritoti i t i t d prdiwtioii i> cxpcctctl j i i iicutroii-irradiatctl wrriples of lithirini salts uf the :tniiiio acid.. There tritons tlieii rewt with the oxygen a t t r m a i i i the molecules t o yield FIE. The "hot" FIB:itoms randumly substitute hydrogen atonis iii the (11) niolec~les.~ ~ The second is direct production of F'8-labeled compounds from stable organic: fluoro compounds by irradiation with fast nrutroiis in the reartor tlo produre the iiurlear reari ion (.I,
.~ ~
.
!Xi) J i . (:. .Joslii and S. ( ' . Ualif.1. , J . Indian C i ~ ~ r i (24) II. Irving anti J. J. It. F. llaSil\a, J . C h e m . i 2 5 ) 11.Viscontini and H. Kascliig, Heli'. Chtm. Acta, 42, 570 (lW$!). (26) M.Anbar and P. Neta, J . AT^. Cliem. Soc., 84, 2673 (1962).
.
March 1966
UIOLOGICllL ~ C r P I V I T 1 E SO F
CHLOliAhIPHENICOL ANALOGS BY h I I c a o ~ 1 . 4I p-SCH9 > p-I > p-Br m-N02 > p-OCHo > p-C1 > p-i-C3H7 > p-S02CH3> p-NH2, where the p-SCH,, p-S02CH,, and m-N02 analogs are incongruent with the Hansch equation. The individual inhibitory constants, k , , can be used to predict quantitatively the rate of E. coli growth in any admixture of analogs, e.g., k = ko 2ik,[I], and clearly demonstrate that, in this series, dose effects are additive with respect to rates.
-
The interest in correlating substituent constants and partition coefficients with biological activity as evidenced by the recent publications of Hansch arid coworkers2 has shown the vital need for the quantitative and precise evaluation of the biological activity of various subktituted compounds. Only when such data are available can the niodels for such correlations be adequately tested and modified. The complaint'" is legitimate that there is a severe limitation o n the number of adequate examples in the literature where a particular compound has been modified with a, wide variety of substituents and where these comFounds have been tested under a standard set of conditions to yield quantitative results. I n fact, an outstanding exception is consideredza to be a study of the action of chloramphenicol analogs on microorganisms by the serial-dilution method where the accuracy of the determinations was f25%.3 The need for quantification at a higher order of accuracy is readily apparent when parameters are to be correlated in the intermediate ranges of activities, ie., 25-10070 of the most active analog. (1) This is number IT' in a series entitled Kinetics and Mechanisms of Action of hntibiotics on Microorganisms. The previous publications in this series were (a) E. R. Garrett and &I. R. W. Brown. J . P h a r m . Pharmacol., 16, 185T (1963); (b) M. R. W. Brown and E. R. Garrett, J . P h a r m . Sei., 63, 179 (1964); (c) E. R. Garrett a n d G. H. Miller, ibid., 64, 427 (1965). ( 2 ) (a) C. Hansch, R. AI. Muir, T. Fujita, P . P. Maloney, F. Geiger, and BI. Streich, J . A m . Chem. Soc., 85, 2817 (1963); (b) C. Hrtnsch and T. Fujita, ibid., 86, 1616 (1964); IC)C. Hansch and A. R. Steward, J . M e d . Ckem.. 7 , 691 (1964); (d) T. Fujita, J. I a a s a , and C. Hansch. J . A m . Chem. Soc., 86, 5175 (1964); (e) J. Iwasa, T . Fujita, and C . Hansch, J . .Wed. Ckem., 8 , 150 (1965). (3) A l . N. Shemyakin, 11. S. Iiolosov, 11. AI. Levitov, I
