9-amino-4,5-diazafluorene-9-carboxylic acid (Daf), a new Cα,α-disubstituted glycine containing a spatially constrained bipyridine-like ligand for transition metals - Synthesis and evaluation of peptide-coupling conditions at its C- and N-termin

Article abstract of DOI:10.1002/1099-0690(200105)2001:10<1821::aid-ejoc1821>3.0.co;2-9

Acylation of the anion of N-benzyl-4,5-diazafluorene-9-methyleneamine with methyl or benzyl chloroformate, followed by acidic hydrolysis, resulted in 9-amino-4,5-diazafluorene-9-carboxylic acid methyl ester (H-Daf-OMe) and benzyl ester (H-Daf-OBzl), respe

Full text of DOI:10.1002/1099-0690(200105)2001:10<1821::aid-ejoc1821>3.0.co;2-9

FULL PAPER
9-Amino-4,5-diazafluorene-9-carboxylic Acid (Daf), a New Cα,α-Disubstituted
Glycine Containing a Spatially Constrained Bipyridine-Like Ligand for
Transition Metals ؊ Synthesis and Evaluation of Peptide-Coupling Conditions
at its C- and N-Termini
Jean-Paul Mazaleyrat,*[a] Karen Wright,[a] Michel Wakselman,[a] Fernando Formaggio,[b]
Marco Crisma,[b] and Claudio Toniolo[b]
Keywords: α,α-Dialkylated amino acids / Transition metal receptors / Bipyridine ligands / Peptides
Acylation of the anion of N-benzyl-4,5-diazafluorene-9-
methyleneamine with methyl or benzyl chloroformate, fol- with H-Ala-OMe by the acylazide method. Coupling of Boc-
lowed by acidic hydrolysis, resulted in 9-amino-4,5-diaza- Ala-OH at the N-terminus of Daf could also be performed
fluorene-9-carboxylic acid methyl ester (H-Daf-OMe) and by the mixed anhydride method. However, coupling of the
tion afforded Boc-Daf-NHNH2, which was efficiently coupled
benzyl ester (H-Daf-OBzl), respectively. Nα-protection with
Boc2O at 60 °C gave Boc-Daf-OMe and Boc-Daf-OBzl, sa-
ponification or hydrogenolysis of which resulted in complete
decarboxylation. However, hydrazinolysis of the ester func-
crowded Aib residue required the use of Boc-Aib-NCA. Daf,
a new Cα,α-disubstituted glycine, is the first α-amino acid
containing a rigid bipyridine ligand in a totally controlled
spatial disposition relative to the Cα atom.
Introduction
intramolecular luminescence quenching.[9Ϫ24] In this paper,
we wish to report our detailed experimental procedures for
the synthesis of terminally protected derivatives of 9-amino-
4,5-diazafluorene-9-carboxylic acid (Daf) (Figure 1),[25] and
an initial evaluation of their peptide coupling conditions.
Daf is characterized by a 4,5-diazafluorene architecture
as a transition metal receptor[26Ϫ27] and presents new and
interesting features: (i) the metal ligand site is in a totally
rigid and controlled spatial disposition relative to the Cα
atom of the amino acid, in contrast with previously de-
scribed flexible amino acid metal receptors;[9Ϫ24] (ii) Daf
can be inserted into the main chain of a peptide, not into
a side chain, which should impose distinct conformational
constraints on the backbone geometry,[24] (iii) it belongs to
the class of Cα,α-disubstituted glycines, well-known for their
strong conformational preferences and their very strong
tendency to induce β-bends and α/310-helices in
peptides;[28Ϫ35] this would be expected to allow better con-
trol over the spatial organization of these metal receptors
in peptide supramolecular architectures. In this context, we
have previously designed the [20-C-6]-Bip residue, in which
The utility of peptide synthesis for the facile assembly of
combinations of natural amino acids and artificial amino
acids with functionalized side chains, permitting the con-
struction of supramolecular devices and catalysts, has in the
past few years been recognized as an important
strategy.[1Ϫ3] Unnatural amino acids that can bind trans-
ition metals are especially interesting targets for the de novo
design of metalloproteins,[4] as well as peptide-based elec-
tronic devices and molecular switches.[1Ϫ3,5] In this context,
peptides containing a variety of 2,2Ј-bipyridine-type trans-
ition metal receptors,[6Ϫ8] whether covalently attached to
their C- or N-termini, incorporated within the sequence as
side chain-modified α-amino acids, or directly inserted into
the main chain, have been extensively studied and shown to
display supramolecular properties of templated self-organ-
ization, photoinduced intramolecular electron transfer and
a crown ether effector is carried by an axially chiral Cα,α
-
disubstituted glycine.[36] We have also recently exploited one
such scaffold for the synthesis of a peptide-based ‘‘rigid
donor/rigid interchromophore spacer/rigid acceptor’’ sys-
tem.[37]
Figure 1. Structure of the 9-amino-4,5-diazafluorene-9-carboxylic
acid residue (Daf).
[a]
ˆ
´
SIRCOB, ESA CNRS 8086, Bat. Lavoisier, Universite de
Versailles,
Results and Discussion
78000 Versailles, France
Fax: (internat.) ϩ33Ϫ01 39 25 44 52
E-mail: mazaleyrat@chimie.uvsq.fr
Biopolymer Research Centre, CNR, Department of Organic
Chemistry, University of Padova,
35131 Padova, Italy
For the synthesis of Daf, we considered a route involving
acylation of the anion of N-benzyl-4,5-diazafluorene-9-
methyleneamine as the key step. This method had been pro-
[b]
Eur. J. Org. Chem. 2001, 1821Ϫ1829
WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ193X/01/0510Ϫ1821 $ 17.50ϩ.50/0
1821

J.-P. Mazaleyrat et al.
FULL PAPER
posed by DuPriest et al.[38] for acylation of the delocalized (NaHMDS/THF, then ClCOOMe) did not give any amino
anion of N-benzyl-fluorene-9-methyleneamine and was pre- ester 3a after acidic hydrolysis, although benzophenone was
viously applied by us to the preparation of 9-aminofluor- isolated in 40% yield, together with ketone 1 (60%). Acyl-
ene-9-carboxylic acid (Afc) methyl ester.[39Ϫ40]
ation of the delocalized anion of imine 2a (treated with
In this context, 4,5-diazafluoren-9-one 1 (Figure 2) was NaHMDS/THF) by benzyl chloroformate, which was also
first prepared from phenanthroline,[26,41] and condensed attempted in order to obtain a Daf ester function cleavable
with benzylamine. Surprisingly, the reaction did not pro- by hydrogenolysis (vide infra), gave H-Daf-OBzl (OBzl,
ceed in the presence of BF3 catalyst,[42] but did when use benzyloxy) 3b in a slightly higher yield (38%). Here, only
was made of TiCl4 in CH2Cl2 at 0 °C Ϫ as in the case of traces of compound 4 were present and the main isolated
the preparation of N-benzyl-fluorene-9-methyleneamine[38] side product was 9-amino-9-benzyl-4,5-diazafluorene 5
Ϫ producing 4,5-diazafluorenone-benzyl-imine 2a in 68% (11%), presumably originating from alkylation of the anion
yield after crystallization from CH2Cl2/hexane. It was found of 2a by benzyl chloride present in the benzyl chloroformate
to be very important to keep the reaction mixture at 0 °C, as a contaminant.
since at 25 °C a strong, blue-green colour sometimes de-
veloped, with unidentified side products then mostly being
obtained after workup. In the same manner, condensation
of 1 with benzhydrylamine produced the benzhydrylimine
of 4,5-diazafluorenone 2b, in 53% yield. When p-methoxy-
benzylamine was used, however, the corresponding imine
could not be purified without extensive decomposition.
Figure 2. Preparation of N-benzyl- and N-benzhydryl-4,5-diaza-
fluorene-9-methyleneamines 2a and 2b from phenanthroline. (i)
KMnO4; KOH/H2O; 100 °C (ii) PhCH2ϪNH2 or Ph2CHϪNH2;
TiCl4; CH2Cl2; 0 °C.
In the next key step, the delocalized anion obtained after
abstraction of a benzylic proton of imine 2a with the aid
of NaHMDS (sodium hexamethyldisilazane) in THF was
treated with methyl chloroformate (Figure 3). Acylation
presumably occurred on both carbanionic sites, since both
benzaldehyde and ketone 1 were present, in a ca. 2:1 ratio,
after acidic hydrolysis of the reaction mixture. However,
phenylglycine methyl ester resulting from acylation at the
benzylic carbon could not be isolated. In several duplicate
experiments the desired amino ester H-Daf-OMe (OMe,
methoxy) 3a was the only such compound obtained in a
more or less constant yield of 25Ϫ30% after chromato-
graphy, while 9-benzoylamino-4,5-diazafluorene (4), re-
Figure 3. Synthesis of 9-amino-4,5-diazafluorene-9-carboxylic acid
sulting from oxidation of the anion of imine 2a, was always
present as the main side product. Such a relatively low
yield, in comparison with that obtained in the related acyl-
methyl and benzyl esters, and their N-Boc protected derivatives,
from imine 2a. (i) NaHMDS; THF; 0 °C (ii) (1) ClCOOCH3 or
ClCOOCH2Ph; room temp. (2) 1  HCl; room temp.; 2 h (iii)
Boc2O; CH3CN; 60 °C (iv) TFA/CH2Cl2 1:1.
ation
of
the
anion
of
N-benzyl-fluorene-9-
methyleneamine,[39Ϫ40] is probably the result of a decreased
As expected, N-protection of the sterically hindered
reactivity of the delocalized anion, stabilized by the strongly amine function of the amino esters 3a and 3b by a tert-
electron-withdrawing effect of the two nitrogens at the 4,5- butyloxycarbonyl (Boc) group required prolonged heating
positions. Attempts to improve the yield by varying the ex- at 60 °C in the presence of a large excess of Boc2O in
perimental conditions {temperature, reaction time, use of acetonitrile[43Ϫ45] to furnish the fully protected amino acid
other base/solvent systems such as NaH/THF, NaH/ residues Boc-Daf-OMe 7a (73%) and Boc-Daf-OBzl 7b
DMSO, NaHMDS/DMSO or KHMDS (potassium hexa- (87%), respectively. Such treatment was also a convenient
methyldisilazane)/toluene-THF} were unsuccessful. Treat- means for further purification of the Daf amino esters 3a
ment of imine 2b under the same reaction conditions as 2a and 3b, which could easily be recovered from chromato-
1822
Eur. J. Org. Chem. 2001, 1821Ϫ1829

9-Amino-4,5-diazafluorene-9-carboxylic Acid (Daf), a New Cα,α-Disubstituted Glycine
FULL PAPER
graphically purified 7a and 7b after Nα-deprotection in TFA
(trifluoroacetic acid)/CH2Cl2 1:1. In the same manner, the
amine 5 was acylated with Boc2O to give 6 (80%) and reco-
vered by acidic cleavage of the Boc protecting group.
The C-deprotection conditions of 7a and 7b were exam-
ined next. From previous studies we suspected that spon-
taneous decarboxylation of the desired Nα-protected free
amino acid Boc-Daf-OH 8 (Figure 4) would be a problem,
since the analogous 9-hydroxy-4,5-diazafluorene-9-carb-
oxylic acid had been shown to decarboxylate at room tem-
perature and had been impossible to isolate.[46] Decarb-
oxylation had also previously been observed, but only to a
certain extent, during saponification of the ester function
of the related 9-tert-butyloxycarbonyl-aminofluorene-9-
carboxylic acid methyl ester, Boc-Afc-OMe, as well as dur-
ing coupling of the N-protected amino acid Boc-Afc-
OH.[39Ϫ40] Saponification of the ester function of 7a in 1 
aqueous NaOH/MeOH at room temperature, followed by
acidic hydrolysis of the reaction mixture, resulted in 9-tert-
butyloxycarbonylamino-9-methoxy-4,5-diazafluorene
Figure 4. Hydrolysis, hydrogenolysis and hydrazinolysis of the ester
functions of Boc-Daf-OMe 7a or Boc-Daf-OBzl 7b. (i) (1) 1 
NaOH; MeOH; room temp. (2) Hϩ(ii) H2/Pd-C; MeOH; room
temp. (iii) spontaneous decomposition in CDCl3 solution (iv)
H2NNH2,H2O; MeOH; room temp.
9
(43%) and ketone 1 (41%) as the only products, with no
trace of the desired Boc-Daf-OH 8. In a duplicate experi-
ment, saponification of 7b under similar experimental con-
ditions produced compound 9 (13%), 9-tert-butyloxycar-
bonylamino-4,5-diazafluorene 10 (15%) and ketone 1 (21%)
as the only products. The different product distributions in
the two experiments are not surprising, since we have ob-
served in control experiments that the decarboxylated prod-
uct 10 is very easily oxidized to 1 (such decomposition oc-
curs in CDCl3 solution, albeit slowly). As it is reasonable
to assume that the mechanism of formation of compound
9 involves decarboxylation of the COOϪ function of the
Boc-Daf-OϪ Naϩ salt, followed by oxidation and reaction
with methanol, it appears that all reaction products arise
from decarboxylation of Boc-Daf-OϪ Naϩ, or Boc-Daf-
OH, or both. The presence of a benzyl ester function in 7b
made its C-deprotection by hydrogenolysis under neutral
conditions (Pd/C; MeOH) possible, but complete decarb-
oxylation was again observed and compound 10 was ob-
tained in 100% yield (crude).
same experimental conditions failed to give the desired Boc-
Daf-Aib-OMe. Presumably, activation of Daf through its
acylazide is not efficient enough to allow a fast acylation of
Aib to compete with the decarboxylation reaction.
Figure 5. Coupling at the C- and N-termini of Daf. (i) (1)
iPr(CH2)2ONO; HCl; DMF; Ϫ40 °C (2) H-Ala-OMe·HCl; DIEA;
DMF; Ϫ40 °C to room temp. (ii) (1) Boc-Ala-OH; NMM; EtOC-
OCl; THF; Ϫ10 °C (2) H-Daf-OMe; CH2Cl2; Ϫ10 °C to room
temp. (iii) H-Daf-OBzl; Boc-Aib-NCA (excess); DIEA; THF;
60 °C.
As coupling of an amino acid residue at the C-terminus
of Daf is a prerequisite for its incorporation into an internal
position in a peptide fragment, we turned to the synthesis
of the N-protected amino hydrazide of Daf, the key pre-
cursor for the exploitation of the acylazide method.[47] In-
terestingly, treatment of 7a with a large excess of hydrazine
For coupling of Ala at the N-terminus of Daf, we consid-
hydrate in methanol at room temperature afforded Boc- ered the mixed anhydride[49] and the symmetrical anhyd-
Daf-NHNH2 11 in 91% yield, with only traces of the de- ride[50] methods, previously shown to be much more effici-
carboxylation product 10.
Conversion of the hydrazide 11 to the corresponding acy- carbodiimide]/HOBt
lazide Boc-Daf-N3 by the HonzlϪRudinger method[48] and method[51] for the coupling of Ala at the N-terminus of
in situ acylation of a large excess of alanine methyl ester Cα,α-diphenylglycine (Dph),[52] Afc[39Ϫ40] and other Cα,α
ent than the EDC [N-ethyl-NЈ-(3-dimethylaminopropyl)-
(1-hydroxy-1,2,3-benzotriazole)
-
(H-Ala-OMe) by the azide was successful in producing Boc- disubstituted glycines.[53Ϫ54] Therefore, an excess of the
Daf-Ala-OMe 12 (Figure 5) in reasonable yield (64%), ac- mixed anhydride Boc-Ala-OCOEt, prepared in situ, was
companied only by small quantities of the decarboxylation used to check the acylation of 3a, resulting without any
product 10 and ketone 1. However, coupling of the hy- particular difficulty in the desired dipeptide Boc-Ala-Daf-
drazide 11 with the more hindered H-Aib-OMe (Aib, α- OMe 13. In situ acylation of 3b by Aib using an excess of
aminoisobutyric acid) by the acylazide method under the the symmetrical anhydride (Boc-Aib)2O was also at-
Eur. J. Org. Chem. 2001, 1821Ϫ1829
1823

J.-P. Mazaleyrat et al.
FULL PAPER
(0.040Ϫ0.063 mm) (Merck) respectively, with the following eluent
systems: 2.5% MeOH/97.5% CH2Cl2 (I); 5% MeOH/95% CH2Cl2
(II); 7.5% MeOH/92.5% CH2Cl2 (III); 10% MeOH/90% CH2Cl2
(IV); 50% EtOAc (ethyl acetate)-50% CH2Cl2 (V). UV light
(254 nm) was used for all compounds for viewing spots after thin
layer chromatography (TLC).
tempted, but failed. However, efficient coupling of Aib at
the N-terminus of Daf could be achieved with the aid of
the protected N-carboxyanhydride Boc-Aib-NCA,[55] which
reacted with 3b in THF at 60 °C to furnish the dipeptide
Boc-Aib-Daf-OBzl 14 in excellent yield (86%).
1
It is noteworthy that the H NMR spectra of the Daf
4,5-Diazafluoren-9-one (1): According to the procedure of Hender-
son et al.,[26] a hot solution of KMnO4 (64.5 g) in water (1 L) was
added dropwise over ca. 2 h to a magnetically stirred, boiling solu-
tion of phenanthroline monohydrate (25 g) and KOH (13 g) in
water (1.3 L). After addition was complete, the solution was re-
fluxed for 1 h, filtered hot through paper and left at room temper-
ature overnight. The yellow crystals were filtered, thoroughly
washed with water and air dried. The crystals from five identical
experiments were combined (yield 27.20 g) and dissolved in boiling
water (2 L). The boiling, clear solution was concentrated to 1.5 L
and left at room temperature overnight. The resulting crystals were
filtered, washed with water and air dried (yield 24.13 g, 21%). M.p.
215Ϫ216 °C (ref.[26] M.p. 212Ϫ213 °C; ref.[41] M.p. 215Ϫ216 °C).
Rf ϭ 0.43 (II). Ϫ 1H NMR (CDCl3): δ ϭ 8.79 [dd, J ϭ 5.0 Hz and
1.2 Hz, 2 H, ArH3H6], 7.98 [dd, J ϭ 7.5 Hz and 1.2 Hz, 2 H,
ArH1H8], 7.35 [dd, J ϭ 5.0 Hz and 7.5 Hz, 2 H, ArH2H7]. Ϫ 13C
NMR (CDCl3): δ ϭ 189.5 (CϭO), 163.3, 155.1, 131.5, 129.3,
124.7 (CAr).
derivatives and peptides with the N-terminal Boc-Daf se-
quence (compounds 7a, 7b, 11, 12), in CDCl3 at room tem-
perature, displayed the same striking feature as previously
observed for the Afc derivatives,[39Ϫ40] namely the trans
(anti)-cis (syn) CO-NH isomerization of the urethane (car-
bamate) moiety, which resulted in broadened and split Boc
CH3 and Daf NH singlets. Such isomerization was not ob-
served for the peptides 13 and 14, with amide functions
rather than carbamate moieties at the Daf N-termini.
Conclusions
This study has shown that Daf synthons, readily obtained
(although in relatively low yield) by treatment of the deloc-
alized anion of the N-benzyl Schiff base of 4,5-diazaflu-
oren-9-one with methyl or benzyl chloroformate, may suc-
cessfully be subjected to peptide coupling at both the N-
and the C-termini. Because of the high instability of the
protected amino acid Boc-Daf-OH, which spontaneously
decarboxylated under both basic and neutral conditions
and could not be isolated, coupling at the C-terminus could
only be achieved by the acylazide method, using the hy-
drazide Boc-Daf-NHNH2. This method was shown to be
efficient for acylation of unhindered proteinogenic amino
N-Benzyl-4,5-diazafluorene-9-methyleneamine (2a): Benzylamine
(8.2 mL; 75 mmol) was added to a magnetically stirred solution
(under argon) of ketone 1 (3.64 g, 20 mmol) in CH2Cl2 (100 mL).
The solution was cooled to 0 °C (ice-water bath) and a solution of
TiCl4 (1.4 mL; 12.7 mmol) in CH2Cl2 (15 mL) was added dropwise
over ca. 0.5 h. The resulting pale yellow, milky suspension was
stirred under argon at 0 °C for an additional hour, then rapidly
filtered over Celite in a Büchner funnel (water aspirator), the pre-
cipitate being washed with CH2Cl2 (2 portions of 100 mL), then
acids such as Ala, but failed for acylation of Aib. On the with Et2O (diethyl ether) (2 portions of 100 mL). The filtrate, which
contained a small amount of fluffy precipitate, was concentrated in
vacuo at 40 °C to ca. 100 mL, filtered again through cotton wool,
then diluted with hexane (50 mL) and concentrated to ca. 60 mL
(removing most CH2Cl2), resulting in crystallization. The mixture
was left in a refrigerator overnight, the crystals were filtered in a
Büchner funnel, washed with several portions of hexane/CH2Cl2
4:1 (100 mL) and air dried, to give 3.69 g (68%) of pure imine 2a
as white crystals. M.p. 124Ϫ128 °C (decomp). Rf ϭ 0.37 (II). Ϫ 1H
NMR (CDCl3): δ ϭ 8.78 and 8.76 [2 dd (dt-like), J ഠ 5.1 Hz and
1.4 Hz, 2 H, ArH3H6], 8.26 and 8.20 [2 dd, J ഠ 7.5 Hz and 1.2 Hz,
other hand, both Ala and Aib could be coupled in reason-
able yield at the Daf N-terminus, the latter through its pro-
tected N-carboxyanhydride. Synthesis of Daf-rich peptides
incorporating Ala, Gly and Aib and conformational ana-
lysis of these in solution is currently being investigated.[56]
It will be of interest to compare the conformational behavi-
our of Daf peptides and Afc peptides, which Ϫ according
to our previous results[39Ϫ40] and those of Lombardi et
al.[57] Ϫ mostly adopt an extended (C5) conformation with
only a modest helical tendency. Possible conformational 2 H, ArH1H8], 7.52 [m, 2 H, ArH (Ph)], 7.42 [m (t-like), 2 H,
ArH2H7], 7.33 [m, 3 H, ArH (Ph)], 5.42 [s, 2 H, PhCH2N]. Ϫ 13C
NMR (CDCl3): δ ϭ 161.5 (CϭN), 159.1, 158.7, 152.6, 152.2, 139.2,
134.4, 130.0, 128.7, 127.6, 127.2, 126.7, 124.3, 123.5 (CAr), 57.3
changes induced by metal complexation of Daf peptides
will also be examined.
(PhCH2N). Ϫ C18H13N3 (271.308): calcd. C 79.68, H 4.83, N 15.49;
found C 79.52, H 4.85, N 15.26. More imine 2a was present in the
mother liquor (by TLC), but was very difficult to separate from
Experimental Section
unidentified side products either by further crystallization or by
flash chromatography. Other experiments under identical experi-
mental conditions gave yields of between 50% and 70% after crys-
tallization. In initial experiments, in which the reaction mixture was
allowed to warm to ca. 25 °C under argon for 1 h, a strong blue
colour sometimes appeared and 2a, which in this case was present
in only a minor proportion relative to the side products, was isol-
ated in much lower yield.
General: Melting points were determined by means of a capillary
tube immersed in an oil bath (Tottoli apparatus, Büchi) with a final
temperature raise of 3 °C/min and are uncorrected. Ϫ 1H NMR
and 13C NMR spectra were recorded at 300 MHz and 75 MHz,
respectively, the solvent (CDCl3 or CD3OD) being used as internal
standard (δ ϭ 7.27 or 3.31 for 1H; δ ϭ 77.00 or 49.00 for 13C).
Splitting patterns are abbreviated as follows: (s) singlet, (d) doublet,
(t) triplet, (q) quadruplet, (m) multiplet. Ϫ The optical rotations
were measured with an accuracy of 0.3%, in a 1 dm thermostatted
cell. Ϫ Analytical TLC and preparative column chromatography
N-Benzhydryl-4,5-diazafluorene-9-methyleneamine (2b): Benzhydryl-
amine (3.22 mL; 18.7 mmol) was added to a magnetically stirred
solution of ketone 1 (0.91 g, 5 mmol) in CH2Cl2 (25 mL) under
were performed on
1824
F 254 Kieselgel and on Kieselgel 60
Eur. J. Org. Chem. 2001, 1821Ϫ1829

9-Amino-4,5-diazafluorene-9-carboxylic Acid (Daf), a New Cα,α-Disubstituted Glycine
FULL PAPER
argon. The solution was cooled to 0 °C (ice-water bath) and a solu-
tion of TiCl4 (0.344 mL; 3.13 mmol) in CH2Cl2 (5 mL) was added
dropwise over 1 h. The resulting pale yellow-green, milky suspen-
sion was stirred under argon at 25 °C for 45 min, then rapidly
filtered over Celite in a Büchner funnel, the precipitate being
washed with CH2Cl2 (2 portions of 25 mL) and then Et2O (50 mL).
which readily crystallised from acetonitrile/Et2O, furnishing 0.073 g
(2.6%) of pure 4 as white crystals from which an analytical sample
(0.054 g) was obtained after a further crystallisation from meth-
anol; (ii) 0.707 g (30.4%) of amino ester 3a, pure by NMR, but
contaminated by a dark brown greenish tar. Crystallisation from
acetonitrile/Et2O furnished an analytical sample as pale yellow-
The clear filtrate was concentrated to ca. 25 mL in vacuo at 40 °C brown plates, but with a decreased yield. It was found more con-
and hexane (5 mL) was added in portions. Concentration in vacuo
resulted in a turbid solution with no crystallization. More hexane
(20 mL) was added and the resulting white precipitate was filtered
out in a Büchner funnel, washed with several portions of hexane/
CH2Cl2 3:1 (100 mL) and air dried, to give 2.23 g of a white pow-
der. Dissolution of this solid in CH2Cl2 (75 mL) resulted in some
decomposition, giving a turbid purple solution, which was stirred
in the presence of activated charcoal and filtered through paper.
The pale yellow solution obtained was concentrated to ca. 5 mL in
vacuo at 40 °C, and Et2O (50 mL) was added in portions. Crys-
tallisation occurred at room temperature within 1 hour and was
accelerated by addition of hexane (20 mL). The crystals were fil-
tered, washed with several portions of hexane/Et2O 1:2 (60 mL)
and air dried (yield 0.798 g). More crystals were obtained after con-
centration of the mother liquor, to give a total of 0.917 g (53%) of
pure imine 2b as white crystals. M.p. 185Ϫ190 °C (dec.). Rf ϭ 0.41
venient to proceed directly to the next step (Nα-Boc protection),
purify the resulting fully protected compound 7a by chromato-
graphy and deprotect the amino group of 7a to give back 3a if
necessary (vide infra).
Six duplicate experiments on similar scales (3Ϫ13 mmol of 2a) gave
32%, 24%, 26%, 21%, 29% and 27% yields of 3a after chromato-
graphy, always as a dark-brown glass (NMR pure) that slowly crys-
tallized. Initial experiments on a smaller scale (0.25Ϫ0.5 mmol of
2a) gave lower yields (ca. 7Ϫ10%), all of which were similar
whether NaH/THF, NaH/DMSO or NaHMDS/DMSO were used
as base/solvent systems. In the same manner, treatment of a solu-
tion of 2a (0.271 g, 1 mmol) in THF (20 mL) with a solution of 0.5
 KHMDS in toluene (4.5 mL; 2.25 mmol), followed by addition
of 2 mL (3.37 mmol) of a solution of ClCOOMe (0.75 mL)/THF
(5 mL)/K2CO3, using the same experimental procedure and workup
as above, gave 0.067 g (28%) of NMR-pure 3a after chromato-
graphy.
1
(II). Ϫ H NMR (CDCl3): δ ϭ 8.76 [dd, J ϭ 5.0 Hz and 1.5 Hz, 1
H] and 8.72 [dd, J ϭ 5.0 Hz and 1.3 Hz, 1 H] [ArH3H6], 8.31 [dd,
J ϭ 7.5 Hz and 1.5 Hz, 1 H] and 8.20 [dd, J ϭ 7.9 Hz and 1.3 Hz,
1 H] [ArH1H8], 7.50 [m (d-like), 4 H, ArH (Ph)], 7.36 [m (t-like), 5
H, ArH (Ph) and ArH2H7], 7.26 [m (t-like), 3 H, ArH (Ph)], 6.71
[s, 1 H, Ph2CHN]. Ϫ 13C NMR (CDCl3): δ ϭ 161.5 (CϭN), 159.0,
157.6, 152.7, 152.1, 143.6, 134.6, 133.8, 130.4, 128.7, 127.3, 126.4,
126.7, 124.2, 123.4 (CAr), 69.4 (Ph2CHN). Ϫ C24H17N3 (347.400):
calcd. C 82.97, H 4.93, N 12.10; found C 82.71, H 4.98, N 11.79.
Treatment of a solution of imine 2b (0.174 g, 0.5 mmol) in THF
(10 mL) with a solution of 1  NaHMDS in THF (1 mL; 1 mmol),
followed by addition of 1 mL (1.8 mmol) of
a solution of
ClCOOMe (0.7 mL)/THF (5 mL)/K2CO3, using the same experi-
mental procedure and workup as above, followed by preparative
TLC of the crude product, gave 0.036 g (40%) of benzophenone,
0.055 g (60%) of ketone 1 and several minor unidentified side prod-
ucts, but no trace of 3a. Similar results were obtained with
NaHMDS/DMSO as base/solvent system.
Acylation of the Anions of the Imines 2a and 2b: A solution of 2a
(2.612 g, 9.64 mmol) in anhydrous THF (175 mL) was magnetically
stirred under argon at 0 °C (ice/water bath) and a solution of 1 
NaHMDS in THF (21 mL; 21 mmol) was added by syringe. The
resulting dark red-brown solution was stirred at room temperature
for 1 h, then cooled to 0 °C. Then, 20 mL (33 mmol) of a solution
of ClCOOMe (3 mL) in THF (20 mL), stirred under argon in the
presence of a large excess of powdered K2CO3 for a few min imme-
diately previously, was added by syringe. The resulting solution,
which turned from dark red-brown to dark green within ca. 15 min,
was stirred under argon at room temperature overnight, and cooled
again to 0 °C. A solution of 1  HCl (225 mL) was then added.
The resulting solution was stirred for no more than 2 h at room
temperature and rapidly extracted with CH2Cl2 (3 portions of
100 mL) and then with Et2O (100 mL). The organic phase was
dried over MgSO4, filtered and evaporated in vacuo at 40 °C, to
give 1.72 g of crude neutral residue which showed two main UV-
positive spots on analytical TLC (eluent II), corresponding to ke-
tone 1 (Rf ϭ 0.45) and benzaldehyde (Rf ϭ 0.80) in a ratio of ca.
1:3.5 by 1H NMR (other runs gave ratios 1:benzaldehyde of ca. 1:2
to 1:3). The aqueous acidic phase was made basic by the addition
of a large excess of NaHCO3 (in portions) with magnetic stirring
and extracted with CH2Cl2 (3 portions of 100 mL). The organic
phase was dried over MgSO4, filtered and evaporated in vacuo at
40 °C, to give 1.56 g of crude basic residue which on analytical
TLC (eluent II) showed a main UV-positive spot corresponding
to the desired 9-amino-4,5-diazafluorene-9-carboxylic acid methyl
ester 3a and several minor spots corresponding to 9-benzoylamino-
4,5-diazafluorene 4, ketone 1, and other unidentified products. This
mixture was chromatographed on a 3 ϫ 65 cm column of silica gel
with eluent (II) to give: (i) 0.221 g of a dark brown, glassy oil
Treatment of a solution of imine 2a (1.084 g, 4 mmol) in THF
(75 mL) with a solution of 1  NaHMDS in THF (10 mL;
10 mmol), followed by addition of 10 mL (14.7 mmol) of a solution
of ClCOOCH2C6H5 (8 mL; 50% in toluene)/THF (8 mL)/K2CO3,
using the same experimental procedure and workup as above, gave
after column chromatography of the basic extraction product on
silica gel with eluent (I): (i) 0.479 g (37.6%) of NMR-pure 9-am-
ino-4,5-diazafluorene-9-carboxylic acid benzyl ester 3b as a brown-
orange glass which slowly crystallized; and (ii) 0.123 g (11.2%) of
NMR-pure 9-amino-9-benzyl-4,5-diazafluorene 5 as a yellow glass
which slowly crystallized. Traces of compound 4 together with
other unidentified impurities were observed by analytical TLC. As
in the case of 3a, it was found convenient to proceed directly to the
Nα-Boc protection step of the obtained samples of 3b and 5 for
further chromatographic purification of the resulting compounds
7b and 6, respectively, with subsequent deprotection of the amino
group to give back 3b and 5 (vide infra).
Nα-Boc Protection: Boc2O (0.534 g, 2.45 mmol) and CH3CN
(20 mL) were added to a sample of amino ester 3a (0.266 g,
1.104 mmol), obtained as described above in 32% yield after chro-
matography, pure by NMR but contaminated with dark-brown tar
(vide supra). The mixture was magnetically stirred on a water bath
at 60 °C (complete dissolution required ca. 15 min), the reaction
progress being monitored by analytical TLC. After 6 h, more
Boc2O (0.256 g, 1.17 mmol) was added and the solution was stirred
at 60 °C for 16 h, then for a further 16 h after addition of a third
quantity of Boc2O (0.248 g, 1.02 mmol). The solution was evapor-
ated to dryness in vacuo and the residue purified by chromato-
Eur. J. Org. Chem. 2001, 1821Ϫ1829
1825

J.-P. Mazaleyrat et al.
FULL PAPER
graphy on a 3 ϫ 60 cm column of silica gel with eluent (I), to give
0.275 g (73%) of pure methyl 9-tert-butyloxycarbonylamino-4,5-di-
azafluorene-9-carboxylate (7a) as a white solid (overall yield from
2a: 23%). Crystallisation of an aliquot from CH2Cl2/Et2O gave an
analytical sample as white crystals. Several other duplicate experi-
ments gave similar yields.
Ϫ C19H15N3O2 (317.334): calcd. C 71.91, H 4.76, N 13.24; found
C 71.89, H 4.82, N 13.24.
9-Benzoylamino-4,5-diazafluorene (4): White crystals (MeOH).
1
M.p. 239Ϫ242 °C. Rf ϭ 0.24 (II). Ϫ H NMR (CDCl3): δ ϭ 8.58
[dd, J ϭ 4.9 Hz and 1.3 Hz, 2 H, ArH3H6], 8.02 [m(d-like), 2 H,
ArH Bz], 7.93 [dd, J ϭ 7.6 Hz and 1.3 Hz, 2 H, ArH1H8], 7.53 [m,
3 H, ArH Bz], 7.21 [dd, J ϭ 4.9 Hz and 7.6 Hz, 2 H, ArH2H7],
7.11 [d, J ϭ 9.0 Hz, 1 H, NH], 6.47 [d, J ϭ 8.7 Hz, 1 H, Ar2CHN].
In the same manner, the sample of NMR-pure amino ester 3b
(0.479 g, 1.51 mmol) obtained in 37.6% yield from 2a as described
above, was treated with Boc2O (0.494 g, 2.26 mmol) in CH3CN
(25 mL). The solution was magnetically stirred at 60 °C for 6 h.
More Boc2O (0.494 g, 2.26 mmol) was then added and the solution
was stirred at 60 °C for 17 h. More Boc2O (0.329 g, 1.51 mmol)
was once again added, and the solution was stirred at 60 °C for
24 h. The solution was evaporated to dryness in vacuo and the
residue was chromatographed on a 2.3 ϫ 52 cm column of silica
gel with eluent (I), to give 0.548 g (87%) of pure benzyl 9-tert-bu-
Ϫ
13C NMR (CDCl3): δ ϭ 168.3 (CϭO), 157.9, 150.8, 139.2, 133.4,
133.1, 132.1, 128.7, 127.4, 123.4 (CAr), 50.9 (Ar2CHN). Ϫ ESIϩ
MS; m/z (relative intensity): 597 (100) [2M,Na]ϩ; 575 (32) [2M,H]ϩ;
310 (25) [M,Na]ϩ; 288 (83) [M,H]ϩ. Ϫ C18H13N3O (287.308): calcd.
C 75.24, H 4.56, N 14.63; found C 74.71, H 4.56, N 14.59.
9-Amino-9-benzyl-4,5-diazafluorene (5): White crystals (Et2O). M.p.
1
171Ϫ173 °C. Rf ϭ 0.18 (II). Ϫ H NMR (CDCl3): δ ϭ 8.63 [dd,
J ϭ 5.0 Hz and 1.3 Hz, 2 H, ArH3H6], 7.61 [dd, J ϭ 7.7 Hz and
1.5 Hz, 2 H, ArH1H8], 7.22 [dd, J ϭ 5.0 Hz and 7.7 Hz, 2 H,
ArH2H7], 7.18 [m, 3 H, ArH Bzl], 6.95 [m, 2 H, ArH Bzl], 3.16 [s,
2 H, CH2 Bzl], 2.32 [s (broad), ഠ2 H, NH2]. Ϫ 13C NMR (CDCl3):
δ ϭ 156.8, 150.2, 145.0, 135.7, 131.5, 130.4, 127.8, 126.9, 122.9
(CAr), 61.9 (Ar2CHN), 46.1 (CH2 Bzl). Ϫ C18H15N3·0.2 H2O
(276.927): calcd. C 78.06, H 5.61, N 15.17; found C 78.25, H 5.59,
N 15.03.
tyloxycarbonylamino-4,5-diazafluorene-9-carboxylate (7b) as
a
white solid (overall yield from 2a: 33%). Crystallisation from a
highly concentrated CH2Cl2 solution with addition of Et2O/hexane
1:1 gave an analytical sample (white crystals; 0.508 g).
Treatment of the amine 5 (0.143 g, 0.523 mmol) with Boc2O
(0.165 g, 0.75 mmol) in CH3CN (10 mL) at room temperature for
22 h gave a mixture containing ca. 50% (by TLC) of the starting
amine. More Boc2O (0.165 g, 0.75 mmol) was added and the solu-
tion was stirred at 60 °C for 9 h. It was evaporated to dryness in
vacuo and the residue was purified by chromatography on a 2.3 ϫ
49 cm column of silica gel with eluent (I), to give 0.156 g (80%) of
pure 9-tert-butyloxycarbonylamino-9-benzyl-4,5-diazafluorene (6)
as a colourless glass. Crystallisation from Et2O/hexane furnished
an analytical sample (white crystals; 0.114 g).
9-tert-Butyloxycarbonylamino-9-benzyl-4,5-diazafluorene (6): White
crystals (Et2O/hexane). M.p. 194Ϫ196 °C. Rf ϭ 0.37 (II). Ϫ 1H
NMR (CDCl3): δ ϭ 8.65 [dd, J ϭ 5.0 Hz and 1.5 Hz, 2 H,
ArH3H6], 7.68 [m (broad), 2 H, ArH1H8], 7.22 [dd, J ϭ 5.0 Hz and
7.7 Hz, 2 H, ArH2H7], 7.12 [m, 3 H, ArH Bzl], 6.80 [m, 2 H, ArH
Bzl], ഠ 5.7 [s (very broad), 1 H, NH], 3.33 [s (broad), 2 H, CH2
Bzl], ഠ1.0 [s (very broad), 9 H, Boc CH3]. Ϫ 13C NMR (CDCl3):
δ ϭ 157.4 (CAr), 154.7 (CϭO Boc), 150.3, 143.0, 134.3, 131.4,
130.5, 127.9, 127.2, 122.7 (CAr), 80.2 (OϪC Boc), 62.9
[Ar2C(Bzl)N], 44.5 (CH2 Bzl), 27.7 (CH3 Boc). Ϫ C23H23N3O2
(373.438): calcd. C 73.97, H 6.21, N 11.25; found C 73.75, H 6.16,
N 11.14.
Nα-Boc Deprotection: Acidolysis of the Boc protecting group of
compound 6 (0.096 g, 0.26 mmol) was accomplished by treatment
with TFA (2 mL)/CH2Cl2 (2 mL) at room temperature for 4 h. The
solution was evaporated to dryness in vacuo at 25 °C. The residue
was dissolved in CH2Cl2, the solution was extracted with 5%
NaHCO3, dried over MgSO4, filtered and evaporated in vacuo at
40 °C to give 0.061 g (87%) of pure amine 5 as a white solid.
Recrystallisation from Et2O gave an analytical sample.
Methyl
9-tert-Butyloxycarbonylamino-4,5-diazafluorene-9-carb-
oxylate (7a): White crystals (CH2Cl2/Et2O). M.p. 192Ϫ196 °C.
Rf ϭ 0.28 (II). Ϫ 1H NMR (CDCl3): δ ϭ 8.76 [dd, J ϭ 4.9 Hz and
1.5 Hz, 2 H, ArH3H6], 8.2Ϫ7.6 [m (very broad), 2 H, ArH1H8],
7.32 [dd, J ϭ 4.9 Hz and 7.7 Hz, 2 H, ArH2H7], 6.25 and 5.76 [s
(very broad), 1 H, NH from anti and syn conformers], 3.61 [s
(broad), 3 H, OCH3], 1.41 and 0.88 [s (very broad), 9 H, Boc CH3
from anti and syn conformers]. Ϫ 13C NMR (CDCl3): δ ϭ 170.0
(CϭO Daf), 158.5 (CAr), 155.1 (CϭO Boc), 151.4, 139.3, 132.4,
130.4, 123.7 (CAr), 81.0 (OϪC Boc), 65.6 (C α), 53.6 (OCH3), 28.0
(CH3 Boc). Ϫ ESIϩ MS; m/z (relative intensity): 364 (97) [M,Na]ϩ;
342 (100) [M,H]ϩ. Ϫ C18H19N3O4 (341.356): calcd. C 63.33, H
5.61, N 12.31; found C 63.17, H 5.52, N 12.59.
Treatment of 7a (0.034 g, 0.1 mmol) as above gave 0.017 g (72%)
of pure 3a as a pale yellow solid. In the same manner, 7b (0.152 g,
0.36 mmol) gave 0.095 g (82%) of pure 3b as a pale yellow solid.
Recrystallization from Et2O gave an analytical sample.
Methyl 9-Amino-4,5-diazafluorene-9-carboxylate (3a): Pale yellow
crystals (CH3CN/Et2O). M.p. 156Ϫ160 °C. Rf ϭ 0.18 (II). Ϫ 1H
NMR (CDCl3): δ ϭ 8.70 [dd, J ϭ 4.9 Hz and 1.5 Hz, 2 H,
ArH3H6], 7.91 [dd, J ϭ 7.7 Hz and 1.5 Hz, 2 H, ArH1H8], 7.31 [dd,
J ϭ 4.8 Hz and 7.7 Hz, 2 H, ArH2H7], 3.58 [s, 3 H, OCH3], 2.63
[s (broad), ഠ3 H, NH2]. Ϫ 13C NMR (CDCl3): δ ϭ 172.2 (CϭO),
157.9, 151.2, 142.0, 131.5, 123.7 (CAr), 65.6 (C α), 53.2 (OCH3). Ϫ
C13H11N3O2 (241.242): calcd. C 64.72, H 4.60, N 17.42; found C
64.49, H 4.59, N 17.46.
Benzyl
9-tert-Butyloxycarbonylamino-4,5-diazafluorene-9-carb-
oxylate (7b): White crystals (CH2Cl2/Et2O/hexane). M.p. 159Ϫ161
1
°C. Rf ϭ 0.30 (II). Ϫ H NMR (CDCl3): δ ϭ 8.73 [d, J ϭ 4.9 Hz,
2 H, ArH3H6], 8.2Ϫ7.6 [m (very broad), 2 H, ArH1H8], 7.26 [dd,
J ϭ 4.9 Hz and 7.7 Hz, 2 H, ArH2H7], 7.22 [m broad), 3 H, ArH
Bzl], 7.02 [m (very broad), 2 H, ArH Bzl], 6.21 and 5.78 [s (very
broad), 1 H, NH from anti and syn conformers], 5.05 [s (broad), 2
H, OCH2Ph], 1.39 and 0.87 [s (very broad), 9 H, Boc CH3 from
Benzyl 9-Amino-4,5-diazafluorene-9-carboxylate (3b): Pale yellow
crystals (Et2O). M.p. 125Ϫ127 °C. Rf ϭ 0.24 (II). Ϫ 1H NMR
(CDCl3): δ ϭ 8.71 [dd, J ϭ 4.9 Hz and 1.5 Hz, 2 H, ArH3H6], 7.84
[dd, J ϭ 7.7 Hz and 1.5 Hz, 2 H, ArH1H8], 7.24 [dd, J ϭ 4.9 Hz anti and syn conformers]. Ϫ 13C NMR (CDCl3): δ ϭ 168.9 (CϭO
and 7.7 Hz, 2 H, ArH2H7], 7.22 [m, 3 H, ArH Bzl], 6.98 [m, 2 H,
Daf), 158.1 (CAr), 155.0 (CϭO Boc), 151.0, 139.0, 134.4, 132.5,
ArH Bzl], 5.03 [s, 2 H, OCH2Ph], 2.41 [s (broad), ഠ2 H, NH2]. Ϫ 130.3, 128.1, 127.4, 123.3 (CAr), 80.6 (OϪC Boc), 67.8 (OCH2Ph),
13C NMR (CDCl3): δ ϭ 171.2 (CϭO), 157.7, 150.9, 141.7, 134.6,
65.4 (C α), 27.8 (CH3 Boc). Ϫ C24H23N3O4·0.2 H2O (421.051):
calcd. C 68.46, H 5.60, N 9.98; found C 68.09, H 5.46, N 9.78.
131.3, 128.2, 128.0, 127.3, 123.4 (CAr), 67.3 (OCH2Ph), 65.6 (C α).
1826
Eur. J. Org. Chem. 2001, 1821Ϫ1829

9-Amino-4,5-diazafluorene-9-carboxylic Acid (Daf), a New Cα,α-Disubstituted Glycine
FULL PAPER
Hydrolysis, Hydrogenolysis and Hydrazinolysis of the Ester Func-
tions of (7a) and (7b). ؊ (a) Hydrolysis: A solution of 7a (0.015 g,
[d, J ϭ 4.2 Hz, 2 H, ArH3H6], 8.03 [d, J ϭ 7.5 Hz, 2 H, ArH1H8],
7.43 [dd, J ϭ 4.9 Hz and 7.0 Hz, 2 H, ArH2H7], 5.79 [s, 1 H,
0.04 mmol) in MeOH (2 mL) and 1  NaOH (0.06 mL) was stirred Ar2CHN], 1.51 [s, 9 H, Boc CH3]. Ϫ 13C NMR (CDCl3): δ ϭ 157.7
at room temperature for 20 h, then acidified by addition of an ex-
(CAr), 155.0 (CϭO Boc), 150.7, 139.3, 132.7, 123.2 (CAr), 80.4
cess of 0.1  HCl (25 mL) and rapidly extracted with CH2Cl2 (two (OϪC Boc), 52.1 (Ar2CHN), 28.2 (CH3 Boc). Ϫ ESIϩ MS; m/z
portions of 25 mL). The CH2Cl2 solution was washed with water,
dried over MgSO4, filtered and evaporated in vacuo at 40 °C. The
residue, which showed two spots on analytical TLC, was purified
by chromatography on a preparative TLC plate of silica gel with
eluent (II), to give 0.0054 g (43%) of pure 9-tert-butyloxycarbonyl-
amino-9-methoxy-4,5-diazafluorene (9) and 0.0030 g (41%) of ke-
tone 1. In a duplicate experiment, a solution of 7b (0.042 g,
0.1 mmol) in MeOH (5 mL) and 1  NaOH (0.15 mL) was stirred
at room temperature for 3 h, then acidified by addition of an excess
of 0.5  HCl (25 mL) and treated as above, to give 9 (0.0040 g,
13%), 1 (0.0038 g, 21%), 9-tert-butyloxycarbonylamino-4,5-diaz-
afluorene 10 (0.0041 g, 15%) and another unidentified compound
(0.0024 g), as the only products after preparative TLC on silica gel,
again with no 9-tert-butyloxycarbonylamino-4,5-diazafluorene-9-
carboxylic acid being detected. The N-Boc-protected amine 10 was
found to decompose slowly in CDCl3 solution, resulting in up to
(relative intensity): 306 (100) [M,Na]ϩ; 284 (19) [M,H]ϩ. Ϫ
C16H17N3O2 (283.320): calcd. C 67.82, H 6.05, N 14.83; found C
67.91, H 6.25, N 14.67.
9-tert-Butyloxycarbonylamino-4,5-diazafluorene-9-carboxylic Acid
Hydrazide (11): Pale yellow solid (crude). M.p. 228Ϫ230 °C. Rf ϭ
0.08 (II). Ϫ 1H NMR (CDCl3): δ ϭ 8.34 [d (broad), 2 H, ArH3H6],
8.20 [s (broad), 1 H, CONH hydrazide], 7.92 [d (broad), 2 H,
ArH1H8], 7.23 [dd, J ϭ 4.9 Hz and 7.7 Hz, 2 H, ArH2H7], 6.8Ϫ6.2
[s (very broad), 1 H, NH Daf], 3.97 [s (broad), 2 H, NH2 hy-
drazide], 1.30 and 0.74 [s (very broad), 9 H, Boc CH3 from anti
and syn conformers]. Ϫ 13C NMR (CDCl3): δ ϭ 168.1 (CϭO Daf),
158.1 (CAr), 153.6 (CϭO Boc), 151.1, ഠ140, 131.5, 123.6 (CAr),
α
80.4 (OϪC Boc), 65.4 (C ), 27.9 (CH3 Boc). Ϫ ESIϩ MS; m/z
(relative intensity): 364 (100) [M,Na]ϩ; 342 (20) [M,H]ϩ. Ϫ
C17H19N5O3 (341.362): calcd. C 59.81, H 5.61, N 20.52; found C
59.76, H 5.61, N 20.54.
1
ca. 50% conversion into ketone 1 (by TLC and H NMR) after 6
days at room temperature. Such decomposition also occurred when
10 was absorbed on silica gel for several days, although it could be
purified on TLC plates of silica gel without significant problems.
Coupling at the C-Terminus of Daf: A suspension of Boc-Daf-
NHNH2 11 (0.392 g, 1.15 mmol) in DMF (15 mL) was heated for
a few min, until a clear solution was obtained. The solution was
magnetically stirred under argon and cooled to ca. Ϫ50 ° C, the
flask being sealed with a rubber septum. A solution of 3.8  HCl
in EtOAc (2.5 mL; 9.5 mmol) was then slowly added by syringe.
The resulting clear pale yellow solution was stirred at Ϫ45 °C for
10 min and isoamyl nitrite (0.230 mL; 1.71 mmol) was added by
syringe. The resulting solution was stirred at ca. Ϫ40 °C for 0.5 h,
and a cold solution of HCl·H-Ala-OMe (1.606 g, 11.5 mmol) and
diisopropylethylamine (DIEA) (4.2 mL; 24 mmol) in DMF
(10 mL) was rapidly added by syringe. The resulting solution was
stirred under argon, warming from Ϫ40 °C to room temperature
overnight, and evaporated in vacuo at 40 °C. The pale yellow glassy
residue was dissolved in EtOAc (100 mL) and H2O (100 mL). The
separated aqueous phase was extracted with EtOAc (100 mL). The
organic phase was washed with H2O (3 portions of 50 mL), dried
over MgSO4, filtered and evaporated in vacuo. The residue was
dissolved in EtOAc. The solution was then concentrated in vacuo
to a very small volume (ca. 1 mL), and Et2O (ca. 15 mL) was ad-
ded. Crystallisation occurred at room temperature. The white crys-
tals of pure Boc-Daf-Ala-OMe 12 were filtered, thoroughly washed
with Et2O and air dried (yield: 0.168 g). More crystals were ob-
tained from the filtrate after evaporation in vacuo, dissolving the
residue in EtOAc, concentration of the solution to ca. 1 mL, dilu-
tion with Et2O (ca. 15 mL) and standing at room temperature for
2 h and then in a refrigerator overnight. The clear supernatant solu-
tion was collected by pipette and the remaining crystals were tritur-
ated in 10 mL of hexane/Et2O 4:1, filtered, washed with 50 mL of
hexane/Et2O 4:1, and air dried (yield: 0.096 g). The combined fil-
trates were evaporated to dryness in vacuo and the residue was
purified by chromatography on a preparative TLC plate of silica
gel with eluent (V) to give more 12 (0.040 g, total amount: 0.304 g,
64.1%) as well as compounds 10 (0.018 g, 5.5%) and 1 (0.028 g,
13.5%). A duplicate run gave slightly different yields of 12 (54.6%),
10 (29.5%) and 1 (3%).
(b) Hydrogenolysis: 10% Pd/C (0.050 g) was added to a solution
of 7b (0.090 g, 0.21 mmol) in MeOH (50 mL). The mixture was
hydrogenated in a Parr apparatus at room temperature for 4 h, fil-
tered through paper and evaporated in vacuo at 40 °C, to give crude
10 (0.065 g, 100%) as the single reaction product by TLC and
NMR. Trituration in Et2O/hexane gave an analytical sample as a
white solid.
(c) Hydrazinolysis: Hydrazine hydrate (2.0 mL; 41.3 mmol) was ad-
ded to a solution of 7a (1.238 g, 3.63 mmol) in MeOH (85 mL).
The solution was stirred at room temperature for 17 h, then evapor-
ated in vacuo at 30 °C. The residue was repeatedly dissolved in
MeOH and the solution evaporated in vacuo at 30 °C, until most
of the excess of hydrazine was removed. The residue was then dis-
solved in eluent (II) (100 mL) and the solution purified by chroma-
tography on a 3 ϫ 40 cm column of silica gel with eluent (II),
followed by (IV), to give 0.047 g (4.6%) of decarboxylation product
10 and 1.133 g (91.5%) of pure 9-tert-butyloxycarbonylamino-4,5-
diazafluorene-9-carboxylic acid hydrazide 11 as a pale yellow solid.
Duplicate runs gave isolated products 10:11 with similar yields
(7%:88% and 2%:89%).
9-tert-Butyloxycarbonylamino-9-methoxy-4,5-diazafluorene (9): Pale
yellow solid (crude). M.p. 137Ϫ141 °C. Rf ϭ 0.25 (II). Ϫ 1H NMR
(CDCl3): δ ϭ 8.75 [dd, J ϭ 4.9 Hz and 1.5 Hz, 2 H, ArH3H6], 8.09
[m (broad), 2 H, ArH1H8], 7.32 [dd, J ϭ 4.9 Hz and 7.7 Hz, 2 H,
ArH2H7], 5.56 [s, 1 H, NH], 2.99 [s, 3 H, OCH3], 1.26 [s, 9 H, Boc
CH3]. Ϫ 13C NMR (CDCl3): δ ϭ 158.3 (CAr), 153.4 (CϭO Boc),
152.0, 138.4, 132.2, 123.8 (CAr), 89.6 (Ar2CNO), 80.7 (OϪC Boc),
51.9 (OCH3), 28.0 (CH3 Boc). Ϫ ESIϩ MS; m/z (relative intensity):
336 (100) [M,Na]ϩ; 314 (27) [M,H]ϩ. Ϫ C17H19N3O3 (313.346):
calcd. C 65.16, H 6.11; found C 65.06, H 6.89.
9-tert-Butyloxycarbonylamino-4,5-diazafluorene (10): White solid
(hexane/Et2O). M.p. 161Ϫ166 °C. Rf ϭ 0.22 (II); 0.15 (V). Ϫ 1H
The hydrazide 11 (0.085 g, 0.25 mmol) was also treated under the
NMR (CDCl3): δ ϭ 8.61 [d, J ϭ 4.6 Hz, 2 H, ArH3H6], 7.89 [d, same experimental conditions and workup as above, except that
J ϭ 7.6 Hz, 2 H, ArH1H8], 7.22 [dd, J ϭ 4.8 Hz and 7.6 Hz, 2 H,
HCl·H-Aib-OMe (0.154 g, 1 mmol) was used instead of HCl·H-
ArH2H7], 5.81 [d, J ϭ 9.2 Hz, 1 H, Ar2CHN], 5.23 [d, J ϭ 9.0 Hz, Ala-OMe. Preparative TLC of the crude reaction product showed
1 H, NH], 1.49 [s, 9 H, Boc CH3]. Ϫ 1H NMR (CD3OD): δ ϭ 8.64
the presence of 1, 10 and other unidentified compounds, but the
Eur. J. Org. Chem. 2001, 1821Ϫ1829
1827

J.-P. Mazaleyrat et al.
FULL PAPER
desired Boc-Daf-Aib-OMe could not be isolated from any of the
chromatographic fractions.
Ϫ ESIϩ MS; m/z (relative intensity): 435 (100) [M,Na]ϩ; 413 (96)
[M,H]ϩ. Ϫ C21H24N4O5 (412.434): calcd. C 61.15, H 5.87; found
C 61.31, H 6.11.
Coupling at the N-Terminus of Daf: N-Methylmorpholine (NMM)
(0.024 mL; 0.22 mmol) was added to a solution of Boc-Ala-OH
(0.041 g, 0.21 mmol) in THF (1 mL), cooled to Ϫ10 °C (ice/salt
bath). EtOCOCl (0.019 mL; 0.20 mmol) was then added. The reac-
tion mixture was stirred under argon at Ϫ10 °C for 5 minutes and
a solution of H-Daf-OMe 3a (0.017 g, 0.071 mmol) in CH2Cl2
(1 mL) was added. The reaction mixture was stirred while warming
from Ϫ10 °C to 0 °C over 1 h, then at room temperature for 5 h.
EtOAc (100 mL) was added and the organic solution was success-
ively washed with 0.5  HCl (2 portions of 50 mL), H2O (50 mL),
5% NaHCO3 (2 portions of 50 mL) and H2O (2 portions of
50 mL), dried over MgSO4, filtered and evaporated in vacuo at 40
°C. The residue was purified on a preparative TLC plate of silica
gel with eluent (IV), to give 0.015 g (52%) of pure Boc-Ala-Daf-
OMe 13 as a pale yellow solid.
Boc-Aib-Daf-OBzl (14): White crystals (CH2Cl2/Et2O). M.p.
1
189Ϫ190 °C. Rf ϭ 0.33 (III). Ϫ H NMR (CDCl3): δ ϭ 8.71 [d,
J ϭ 4.8 Hz, 2 H, ArH3H6], 8.11 [s (broad), 1 H, NH Daf], 7.97 [d,
J ϭ 7.4 Hz, 2 H, ArH1H8], 7.24 [m, 5 H, ArH2H7 and 3 ArH Bzl],
7.02 [m, 2 H, ArH Bzl], 5.06 [s, 2 H, OCH2Ph], 4.94 (s, 1 H, NH
Aib), 1.44 [s, 9 H, Boc CH3], 1.34 [s, 6 H, Aib CH3]. Ϫ 13C NMR
(CDCl3): δ ϭ 174.9 and 168.9 (CϭO Aib and Daf), 158.3 (CAr),
154.9 (CϭO Boc), 151.4, 138.9, 134.7, 132.5, 128.3, 128.5, 127.7,
123.6 (CAr), 80.5 (OϪC Boc), 68.0 (OCH2Ph), 65.4 (C α Daf), 56.7
α
(C
Aib), 28.3 (CH3 Boc), 25.3 (CH3 Aib). Ϫ C28H30N4O5
(502.552): calcd. C 66.91, H 6.01, N 11.15; found C 66.88, H 5.85,
N 11.17.
[1]
N. Voyer, Topics Current Chem. 1996, 184, 1Ϫ37.
N. Voyer, J. Lamothe, Tetrahedron 1995, 51, 9241Ϫ9284.
M. J. I. Andrews, A. B. Tabor, Tetrahedron 1999, 55,
11711Ϫ11743.
L. Regan, Annu. Rev. Biophys. Biomol. Struct. 1993, 22,
257Ϫ281.
A solution of Boc-Aib-OH (0.0183 g, 0.09 mmol) and EDC
(0.0086 g, 0.045 mmol) in CH3CN (0.5 mL) was stirred at room
temperature for 1 h and H-Daf-OBzl 3b (0.0095 g, 0.03 mmol) was
added. The mixture was heated at 60 °C for 18 h, then evaporated
in vacuo. Analytical TLC of the crude reaction product showed
only one UV-positive spot, corresponding to the starting amino
ester 3b, with no trace of the desired coupled product 14.
[2]
[3]
[4]
[5]
For a recent review of transition metals as switches, see: L.
Fabbrizzi, M. Licchelli, P. Pallavicini, Acc. Chem. Res. 1999,
32, 846Ϫ853.
[6]
Boc-Aib-NCA (0.897 g, 3.92 mmol) was added to a solution of H-
Daf-OBzl 3b (0.207 g, 0.65 mmol) and DIEA (0.115 mL;
0.65 mmol) in THF (3.7 mL). The solution was heated at 60 °C
for 24 h, then concentrated in vacuo. The residue was dissolved in
CH2Cl2, washed successively with 0.5  HCl, H2O, 5% NaHCO3
and H2O, dried over MgSO4, filtered and evaporated in vacuo at
40 °C. The residue was purified by chromatography on a 1 ϫ 35 cm
column of silica gel with eluent (III), to give 0.281 g (86%) of Boc-
Aib-Daf-OBzl 14 as a white solid, which was recrystallised from
CH2Cl2/Et2O to give colourless crystals.
E. C. Constable, Adv. Inorg. Chem. 1989, 34, 1Ϫ63.
E. C. Constable, P. J. Steel, Coord. Chem. Rev. 1989, 93,
205Ϫ223.
F. C. J. M. Van Veggel, W. Verboom, D. N. Reinhoudt, Chem.
Rev. 1994, 94, 279Ϫ299.
B. Imperiali, S. L. Fisher, J. Am. Chem. Soc. 1991, 113,
8527Ϫ8528.
M. Lieberman, T. Sasaki, J. Am. Chem. Soc. 1991, 113,
1470Ϫ1471.
B. Imperiali, S. L. Fisher, J. Org. Chem. 1992, 57, 757Ϫ759.
M. R. Ghadiri, C. Soares, C. Choi, J. Am. Chem. Soc. 1992,
114, 825Ϫ831.
M. R. Ghadiri, M. A. Case, Angew. Chem. Int. Ed. Engl. 1993,
32, 1594Ϫ1597.
T. Sasaki, M. Lieberman, Tetrahedron 1993, 49, 3677Ϫ3689.
J. P. Schneider, J. W. Kelly, J. Am. Chem. Soc. 1995, 117,
2533Ϫ2546.
M. D. Struthers, R. P. Cheng, B. Imperiali, J. Am. Chem. Soc.
1996, 118, 3073Ϫ3081.
R. B. Hopkins, J. S. Albert, D. Van Engen, A. D. Hamilton,
Bioorg. Med. Chem. Lett. 1996, 4, 1121Ϫ1128.
[7]
[8]
[9]
[10]
[11]
[12]
[13]
Boc-Daf-Ala-OMe (12): White crystals (EtOAc/Et2O). M.p.
25
25
197Ϫ200 °C. Rf ϭ 0.22 (II); 0.08 (V). [α] ϭ Ϫ43, [α] ϭ Ϫ47,
589
578
[14]
25
25
25
[α] ϭ Ϫ58, [α] ϭ Ϫ119, [α] ϭ Ϫ277 (c ϭ 0.1; MeOH). Ϫ
546
436
365
[15]
1H NMR (CDCl3): δ ϭ 8.38 [m (broad), 2 H, ArH3H6], 7.87 [m
(broad), 2 H, ArH1H8], 7.27 and 7.25 [dd, J ϭ 4.9 Hz and 7.7 Hz,
2 H, ArH2H7], 6.80 [m (very broad), 1 H, NH Ala], 6.72 [m (very
broad), 1 H, NH Daf], 4.48 [dq, J ഠ 7.1 Hz and ഠ 7.1 Hz, 1 H,
Hα Ala], 3.67 [s, 3 H, OCH3], ഠ 1.4 and ഠ 0.70 [s (very broad), 9
H, Boc CH3 from anti and syn conformers], 1.24 [d, J ϭ 7.2 Hz, 3
H, CH3 Ala]. Ϫ 13C NMR (CDCl3): δ ϭ 172.6 and 167.8 (CϭO
Ala and Daf), 158.5, 158.3 (CAr), 153.3 (CϭO Boc), 150.9, 141.9,
131.6, 130.9, 123.7, 123.6 (CAr), 80.1 (OϪC Boc), 65.8 (Cα Daf),
52.4 (OCH3), 49.2 (Cα Ala), 27.6 (broad; CH3 Boc), 16.8 (CH3
Ala). Ϫ ESIϩ MS; m/z (relative intensity): 435 (100) [M,Na]ϩ; 413
(27) [M,H]ϩ. Ϫ C21H24N4O5 (412.434): calcd. C 61.15, H 5.87, N
13.58; found C 60.76, H 5.79, N 13.55.
[16]
[17]
[18]
N. Nishino, T. Kato, T. Murata, H. Nakayama, T. Arai, T.
Fujimoto, H. Yamamoto, S. Yoshikawa, Chem. Lett. 1996,
49Ϫ50.
[19]
A. B. Gretchikhine, M. Y. Ogawa, J. Am. Chem. Soc. 1996,
118, 1543Ϫ1544.
C. A. Slate, D. R. Striplin, J. A. Moss, P. Chen, B. W. Erickson,
T. J. Meyer, J. Am. Chem. Soc. 1998, 120, 4885Ϫ4886.
K. J. Kise Jr, B. E. Bowler, Tetrahedron: Asymmetry 1998, 9,
3319Ϫ3324.
M. W. Mutz, M. A. Case, J. F. Wishart, M. R. Ghadiri, G. L.
McLendon, J. Am. Chem. Soc. 1999, 121, 858Ϫ859.
[20]
[21]
[22]
Boc-Ala-Daf-OMe (13): Pale yellow solid (crude). M.p. 187Ϫ190
°C. Rf ϭ 0.23 (II); 0.43 (IV). Ϫ 1H NMR (CDCl3): δ ϭ 8.76 [d,
J ϭ 4.4 Hz, 2 H, ArH3H6], 8.06 and 8.00 [d, J ϭ 7.4 Hz, 2 H,
ArH1H8], 7.77 [s (broad), 1 H, NH Daf], 7.27 [dd, J ϭ 4.9 Hz and
7.5 Hz, 2 H, ArH2H7], 5.05 [d (broad), J ഠ 7.3 Hz, 1 H, NH Ala],
4.22 [dq, J ഠ 7.1 Hz and ഠ 7.1 Hz, 1 H, Hα Ala], 3.64 [s, 3 H,
OCH3], 1.41 [s, 9 H, Boc CH3], 1.38 [d, J ϭ 7.1 Hz, 3 H, CH3 Ala].
[23]
A. Polese, S. Mondini, A. Bianco, C. Toniolo, G. Scorrano, D.
M. Guldi, M. Maggini, J. Am. Chem. Soc. 1999, 121,
3446Ϫ3452.
[24]
B. M. Bishop, D. G. McCafferty, B. W. Erickson, Tetrahedron
2000, 56, 4629Ϫ4638.
[25]
Preliminary communication concerning this study: J.-P. Maza-
leyrat, M. Wakselman, F. Formaggio, M. Crisma, C. Toniolo,
Tetrahedron Lett. 1999, 40, 6245Ϫ6248.
Ϫ
13C NMR (CDCl3): δ ϭ 173.1 and 169.4 (CϭO Daf and Ala),
[26]
L. J.. Henderson, Jr, F. R. Fronczek, W. R. Cherry, J. Am.
Chem. Soc. 1984, 106, 5876Ϫ5879.
Y. Wang, W. Perez, G. Y. Zheng, D. P. Rillema, Inorg. Chem.
1998, 37, 2051Ϫ2059.
158.3, 158.2 (CAr), 155.8 (CϭO Boc), 151.6, 151.5, 138.9, 138.7,
132.9, 132.5, 129.7, 123.8, 123.7 (CAr), 80.5 (OϪC Boc), 65.2 (Cα
Daf), 53.5 (OCH3), 49.7 (Cα Ala), 28.2 (CH3 Boc), 17.0 (CH3 Ala).
[27]
1828
Eur. J. Org. Chem. 2001, 1821Ϫ1829

9-Amino-4,5-diazafluorene-9-carboxylic Acid (Daf), a New Cα,α-Disubstituted Glycine
FULL PAPER
[28]
[29]
[45]
[46]
[47]
[48]
[49]
[50]
ˇ
C. Toniolo, E. Benedetti, Macromolecules 1991, 24,
4004Ϫ4009.
J.-P. Mazaleyrat, A. Gaucher, J. Savrda, M. Wakselman, Tetra-
hedron: Asymmetry 1997, 8, 619Ϫ631.
T. S. Eckert, T. C. Bruice, J. Am. Chem. Soc. 1983, 105,
C. Toniolo, M. Crisma, F. Formaggio, G. Valle, G. Cavicchioni,
´
4431Ϫ4441.
G. Precigoux, A. Aubry, J. Kamphuis, Biopolymers 1993, 33,
1061Ϫ1072.
M. Bodansky, ‘‘Principles of Peptide Synthesis’’, Springer-Ver-
lag, Berlin, 1984.
J. Honzl, J. Rudinger, Coll. Czech. Chem. Commun. 1961, 26,
2333Ϫ2344.
J. R. Vaughan Jr, R. L. Osato, J. Am. Chem. Soc. 1952, 74,
676Ϫ678.
T. Wieland, W. Kern, R. Sehring, Justus Liebigs Ann. Chem.
1950, 569, 117Ϫ129.
W. König, R. Geiger, Chem. Ber. 1970, 103, 788Ϫ798.
T. Yamada, Y. Omote, Y. Nakamura, T. Miyazawa, S. Kuwata,
Chem. Lett. 1993, 1583Ϫ1586.
J.-P. Mazaleyrat, A. Gaucher, M. Wakselman, C. Toniolo, M.
Crisma, F. Formaggio, Peptides 1996, (Eds.: R. Ramage, R.
Epton), Mayflower Scientific, Kingswinford, U. K., 1998, pp
623Ϫ624.
[30]
[31]
C. Toniolo, Janssen Chimica Acta 1993, 11, (28) 10Ϫ16.
I. L. Karle, R. Gurunath, S. Prasad, R. Kaul, R. B. Rao, P.
Balaram, J. Am. Chem. Soc. 1995, 117, 9632Ϫ9637.
E. Benedetti, Biopolymers 1996, 40, 3Ϫ44.
I. L. Karle, Biopolymers 1996, 40, 157Ϫ180.
I. L. Karle, Acc. Chem. Res. 1999, 32, 693Ϫ701.
K. A. Bolin, G. L. Millhauser, Acc. Chem. Res. 1999, 32,
1027Ϫ1033.
[32]
[33]
[34]
[35]
[51]
[52]
[36]
[37]
ˇ
J. P. Mazaleyrat, A. Gaucher, Y. Goubard, J. Savrda, M. Wak-
selman, Tetrahedron Lett. 1997, 38, 2091Ϫ2094.
[53]
C. Toniolo, F. Formaggio, M. Crisma, J. P. Mazaleyrat, M.
Wakselman, C. George, J. R. Deschamps, J. L. Flippen-Ander-
son, B. Pispisa, M. Venanzi, A. Palleschi, Chem. Eur. J. 1999,
5, 2254Ϫ2264.
[54]
F. Formaggio, M. Crisma, C. Toniolo, J.-P. Mazaleyrat, M.
[38]
[39]
[40]
M. T. DuPriest, R. E. Conrow, D. Kuzmich, Tetrahedron Lett.
Wakselman, Peptides 1998, (Eds.: S. Bajusz, F. Hudecz), Akad-
1990, 31, 2959Ϫ2962.
´
emiai Kiado, Budapest, 1999, pp 352Ϫ353.
ˇ
J. Savrda, J. P. Mazaleyrat, M. Wakselman, F. Formaggio, M.
[55]
[56]
W. D. Fuller, M. Goodman, F.-R. Naider, Y.-F. Zhu, Biopoly-
Crisma, C. Toniolo, J. Pept. Sci. 1999, 5, 61Ϫ74.
M. Crisma, F. Formaggio, S. Mezzato, C. Toniolo, J. Savrda,
J. P. Mazaleyrat, M. Wakselman, Lett. Pept. Sci. 2000, 7,
mers 1996, 40, 183Ϫ205.
ˇ
Preliminary communication: M. Crisma, F. Formaggio, C.
Toniolo, M. Wakselman, J.-P. Mazaleyrat, Peptides 2000, Pro-
ceedings of the 26 th European Peptide Symposium, in press.
A. Lombardi, G. De Simone, S. Galdiero, F. Nastri, L. Di Co-
stanzo, K. Makihira, T. Yamada, V. Pavone, Biopolymers 2000,
53, 150Ϫ160.
123Ϫ131.
[41]
K. Kloc, J. Mlochowski, Z. Szulc, J. Prakt. Chem. 1977, 319,
959Ϫ967.
Z. Tai, X. Qian, J. Zou, G. Zhang, Mol. Eng. 1994, 3, 285Ϫ291.
D. S. Kemp, T. P. Curran, J. Org. Chem. 1988, 53, 5729Ϫ5731.
D. S. Kemp, R. I. Carey, J. Org. Chem. 1989, 54, 3640Ϫ3646.
[57]
[42]
[43]
[44]
Received November 7, 2000
[O00556]
Eur. J. Org. Chem. 2001, 1821Ϫ1829
1829