Eut 1 Me:d Chem (1997) 32, 301-309
`© Elsevier, Paris
`
`301
`
`Pyrrolidides: synthesis and structure-activity relationship as inhibitors
`of dipeptidyl peptidase IV
`
`KJL Augustynst, AM Lambeir2, M Borlool, I De Meester2, I Vedernikovai,
`G VanhoofZ, D Hendriks2, S Scharpe2, A Haemersi*
`
`,
`ID~partment of Pharmaceutical Chemistry, UnivfTrsiry of Antwerp (UIA);
`LDepartment of Medical BiochemistY}; University of Antwerp (UIA), Universiteitsplein !, B-2610 Antwerp, Belr:;ium
`~Received I August 1996; accepted I October 1996)
`
`Summary - Dipeptidyl peptidase IV cleaves specifically the peptide bond at the carboxyl side of a proline at the penultimate
`N-terr;nina~ position of a peptide. It is thought to be important for the regulation of biologically active peptides. Moreover, it has
`been 1dentdied as an activation marker ofT-lymphocytes (CD26). Pyrrolidides and thiazolidides are known as reversible inhibitors of
`DPP IV. Several homologues, unsaturated, open and 3-substituted analogues were synthesized in order to determine the stmcture(cid:173)
`activity relationship of the P 1 site. L-lsoleucine was taken as P-2 amino acid. l-(L-lsoleucyl)-3(S)-fluoropyrrolidine is about as active
`as the non-fluorinated compound and behaves as a competitive inhibitor. Other changes decrease or abolish the activity.
`
`d:ipeptidyl peptidase IV 1 pyrrolidide I serine protease I proline peptidase I proteast' inhibitor
`
`Introduction
`
`Due to the unique structure of proline among the
`amino acids, the peptide bond before or after a proline
`residue is relatively resistant to breakdown by com(cid:173)
`mon proteases. Therefore, it is not surprising that
`specific enzymes participate in the cleavage of such
`bonds. These p1nline-specific proteases are supposed
`to play an important role in the regulation of the life(cid:173)
`time of biologically active peptides [l].
`Dipeptidyl peptidase IV (DPP IV. EC 3.4.14.5) is
`a serine exopeprirla<;e that cleaves off N-terminal
`dipeptides specifically at the carboxyl side of proline
`(X-Pro) [2]. DPP IV is expressed quite ubiquitously in
`mammalian tissues. On the epithelial cells of intestine
`and kidney, DPP IV participates in the metabolism and
`uptake of proline-containing peptides [31- ln human
`plasma, DPP IV is responsible for the degradation and
`inactivation of growth hormone-releasing humwne
`[4]. It may also be involved in the metabolism of
`several other biologically active peptides [ 5-71. On
`human T-lymphO<:ytes, DPP IV is assigned to the
`
`*Correspondence and reprints
`
`CD 26 cluster involved in T-cell activation [8]. Further(cid:173)
`more, lymphocytic DPP IV/CD 26 is associated with
`the adenosine deaminase protein, the deficiency of
`which is known to result in severe combined immuno(cid:173)
`deficiency in humans [9]. Development of potent,
`selective and biocompatible inhibitors of DPP TV
`contributes to the unravelling of the physiological
`functions of this intriguing ectopeptidase.
`Most of the inhibitors reported for DPP IV are
`dipeptide analogues. Relevant reference compounds
`of different classes of DPP IV inhibitors are shown in
`figure L The boronic acid transition state analogues
`Ala-boroPro, Val-boroPro and Pro-boroPro (1) are
`very potent reversible inhibitors of DPP IV [10].
`Unfortunately, they have a very short half-life in
`aqueous solution at neutral pH due to a cyclization
`reaction between the free amino group of the P-2
`amino acid and the boronic acid [11]. The same insta(cid:173)
`bility problems can be expected for other transition(cid:173)
`state analogues (such as aldehydes or triftuoromethyl(cid:173)
`ketones)
`that are often used as serine protease
`inhibitors. Dipeptide diphenylphosphonates (2) are
`irreversible inhibitors leading to a phosphorylated
`enzyme [12, 13]. They have a half-life of a few hours
`in plasma. Other ineversible inhibitors are proline
`derived diacylhydroxylarnines
`,l4j. Pyrrolidides
`
`AstraZeneca Exhibit 2151
`Mylan v. AstraZeneca
`IPR2015-01340
`
`Page 1 of 9
`
`

`
`302
`
`(J-R
`
`I
`X
`
`X
`p,"
`Pro
`
`lie
`
`lie
`
`Pro
`
`2
`
`5
`
`4-
`
`5
`
`y
`
`CH2
`
`CH2
`CH2
`s
`CH2
`
`R
`
`B(OH),
`
`P(O)(OC0 He)o
`H
`
`H
`
`H
`
`Fig 1. DPP IV inhibitors.
`
`(3 and S) and thiazolidides (4) are reversible, competi(cid:173)
`tive inhibitors of DPP IV [15]. Substitution at position
`2 with a nitrile group also affords competitive inhibi(cid:173)
`tors [16].
`In order to fully explore the S-1 subsite of the
`enzyme and because of the chemical and biological
`instability of reactive electrophiles, we concentrated
`on poLential inhibitors containing substituted pyrro(cid:173)
`lidines, open pyrrolidines or homologucs thereof at
`the P-1 position (6b-23b) (figs 2 and 3). In this study,
`L-Ile was used at the P-2 position as this amino acid
`proved to be an efficient P2-residue. Furthermore, we
`introduced the constrained L-Leu and L-Ile analogues,
`(24b) and
`L-cyclohexylalanine
`trans-3-methyl-L(cid:173)
`proline (25c) (fig 4).
`
`Chemistry
`
`Boc-Ile was coupled to an appropriate secondary amine
`in the. presence of benzotriazol-1-yloxy-tris(dimethyl(cid:173)
`amino}phosphonium hexaftuorophosphate (BOP) to
`afford compounds 6a-15a. Deprotection with trifluo(cid:173)
`roacetic acid gave final compounds 6b-15h (fig 2).
`Coupling of Boc-He. to 3-hydroxypyrrolidine in the
`presence of BOP was possible without hydroxyl
`protection (16a). Various reactions at the hydroxyl
`function and cleavage of the Boc protecting group
`resulted in 3-substituted pyrrolidines 16b-23b (fig 3).
`Chlorine and fluorine were introduced with tripheny1-
`phosphine/CC14 and diethylaminosulphur trifluoride
`(DAST), respectively. An azide was obtained from a
`tosylate intermediate, prepared with tosylsulphonyl
`chloride and sodium azide. A methoxy group was
`introduced with methyl iodide. Treatment with benzoyl
`chloride afforded the benzoate and a periodate
`ruthenium oxide oxidation gave rise to the corre(cid:173)
`sponding ketone.
`
`Boc-llc
`
`Soc-lie-N
`
`_...R
`
`'-R'
`
`-·-
`
`lie-N _......R
`'-Ft·
`
`6b - 15b
`
`lie-<==>
`
`6b
`
`11•-N8
`
`9b
`
`llo-N8
`
`lio-N.::>
`
`71>
`
`8b
`
`lie-N:=>
`
`10b
`
`11,.-NH{J
`
`!1b
`
`t2b
`
`1Jb
`
`t4b
`
`t5b
`
`lie-N
`
`Fig 2. Synthesis and structures of ring-modified pyrro(cid:173)
`lidides: (a) BOP, piperidine; (b) triftuoroacetic acid_
`
`16o
`
`16b - 23b
`
`~oH
`
`-F'
`
`-F
`
`~c1
`
`-Ns
`
`~oc(o)c0H,
`~ocH3
`=0
`
`16b
`
`17b
`
`18b
`
`19b
`
`20h
`
`21b
`
`22b
`
`23b
`
`~
`~
`~
`~
`
`Fig 3. Synthesis and structure of 3-substituted pyrrolidides:
`(a) trifluoroacetic acid; (b) diethylaminosulfur trifluoride;
`(c) identical to the synthesis ofl7, but starting from the (R)-
`3-hydroxypyrrolidine derivative; (d) P(C6H5h, CCI4 ; (e)
`p-toluenesulphonyl chloride, Et3N; (f) NaN3 ; (g) benzoyl
`chloride, pyridine: (h) 1) NaH: 2) CH31; (i) Ru02 , Nal04 •
`
`Examination of the 'H-NMR spectra indicates thar
`several compounds (lOb, 14b, 15b, ISh, 23b) exist in
`solution as a mixture of two rotamers around the
`amide bond, as is frequently observed for X-proline
`amide bonds.
`
`Page 2 of 9
`
`

`
`0
`
`24b
`
`25c
`
`F'ig 4. Structures of compounds 24b and 25c.
`
`Compound 24b was obtained by hydrogenolysis
`of Z-protected N-L-cyclohexylalanylpyrrolidide 24a,
`obtained hy BOP-mediated coupling of pyrrolidine
`and Z-L-cyclohexylalanine. For the synthesis of 25c,
`trans-3-L-methylproline [17] was Z-protected (25a)
`and coupled to pyrrolidine with a water soluble
`carbodiimide. Compou:-1d 25c was obtained by
`hydrogenolysis.
`
`Biological evaluation
`
`DPP IV was isolated from human seminal plasma as
`described previously [18]. The specific activity of the
`preparation was 35 U/mg. One activity unit is defined
`as the amount of enzyme which catalyzes the conver(cid:173)
`sion of 1 ttmol of substrate per minute at 37 oc. The
`IC50 was detennined using the ftuorogenic substrate
`Gly-Pro-4-methoxy-2-naphthylamide-HCI. The IC:m
`value was defined as the inhibitor concentration which
`caused a 50% decrease in the activity. The error on
`repeated IC~8 determinations was around 20%.
`To determine the type of inhibition and the inhibi(cid:173)
`tion constants, the chromogenic substrate Gly-Pro-p(cid:173)
`nitroanilide was used. The decrease in initial rate with
`inhibitor concentration was fitted with equation (1 ):
`
`(l)
`
`where v, and v0 are the initial rates in the presence and
`absence of inhibitor, K,:arr is the apparent inhibition
`constant at the substrate concentration used, and [1] 0 is
`the total inhibitor concentration. 1n case of competi(cid:173)
`tive inhibition, K, .. PP is defined by equation (2):
`
`(2)
`
`where K, is the substrate-independent, 'true' inhibition
`constant, [S]0 is the initial substrate concentration and
`Km
`is
`the Michaelis constant of the substrate.
`Measured in a separate experiment under the same
`conditions, Km was 0.34 ± 0.02 mM.
`
`303
`
`Results and discussion
`
`The IC50 values for DPP IV inhibition of the com(cid:173)
`pounds prepared are summarized in table I. Analogues
`with diiTerent ring sizes or an open ring structure are
`less active than the parent pyrrolidide 3. The IC50
`values of compounds 3, 4 and 6b-9h show a good
`correlation with the ring size, indicating an optimal
`inhibition with pyrrolidine and thiazolidine: the inhi(cid:173)
`bitory capacity increases from azetidine Sb (270 J.lM)
`and pyrroline 9b (100 ~M) to the optimal five-rings
`pyrrolidine 3 (21 ~M) and thiazolidine 4 (18 J.LM). and
`decreases when larger rings like tetrahydropyridine
`lOb (310 ~M), piperidine 6b (510 ~M) or hexamethyl(cid:173)
`eneimine 7b (2700 ~M) are at the P l position.
`Replacing
`the
`ring with
`acyclic
`substituents
`(12b-15b), which could possibly adopt the pyrroli(cid:173)
`dinc conformation, or taking the amino group out of
`the ring as in 11 b also increase the IC 50 considerably.
`Therefore, we believe that the S-1 subsite of DPP IV
`ideally fits a five-membered saturated ring. This
`corresponds to the high specificity of the enzyme for
`proline among the amino acids. In their study on the
`substrate specificity of DPP IV, Rahfeld et a1 [ 19]
`observed a much higher kcat value fm· azetidine
`compared to five-membered rings, but the Km value,
`reflecting the affinity of the enzyme for the substrate,
`was higher. In agreement, we find higher IC 50 values
`for the inhibitors with a four-membered ring.
`Introduction of a substituent at 3-position of the
`pyrrolidine ring (fig 3) (16b-23h) generally decreased
`the inhibitory activity (table 1). Only a small sub(cid:173)
`stituent such as fluorine, isosteric to hydrogen, is
`allowed. L-Ile-3-fluoropyrrolidide (17b) shows about
`the same activity (27 !lM) as its hydrogen (3, 21 !lM)
`or thia analogue (4, 18 J.LM). Moreover, the stereo(cid:173)
`chemistry of the fluorine atom was not important (17b
`versus 18b).
`The allowance of a highly electronegative fluorine
`atom is surprising in view of a recently published
`model of the active site of DPP IV [20]. By compara(cid:173)
`tive molecular field analysis and molecular modelling,
`the authors predict that the proline specificity of
`DPP IV is mainly caused by recognition of proline by
`the tyrosine side chain near the active serine residue,
`and that in the vicinity of PI a positive or an
`uncharged structure clement is favourable. Our data
`clearly show that the presence of the electronegative
`fluorine does not interfere with binding, suggesting
`that the interaction of PI with the Tyr residue is
`mostly sterically determined. More bulky substituents
`as hydroxy, chloro, azido, methoxy, benzoyloxy and
`oxo are deleterious for the inhibitory activity.
`In addition to L-Ile, the prototype P2 amino acid,
`we synthesised two analogous P2 derivatives: L-cyclo(cid:173)
`hexylalanylpyrrolidide 24b and
`trans-3-methyl-L-
`
`Page 3 of 9
`
`

`
`304
`
`Table I. IC50 values for DPP N inhibitors.
`
`Compound
`
`In 50 mM Tris, pH 8.3
`
`in citrated plasma
`
`6b
`7b
`8b
`9b
`lOb
`llb
`12b
`13b
`14b
`15b
`16b"
`17ba
`18b
`19ba
`20b3
`21h•
`22b
`23b
`24b
`25c
`3b
`4b
`5b
`
`Range (mM) (n)
`
`JC5o (J-LM)
`
`10-0.01
`10-0.01
`10-0.01
`2-0.002
`10--0.01
`10--0.01
`10--0.5
`10--0.5
`10--0.5
`10--0.5
`10-0.01
`10-U.Ol
`l-0.001
`10-0.01
`10 0.01
`10--0.01
`10-0.001
`10-0.01
`1-0.001
`l-0.0()1
`10-Q.Dl
`1-0.01
`1-0.001
`
`(4)
`(4)
`(4)
`(lO)
`(lO)
`(4)
`(5)
`(5)
`(5)
`(5)
`(4)
`(4)
`(10)
`(10)
`(10}
`(10)
`(5)
`(10)
`(10)
`{10)
`(4)
`(7)
`{10)
`
`510 :t 50
`2700 ± 600
`270±40
`100±3
`310 ± 30
`14000 ± 1000
`3l:\OO ±YOU
`>10000
`2200± 400
`3500:!. 600
`740 ± 80
`21 ± 3
`25 ± 2
`6!0 ± 30
`1070 ± 80
`6200 ± 700
`> 10000
`320 ± 30
`180 ± 15
`300 ± 20
`21 ±4
`18 ± 2
`126 ± 9
`
`Ran14e
`
`(n)
`
`10--0.02
`
`2-0.005
`1--0.002
`2-0.005
`
`(5)
`
`(5)
`(5)
`(5)
`
`10--0.02
`(5)
`0.2--0.0005 (5)
`0.2--0.0005 (5)
`10--0.02
`(5)
`(5)
`10 0.02
`
`2-0.005
`(5)
`(5)
`2-0.005
`(5)
`2 0.005
`0.2-0.0005 (5)
`0.2-0.0005 (5)
`1-0.002
`(5)
`
`!C,o
`
`470 ± 180
`
`460 ± 70
`73 ± 7
`201 ± 38
`
`1040 ± 70
`35 ± 8
`22 ±6
`500 ± 100
`490 ±50
`
`380 ± 110
`180 ± 40
`350 ±50
`23 ±4
`30 ± LO
`83 ± 7
`
`The activity versus inhibitor concentration was fitted with the equation of a single saturatable binding site. The errors in the
`table are the standard errors of the fit. n = number of different concentrations used. 3Tested as diastereoisomeric mixture;
`breference compounds (prepared as described) from the literatu:-e [15].
`
`prolylpyrrolidide (25c). trans-3-Methyl-L-proline, which
`can be considered as a ring-constrained analogue of
`L-lle, gives a ten-fold decreased inhibition. L-Cyclo(cid:173)
`hex:ylalanine, with its bulky aliphatic side chain,
`fulfills the proposed requirements for binding to the
`S2 subsite. However, this could not be experimentally
`increase
`confirmed. Summarizing, the lC50 values
`from L-Ile-pyrrolidide (3, 21
`j...lM) via L-Pro-pyrro(cid:173)
`lidide (5, 126 !J.M) to L-cydohexylalanyl p}Tiolidide
`(24b, 180 J...lM)] and trans-3-methyl-L-proline pyrro(cid:173)
`lidide (25c, 300 !fM).
`Determination of inhibition constants revealed that
`the compounds of this study are competitive inhibitors
`of DPP IV This is illustrated in figure 5 for the
`3-ftuoro compound 17b: the apparent inhibition constant
`increases linearly with the substrate concentration and
`the presem.:e of the inhibitor causes an increase in Km
`but leaves the Vm= unatlcctcd. The K; vaLues of a
`representative set of inhibitors (3, 4 and 17b) were
`1.5 :!:: 0.1, 1.8:!:: 0.1 and 2.0 ± 0. l, t-espectively. The
`IC50 values listed in table [, therefore. are apparent
`inhibition constants obtained at a [S] 0 = lOKm (using
`the ftuorogenic substrate). The inhibition constants of
`
`these compounds strictly reflect the affinity of the
`P l-P2 residues for substrate binding sites. Examining
`published data, it is clear that a good affinity for the
`substrate binding site is nut always sufficient to make
`a good substrate or inhibitor. Parameters equally
`important for a catalytic efficiency are the rate of
`acylation, the rate of deacylation, or the ability of the
`substrate to adopt the trans-proline conformation.
`Other types of inhibitors that mimic later steps in the
`catalytic mechanism in their mode of action may have
`a different Pl-P2 preference than the compounds of
`this study. A striking example of this is the relative
`potency of dipeptide-derived diphenyl phosphonate
`esters, where cyclohexylalanine and proline were
`better P2 residues than isoleucine [ 13].
`The potency of our compounds to inhibit the intrin(cid:173)
`sic DPP IV activity present in human plasma is
`comparable to that obtained in assay buffer using
`highly purified human CD26 (table I) [18]. Human
`plasma contains a form of DPP IV which dift'ers from
`CD26 by the lack of adenosinedeaminase binding and
`its high subunit molecular weight [21]. The catalytic
`properties of both forms of DPP IV were very similar
`
`Page 4 of 9
`
`

`
`12
`
`10
`
`:lie
`::!..
`
`Q.6
`Q.
`<lt
`::.::::
`
`4
`
`2
`
`0
`
`1.5
`
`r3'
`3
`0.5 ~
`
`)(
`tU
`E
`>
`
`D
`0
`
`0.005
`[I] (mM)
`
`'---L---..t:---l. ____ L ... 1 ..... J ..
`
`-0.4
`
`-0.2
`
`0
`
`0.2
`0.4
`[S] (mM)
`
`0.6
`
`c.s
`
`Fig 5. Competitive inhibition of DPP IV activity hy
`compound l7b. The apparent inhibition constant increases
`with increasing substrate concentration, as predicted by
`equation (2). The inset shows the effect of the inhibitor
`concentration on the Km (e) and \/max (•) of the sub&trate.
`t1m., is expressed in arbitrary units. Conditions are descri(cid:173)
`bed in the Experimental protocols.
`
`using the chromogenic dipeptide substrate Gly-Pro-p(cid:173)
`nitruanilide. Our results with the inhibitors extend
`this observation to include the Pl specificity of both
`enzyme forms.
`The stability of the compounds in plasma is an
`important advantage for their potential use in biologi(cid:173)
`cal systems.
`
`Experimental prol.oculs
`
`Materials and methods
`
`Amino acids and intermediates were obtained from Bachem
`Feinchemikalien AG, Switzerland. Reagents and solvents were
`from ACROS Chimica or Aldrich. Gly-Pro-4-methoxynaphthyl(cid:173)
`amide and Gly-Pro-p-nitroanilide were from Sigma. Fluor(cid:173)
`escence was measured in a Model RF-5000 fluorimeter from
`Shirnadzu Corporation (Tokyo, Japan), excitation wavelength
`was 340 nm. emissicn wavelength 425 nm.
`Melting points were determined on a digital melting point
`apparatus (Electrothermal) and are uncorrected. Tile lH-~MR
`spectra were determined on a Bruker 300 MHz with retra(cid:173)
`methylsilane as internal standard. Column chromatography was
`perfonned on Fluka silica gel H (5-40 J.Lm) under vacuum.
`
`305
`
`Measurement of DPP IV activity
`
`The enzyme was diluted in reaction buffer ir.Jmediately before
`the experiment, typically around 5-10 rnU/mL. Enzymatic
`reactions were carried out in 50 mM Tris·HCl buffer, pH 8.3
`at 37 oc. Substrate and inhibiton were dissolved in DMSO
`and diluted in the reaction buffer immediately before the
`experiment. Substrate stock solutions V>ere stored at - 20 °C
`inhibitor stock solutions were kept at 4 °C.
`'
`
`Determination of JC50 values
`
`To determine the 1C50, 5 J!L of inhibitor solution (between I(XJ
`and 0.01 mM in DMSO) was incubated with 40 IJ.L of DPP IV
`in reaction buffer (or 40 J!L of freshly frozen citrated human
`plasma) at room temperature for 1 h. The remaining activity
`was determined by adding 5 J!L uf 10 mM Gly-Pru-4-melhuxy(cid:173)
`naphthylamide (25% DMSO), then incubation for 20 min
`at 37 "C. rhe reaction was stopped by addition of 0.5 mL of
`100 rnM citrate pH 4.0 before the fluorescence was measured.
`Measurements were perfmmed in duplicate and corrected for
`background fluore"cence using a blank with exactly the same
`composJtJon except that lbe citrate solution was added before
`the substrate.
`
`Type of inhibition, determination of inhibition corzstarzts
`
`To determine the type of inhibition and the inhibition constants,
`the chromogenic suh;;tratc Gly-Pro-p-nitroarilide was used at
`final concentrations of 0.05, 0.1, 0.2, 0.4, 0.6 and 0.8 mM in
`100 mM Tris buffer, pH 8. At every substrate concentration six
`inhibitor concentrations were used between 0.002 and 0.1 rn:M
`(17b, 3, 4), or between 0.05 and 5 mM (8b)
`
`1-[N-(ren-Buryloxycarbony/)-L·isoleucyl]piperidine 6a
`To a mixture of N-(lerJ-butyloxycarbonyi)-L-isoleucine ( 1.16 g.
`5 rnmol), piperidine (0.43 g, 5 mmol) and BOP (2.43 g,
`5.5 mmol) in DMP (20 mL) W&> added triethylamine (1.5 mL.
`11 mmol). After stirring at room temperature overnight, water
`( 100 mL) was added and the mixture was extracted with
`EtOAc (3 x 50 rnL). The combined organic layer was washed
`witb HCl (2 N, 2 x 25 mL), water (25 mL), NaHC03 (5%.
`2 X 25 mL) and brine (25 mL). The organic layer was dried,
`evaporated and purified by column chromatography (CH2Ch;
`CH1CI2/MeOH, 99: l; CH2Cl2/MeOH, 98:2) yielding 1.48 g
`(4.97 mmol, 99%) of the title comp::mnd as a solid. lH-NMR
`(CDCI3 ) o 0.88 (t, J-= 7.6 Hz, 3H, o-CH,), 0.93 (d, J = 6.8 Hz,
`3H, y-CH 3), 1.00-1.20 (m, lH, y-CH), 1.43 (s, 9H, C(CH 3h),
`1.50-1.80 (m, 8H, y-CH,
`(3-CH. 3-CH2, 4-CH1 , 5-CH2 ),
`3.45-3.65 (m, 4H. 2-CH,, 6-CH,). 4.49 (dd, J-= 6.3 and 8.7 Hz.
`I H, a-CH}, 5.32 (d. J-= 8.7 Hz, lH, NH).
`
`1-(L-lsoteucyl)piperidine trifluoroacetate 6[:,. Gener-al proce(cid:173)
`dure for deprotection with trif/uoroacetic acid
`The Boc-protected compound 6a (1.45 g, 4.87 mmol) was
`treated wirh trifluoroacetic acid (TFA) (10 g) for 1 h m room
`temperature. The title compound wa' obtained in quantitative
`yield after evaporation, cocvaporation with toluene (three times}
`and MeOH (three times). The compound was crystallized from
`Me0HfEt20. Mp 147-149 "C. IH-::"J"MR (DMSO-d6 ) o 0.85 (t,
`J = 7.3 Hz, 3H, .S-CH3 ), 0.94 (d, J = 6.9 Hz, 3H, y-CH3),
`1.00-1.25 (m, LH, y-CH), 1.35-1.70 (m, 7H, y-CH, 3-CH2,
`4-CH,, 5-CH2). L70-l.85 (rn, IH, jj-CH), 3.20-3.65 (m, 4H,
`2-CH2, 6-CH2 ), 4.25 (d, J = 4.9 Hz, :H. a-CH), !:L07 (br s. 3H.
`NHj). Anal C 11H 22N 20·C1 HF,0 2 (C. H. N).
`
`Page 5 of 9
`
`

`
`306
`
`Compounds 7a,7b-16a,l6b were prepared in a similar way.
`
`1-[N-( ren-Buryf oxycarbonyl)-L-isoleucyl/hexame thyleneimine
`Ja. From hexamethyleneirnine; 98% yield as an oil. 'H-NMR
`(CDCI1) o 0.89 (t, J = 7.4 Hz, 3H, 8-CH3), 0.92 (d, J = 6.7 Hz,
`3II. y-CH3), 1.05-1.25 (rn, lH, y-CH), 1.43 (,, 9H, C(CH3) 3),
`150--1.95 (m,
`lOH, y-CH, !3-CH, 3-CH2 , 4-CH2, 5-CH,,
`6-CH,). 3.35-3.75 (m, 4H, 2-CH 2, 7-CH1 ), 4.43 (dd, IH,
`a-CH), 5.21 (d, J"" 8.9 Hz, lH, NH).
`
`1-(L-l.mleucyl)he.xamethyleneimine Jrijlumou.cetaie 7b. From
`7a; crystallized from EtOAc/hexane. IH-NMR {DMSO-d6) 15
`0.85 (t, J = 7.3 Hz. 3H, o-CH,), 0.96 (d, J = 6.9 Hz. 3H,
`y-CH,), 1.05-1.25 (m, III, ''-Clil. 1.35-1.90 (m, IOH, -y-CH,
`!3-CH, 3-CH2, 4-CH,, 5-CH2, 6-CH,). 3.20-3.70 (m. 4H,
`2-CH2, 7-CH2), 4.16 (d. J..., 5.1 Hz, 1H, u-CH), 8.!0 (br s, 3H,
`NH;). Anal C 12H 24N 20·C2HF10, (C, H, N).
`
`1-[N-( tert-Butylo.r:ycarbonyl)-L-isoleuc:vl]azetidine 8a. From
`azetidine; 92% yield as an oiL I H-NMR (CDCI3) o 0.85-1.00
`(m, 6H, 0-CH3, y-CH,), l.05-1.20 (m, lH, y-CH), 1.43 (s, 9H,
`C(CH3) 1 ), 1.50-1.80 (m, 2H, y-CH, 13-CH), 2.20--2.40 (m. 2H,
`3-CI-12 ). 3.90-4.30 (m, 4H. 2-CH2 • 4-CH2 ). 4.38 (m, lH,
`u-CH), 5.30 (d, 1 = ~-7 Hz, lH, NH).
`
`1-(L-lsoieucylja2.etidine trijfunroacewle 8b. From 8a; crystal(cid:173)
`lized from MeOHIEt20. IH-NMR (D20) 1l 1.03 (t, J = 7.3 Hz,
`3H, o-CH,). 1.10 (d, J = 6.9 Hz, 3H, y-CH3), 1.20-1.45 (m. 1 H,
`y-CH), 1.55-1.75 (m, l H. y-CH), 1.95-2.15 (m. I H, [:1-CH),
`2.35-2.60 (rn, 2H, 3-CH2}, 3.97 (d, J = 6.3 Hz, IH, a-CH),
`4.05-1-.30 (m, 2H) and 4.35-4.55 (m, 2H) (2 CH2 , 4 CH 2).
`Anal C 9H, 8N 20•C,HF30 2 (C. H, N).
`
`1-[N-(tert-lJutyloxycurbonyl)-L-isoleucyl]-3-pyrroline 'Ja. From
`3-pyrroline: 99% yield as an oil. tH-NMR (CDCl,) o0.80--l.OO
`(m, 6H. 0-CH3 , y-CH3), 1.05-1.25 (m. 1H, "{-CH), 1.43(s, 9H.
`C(CH 3h), 1.50--1.80 (m, 2H, y-CH, J3-CH), 4.10--4.65 (m, 5H,
`a-CH. 2-CH2• 5-CH,), 5.20 (d, J = 8.3 Hz, lH, NH). 5.75-5.90
`(m. 2H, 3-CH, 4-CH>.
`
`l-(L-lwleucyf)-3-pyrmline
`trijfuoranretntf' 9b. From 9a;
`crystallized from E:OAc/hexanc. Mp 140-142 ·Jc. lH-:-fMR
`(DMSO-d6 ) o 0.87 (t, J = 7.3 Hz, 3H, o-CH3 ), 0.96 (d, J =
`6.9 Hz, 3H, y-CH,), 1.05-1.20 (m, IH, y-CH), 1.40-1.60 (m,
`lH, "{-CH), 1.80-2.00 (m, 1 H, f3-CH), 3.92 (d, J = 6.1 H2, I H,
`a-CH), 4.05--4.15 (m, lH) and 4.15--4.35 (m, 2H) and 4.35-
`4.50 (rn. lH) (2-CH,, 5-CH2), 5.93 (s, 2H, 3-CH, 4-CH), 8.13
`(br s, 3 H. NH~). Anal C 10HtsN,O·C,HF30 2 (C, H, :'-J).
`
`1-[ N-( rerr- Butyloxycarbonyi)-L-isoleucyl j -1,2, 5, 6-te 1 rahydru(cid:173)
`pyridine lOa. From I ,2,5,fi-tetrahydropyridine; 99% yield as
`an oil. 1H-NMR (CDCI3) B 0.88 (t, J = 7.3 Hz, 3H, 5-CH,).
`0.93 (d, 3H, y-CH,). 1.00--1.25 (m, IH, y-CHJ. 1.43 (s, 9H,
`C(CH,h), 1.50-1.60 (m, I H, y-CH), 1.60--1.80 (m, lH, ['1-CH),
`2.10-2.30 (m, 2H, 5-CH,), 3.60--3.85 (m. 2H, 6-CH,), 3.95-
`4.20 (m. 2H, 2-CH2), 4.51 (dd, IH, a-CH), 5.33 (d, J = 8.3 Hz,
`III, NH). 5.60--5.95 (m, 2H, 3-Cll, 4-Cin.
`
`1-(L-lsoleucyl )-1,2,5,6-tetrahydropyridine trijluoroacetate lOb.
`From lOa; CTystallizecl from RtOAclhexane. Mp 103-105 °C.
`IH-NMR (DMSO-de• mixture of cis and trans rotamers) i5 0.85
`(t. J = 7.2 Hz, 3H, 5-CH 3 ), 0.93 (d, J = 6.7 Hz) and 0.96 (d,
`J = 6.7 Hz) OH. y-CH 3), 1.05-1.25 (m, IH. y-CH), 1.35-1.55
`(m, IH, y-CH), 1.70-1.90 (m, IH, 13-CH), 2.00--2.30 (m, 2H,
`
`5-CH2), 340--3.75 {m, 2H, 6-CH2), 3.80--4.15 (m, 2H, 2-CH2),
`4.27 (d, J = 4.2 Hz) and 4.34 (d, J = 4.0 Hz) (IH, o:-CH). 5.65-
`5.95 (m, 2H, 3-CH, 4-CH). 8.15 (br s, 3H, NH~). Anal
`C 11H 20N 20•C2HF10 2 (C, H. N).
`
`N-[N-(teri-Butyluxycarbollyl)-L-isoleucyl)cyclopentylam;ne
`lla. From "yclopentylamine; 99% yield a~ a wlid. lH-NMR
`(CDCl:J .S 0.89 (t, J = 7.5 Hz, 3H, 0-CH,), 0.91 (d, J = 6.8 Hz,
`3H, y-CH,), 1.00--L25 (m, lH. y-CH), 1.43 (s, 9H, C(CH3) 3 ),
`1.50--2.05 (m, IOH, y-CH, 13-CH, 4 X CH2 ), 3.88 (dd, lH,
`a-CH). 4.10--4.25 !m, 1H, CH (cyclopentyl)J, 5.32 (d, J =
`8.0 Hz, lH, NH (Ile)), 6.37 [br s, lH, NH (cyclopcntyl)].
`
`tri]luoroacetate !lb. From
`N-(L-lsoleucyl)cyclopentylamine
`lla; crystallized
`from EtOAc/hexane. Mp 199-202 "C.
`IH-NMR (DMSO-d6 ) o 0.86 (t, J = 7.5 Hz, 3H, 8-CH,), 0.88
`(d, J
`6.8 Hz, 3H, y-CH3 ), 1.00-1.20 (m, IH. y-CH), 1.30-
`1.90 (m, lOll, y-CH. 13-CH, 4 x CH2 i, 3.50 (d, J = 6.1 Hz, lH,
`a-CH), 4.00-4.15 Lm,
`lH, CH (cyclopen:yl)], 8.09 (hr s.
`3H, NH;), 8.3.'> Ld, J = '!.1 Hz, IH. NH (cyclopentyl)J. Anal
`C,,H12N 70·C,HF,O, (C, H, N).
`
`N- [N-( tert-Rutyluxycurbunyl)-L-isuleucyl}dimerhyiamine 12a.
`From dimethylarnine hydrochloride (neutralized in situ with
`one eqmvalcnt ot triethylamine); 96% yield as an oil. IH-NMR
`(CDC},) o 0.88 (t, J = 7.4 Hz, 31-I, 0-CH,), 0.92 (d, J = 6.6 Hz.
`3H, y-CH,), 1.05-1.25 (rn, lH, r-CH), 1.43 [s, 9 H, C(CH3 ) 3J,
`1.45-1.60 (m, lH, y-CH}, 1.60-1.75 (m, IH, p-CH), 2.97 (s,
`3H, NCH,), 3.11 (s, 3 H, NCH3 ), 4.48 (dd, lH, a-CH), 5.26 (d,
`J"' 8.1 Hz. lH, NH).
`
`N-1 L-lsoleucyl)dimethylamine
`12/J. From
`rrifluoroacetate
`12a; crystallized
`from EtOAc/hexane. Mp 137-138 °C.
`1H-NMR (DMSO-d.) 15 0.86 (t, J = 7.3 Hz, 3H. o-CH,), 0.94
`(d, J = 7.0 Hz, 3H, y-CH3 ), 1.00-1.25 (m, 1 H, y-CHj, 1.40--
`1.60 (m, lH. y-CH), 1.70-1.90 (m, 1H. f3-CH). 2.90 (s, 3H,
`NCH3), 3.04 (s, 3H, NCH3 ), 4.22 (d, J = 5.5 Hz, tH, a-CH),
`8.06 (br s, 3H. NHj). Anal C.H 1o;N,O·C2HF30 2 (C. H. N).
`
`13a.
`N-[N-(tert-Buryloxycarbonyi)-L-isoleucyljdiethylamfne
`From diethylamine; 94% yield as an oil. lH-NMR (CDCI 3) o
`0.85-1.00 (m, 6R 8-CH1, r-CH,), 1.05-1.25 (m, lH, y-CH),
`1.12 (l, J = 7.1 Hz, 3H. NCCH3 ), 1.22 (l, J = 7.1 HL, 3R
`NCCH,), 1.43 [s, 9H, C(CH,) 3], 1.50-1.80 (m, 2H, y-CH,
`13-CH), 3.lU 3.25 (m, lH) and 3.30 3.50 (m, 2H) and 3.50-
`3.65 (m, lE-I) CNCH2 ), 4.39 (dd, lii. a-CH), 5.19 id, J = 8.8 Hz:.
`IH,NH).
`
`N-(L-lsoleucyl)diethylamine rriftuoroacetate 13b. From 13a
`as an oiL IH-NMR (DMSO-d6) 0 0.36 (t, J = 7.3 Hz, 3H,
`o-CH3), 0.94 (d, 1 = 6.9 Hz, 3H, y-CH3), 1.05-1.20 (m, IH,
`y-CH), 1.06 (t. J = 7.1 Hz, 3H, NCCH3), l.I3 (t, J = 7.0 Hz,
`NCCH3 ), 1.40-1.60 (m, lH, y-CH), 1.70-1.85 (m, lH, 13-CH).
`3.00--3.60 (m, 4H, NCH2 ), 4.08 (m, lH, a-CH), 8.06 (br s, 3H,
`NHj). Anal C 1oHz:N20.C3HF30 2 (C, H, N).
`
`N-(N-(tert-Butyloxycarbonyl)-L-iso/eucyl]-N-ethyl-N-methyl(cid:173)
`amine 14a. From N-ethylmethylamine; 99% yield as an oil.
`'H-NMR (CDC1 3, mixture of cis and trans rotamers) 5 0.85-
`LOO (m, 6H, .S-CH1 , -y-CH,), 1.05-1.20 (m, lH, r-CH), I.! I (t,
`J-= 7.1 H:t., L5H) and 1.22 (t, J = 7.1 Hz. 1.5H) (NCCH,), 1.43
`[s, 9H, C(CH3 ) 3 ], 1.50--1.65 (m, lH, y-CH), 1.65-1.80 (m, lH,
`13-CH), 2.93 (s, L5H, NCH3 ), 3.07 (s, L5H. NCH3 ), 3.25-3.65
`(m, 2H. NCH2), 4.43 (dd. I H. a-CH), 5.15-5.30 (m, IH, NH).
`
`Page 6 of 9
`
`

`
`triftuoroacetate 14b.
`N-(T-lsoleucyl)-N-ethyl-N-methylamine
`From 14a as an oil. lH-NMR (DMSO-d., mixture of cis and
`trans rotamcrs) S 0.86 (t, J = 7.3 Hz:. JH. o-CII 3 ), 0.95 (d. J
`6.9 Hz, 3H, -y-CH 3), 1.04 (t, J = 7.1 Hz, L8H) and I. 13 (t, J =
`7.1 Hz, L2H) (NCCH 3 }, 1.05-1.20 (m, IH, )'-CH), 1.40-1.60
`(m, lH, y-CH), 1.70--1.90 (m, 1 H, 13-CH), 2.87 (s, 1.211) and
`3.02 (>. L8H) (NCH,), 3.20-3.35 (m, lH) and 3.40--3.55 (m,
`lH) (NCH2 ), 4_17 (m, lH, et-CH), 8.11 (br s, 3H, :--rHr). Anal
`C 9H;,oN20·C2HF30 2 (C, H, N).
`
`N-/N·(tert-Dutyloxycarbonylj-L-isoleucyl]-N-methyl-N-prvpyl(cid:173)
`amine l5a. From N-methylpropylamine; 99% yield as an oil.
`IH-NMR (CDC13 , mixture of cis and trans rotamers) o 0.85-
`1.00 (m, 9H, 0-CH3, y-CH3, NCCCH1), 1.05-1.20 (m. lB.
`y-CH), 1.43 (s, 9 H, C(CH3 ) 3 ), !.50--!.80 (rn, 4H, 'f-CH, ~-CH,
`NCCH,), 2.93 (s, L35H) and 3.07 (s. 1.65H) (NCH,), 3.15-
`3.55 (m, 2H, NCH 2 ), -".45 (dd, IH. <X-CH), 5.15-5.30 (m,
`lii, NH).
`
`N-(L-lso{eucyl)-N-methyl-N-propylamine rrifiuoroacetate 15b.
`From 15a as an oil. IH-NMR (DMSO-d0 , mixture of ci:; and
`trans rotamers) o 0.70--0 90 (m, 6H, 6-CH3 , NCCCH 3), 0.95 (d,
`J = 6.8 Hz) and 0.96 (d. J = 6.8 Hz) (3H. y-CH3), 1.05-1.20 (m.
`lH, y-CH), 1.40-1.65 (m, 3H. y-CH, NCCH2), 1.70--1.90 (m,
`!H, j:i-CH), 2.87 (s, 1.05H) end 3.02 (s, L95H) (NCH3), 3.05
`3.25 (m, I H) and 3.30-3.55 (m, lH) (KCll,), 4.17 (m, 1 H,
`u-CH), 8.08 (br s, 3H, NHj) Anal C 1oHnK20•C.,HF30 2
`(C, H, N).
`
`1-1 N-( tert-Butyloxycarbonyli-L-isofeucylJ-3(R, S)-hydmxypyr(cid:173)
`rolidine 16a. From (R.S}-3-hydroxypyrrolidine. Eluent wa~
`gradually increased to CH2Cizf.l\;le0H (96:4); 65% yield as a
`foam. 'H-NMR (CDCI 3 ) o 0.89 (t, J = 7.5 Hz, 3H, 6-CH,), 0.93
`(d, J = 6.6 Hz, 3H, y-CH,), 1.00--1.25 (11:4 IH, r-CH), 1.42 (s.
`9H, C(CH3 ),), 1.50--1.65 (m. lH, y-CH). 1.65-1.80 (m. lH,
`(3-CH), 1.90--2.15 (m, 2H, 4-CH2 ), 3.40--4.15 (m, 4H, 2-CH2,
`5-CH2), 4.15-4.35 (m, I H, o.-CH), 4.48 (br s, 1H, 3-CH), 5.33
`(d. J = 9.3 Hz) and 5.42 (d. J= 9.9 Hz) (lH, NH).
`
`1-(L-lsoleucyl)-3(R,S)-hydroxypyrrolidine trijluomacetate 16b.
`Frum 16a; crystallized from EtOAe/hexane. Mp 124--127 oc.
`III-NMR (DMSO-d,;) o 0.85 (m, 3H, 8-Cll3). 0.94 (d, J
`6.7 Hz, 3H, -y-CH 3 ), 105-1.25 (m, IH, y-CH), 1.40-1.60 (m,
`lH, y-CH), 1.70-2.00 (m, 3H, (5-CH, 4-CH2), 3.15-3.155 (m,
`4H, 2-CH2, 5-CH,), 3.85-4.05 (m, lH, <X-CH), 4.25-4.40 {m.
`!H. 3-CH), 5.00-5.20 (m, IH, OH). 8.17 (br s, 3H. NH;). Anal
`C 10HzoN20 2·C2HF30 2 (C, H, N).
`
`1 -[N-( tert-Butyloxycarhonyl)-L-isoleucyl]-3-( R, S )fluompyrro(cid:173)
`lidine 17a
`A solution of 16a (0.40 g, 1.33 mmol) in dry 1,2-dichlum(cid:173)
`ethane (15 mL) was treated with DAST (0.35 mL, 2.66 mmol)
`at 0 "C under N,. After stirring for 2 h the reaction mixture was
`poured into a NaHC0 1 solution (5%, 100 mL) and stirred for
`15 min. This miKture was eKtracted with CH2Cl 2 (2 x 100 mL)
`and the combined organic layer was dried, evaporated and
`purified by column chromatography (CH2Ch; CH,Cl,/McOH,
`99:1; CH2Cl2fMeOH, 98:2) yielding 0.27 g (0.89 mmol, 67%)
`of the title compound as an oil. lH-NMR (CDCI3 ) () 0.85-1.00
`(m, 6H, o-CH", y-CH,), 1.05-1.25 (m, lH, )'-CH), 1.43 (s, 9H,
`C(CH 3) 3), 1.50--1.65 (m, !H, y-CII), 1.65-1.80 (m, lii, 13-CH),
`1.90--2.45 (m, 2H, 4-CHJ, 3.40--4.10 (m, 4H, 2-CH2 , 5-CH2 ).
`4.10--4.40 (m, lH, rt-CH), 5.20-5.35 (m, I H, NH), 5.28 (dm,
`J = 52.2 Hz, lH, 3-CH).
`
`307
`
`triftuoroacetate 17b.
`1-(L-lsoleucyl)-3(R,S)-ftuoropyrrolidine
`From 17a according to the general procedure; crystallized from
`McOHfEt 20. lH-NMR (DMSO-r/6) S 0.86 (t, J = 7.2 H:z, 3H,
`o-CH3), 0.93 (m, 3H, y-CH3), 1.05-1.25 (m, 1H, y-CH), 1.40-
`1.6.) (m, IH, y-CH), 1.75-1.95 (m, IH, 13-CII), 2.00-2.30 (m,
`211, 4-CH,), 3.35-4.10 (m, 5H, o.-CH. 2-CH2, 5-CH2), 5.38
`(Lim, J = 52.9 Hz, IH, 3-CH), 8.18 (br s, 3H, NH~). Anal
`C 10H,.,FN20·C,HF30 2 (C. H, N).
`.
`
`1-l N-( lert-Butyloxycarbonyl )-1.-i.mieucylj-_~( S )-jhwropyrm(cid:173)
`lidine 18a. Prepared fr-om 3(R)-hydroxypyrrolidine a-; de(cid:173)
`scribed for 16a and 17a.
`
`18b.
`l·(L-lsoieucyl)-3(SI-fluoropyrro/idine
`trifluoroacetate
`From 18a: c.rystalli:z:ed from EtOAc/hexane. IH-NMR (DMSO(cid:173)
`d6, mixture of cis and trans rota mer'} o 0.87 (t, J = 7.3 Hz, 3H,
`o-CH,), 0.92 (d, J = 6.4 liz) and 0.96 (d. J ~ 6.5 Hz:) (3H,
`)'-CH3), 1.05-1.25 (m, IH, y-CH), 1.45-1.65 (m, lH, y-CH),
`1.7:5-1.95 (m, I H, [)-CH}. 2.05-2.30 (m, 2H. 4-CH,), 3.30--
`4.00 (rn, 5H, o.-CH, 2-CH, 5-CH:J, 5.34 (d, J = 52. l Hz,
`0.55H) and 5.42 (d, J = 52.3 Hz, 0.45H) (3-CH), 8.20 (br s,
`3H. NHi). Anal C 1J{ 1"FN20•C2HF,02 ie, H. N)
`
`1-{N-(tert-Butyloxycarbonyl) L isoleucyli-3r R,S)- chloropyrro(cid:173)
`lidin<: 19a
`A mixture uf 16a (0.81 g, 2.7 rmnol) and triphenylphosphine
`(1.42 g, 5.4 mmol) in CCI4 dry (13 g) was refluxed for 2 h.
`After evaporation and purification by column chmmatography
`(Cli,Ch: CH,CiiMeOH, 99:1). the title compound (O.Gl g.
`1.92 mmol, 71%) was obtained an oil. 'H-NMR (CDCI 3) ()
`0.85-1.00 (m, 6H, 8-CH,, y-CH,), !.05-1.25 (m, IH. y-CH).
`1.43 (s. 9H, C(CH1)~). 1.50-1.65 (m, IH, y-CH), 1.65-1.85 (m,
`lH, fj-CH). 2.10-2.45 (m. 2H. 4-CH,). 3.55-4.35 (m, SH,
`{)(-CH, 2-CH2 , 5-CH2 ), 4.55 (br s. 1 H, 3-CH), 5.