19F NMR based pH probes: Lanthanide(III) complexes with pH-sensitive chemical shifts

Article abstract of DOI:10.1039/b802838a

Experimental measurements and theoretical analysis of magnetic properties, structural dynamics and acid-base equilibria for several lanthanide(III) complexes with tetraazacyclododecane derivatives as ¹⁹F NMR chemical shift pH probes are present

Full text of DOI:10.1039/b802838a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
19F NMR based pH probes: lanthanide(III) complexes with pH-sensitive
chemical shiftsw
Alan M. Kenwright, Ilya Kuprov, Elena De Luca, David Parker,*
Shashi U. Pandya, P. Kanthi Senanayake and David G. Smith
Received (in Cambridge, UK) 19th February 2008, Accepted 19th March 2008
First published as an Advance Article on the web 1st May 2008
DOI: 10.1039/b802838a
Experimental measurements and theoretical analysis of mag-
netic properties, structural dynamics and acid–base equilibria
for several lanthanide(III) complexes with tetraazacyclododecane
derivatives as 19F NMR chemical shift pH probes are presented;
pKa values vary between 6.9 and 7.7, with 18 to 40 ppm chemical
shift differences between the acidic and basic forms for Ho(III)
complexes possessing T1 values of 10 to 30 ms (4.7–9.4 T,
295 K).
logy.13 19F NMR chemical shifts and relaxation parameters at
9.4 T with selected lanthanide ions are listed in Table 1 (see
ESI for further details on synthesis, basic properties and NMR
relaxation data over the field range 4.7 to 16.5 Tw).
The para-substituent at the aromatic ring of L1–L3 ligands
regulates the acidity of the amide hydrogen and therefore the pH
sensitivity of the complex. Consistent with the low acidity (pKa
4 9.5) of the amide proton of L2 and L3 ligands (para-NH2 and
-OH groups, respectively), no pH dependence of the 19F chemi-
cal shift of [LnꢀL2] and [LnꢀL3] was observed over the pH range
of 3.5 to 9.0. The nitro-substituted L1 ligand is significantly more
acidic, and the NH group of [LnꢀL1] deprotonates, causing the
CF3 group of [HoꢀL1] to shift from ꢁ55.1 ppm to ꢁ36.8 ppm
under basic conditions (Scheme 2). Because the acid–base
equilibrium is fast, only one signal is observed, with the chemical
shift corresponding to the weighted average of the chemical
shifts of the individual forms. Fitting the acid–base equilibrium
curve to the experimental chemical shift data (Fig. 1) resulted in
a pKa = 7.77 ꢂ 0.02, in 0.1 M NaCl solution.
19F Magnetic resonance probes offer considerable scope for
biological and biomedical studies.1–6 The absence of a back-
ground signal, the high sensitivity of 19F NMR spectroscopy
and the ability to acquire spectra or images on instruments with
1
similar probes and frequencies to H NMR spectroscopy, each
suggest an important role for such probes. Significant progress
has been made recently,6,7 but some difficulties remain, most
notably with longitudinal 19F relaxation times, which are typi-
cally of the order of 1 to 2 seconds at high magnetic fields.2,8 In
high-field 19F MRI studies in biosystems, high concentrations
(B50 mM) are currently required in order to obtain reasonable
signal intensities on the MRI experiment timescale.9,10
While the longitudinal 19F relaxation rate of [HoꢀL1] is
nearly pH-independent, we observed an abrupt increase (by
a factor of 8) in the linewidth (Table 1), which disappears upon
heating, indicating the presence of an intermediate timescale
chemical exchange process in the deprotonated complex. Two
possible chemical exchange processes were considered: lantha-
nide re-coordination, from oxygen to nitrogen, or cis–trans
isomerisation around the {CQN– double bond (Scheme 2).
Based on DFT calculations,wlanthanide re-coordination can
be ruled out, as the DFT energy difference between the two
coordination isomers of [LaꢀL1] is 62 kJ molꢁ1 in favour of the
oxygen-bound isomer. However, the calculated energy differ-
One way to surmount these difficulties involves creating a
paramagnetic centre (such as an open-shell lanthanide ion)
within 7 A of the fluorine nucleus. This enhances the rate of
19F relaxation, predominantly via dipolar mechanisms,11,12 by
about two orders of magnitude.13 As an added benefit, the
pseudocontact shift (PCS)12 induced by the Ln3+ ion amplifies
the 19F chemical shift sensitivity to minor structural changes.
A balance needs to be maintained between the benefits of
relaxation enhancement and the reduced detection sensitivity
associated with the broader line width. The signal intensity
gain by using Tm, Dy or Tb complexes (compared to diamag-
netic Y(III) analogues) in 19F NMR spectroscopy has been
ence between the cis- and trans-isomers is only 18 kJ molꢁ1
,
making the cis–trans isomerisation a more likely explanation
for the additional 19F signals observed. Based on the saddle
point energy, the forward and backward activation energies
for the cis–trans isomerisation are 27 kJ molꢁ1 and 45 kJ
molꢁ1, consistent with the exchange broadening observed in
estimated to be 20–50 fold, bringing the practicable use of 19
F
imaging experiments within grasp13 with sub-millimolar probe
concentrations.
We describe two classes of pH-sensitive paramagnetic com-
plexes that can serve as 19F chemical shift based pH probes.
The ligands used in this work are shown in Scheme 1. They
were prepared using previously published synthetic methodo-
Department of Chemistry, Durham University, South Road, Durham,
UK DH1 3LE. E-mail: david.parker@durham.ac.uk
w Electronic supplementary information (ESI) available: DFT calcu-
lations, calculated structures, relaxation analysis, synthesis and char-
acterisation. See DOI: 10.1039/b802838a
Scheme 1
ꢃc
This journal is The Royal Society of Chemistry 2008
2514 | Chem. Commun., 2008, 2514–2516

View Article Online
Table 1 19F chemical shifts and relaxation parametersa (295 K, 1 mM complex, D2O) for lanthanide(III) complexes with ligands L1 to L4
Complex
[HoL1H] [HoL1]ꢁ
[HoL4H] [HoL4]ꢁ
[TbL2]
[TbL3]b
[DyL3]b
[HoL3]b
[TmL3]b
dF/ppm
ꢁ55.1
92 ꢂ 9
179
ꢁ36.8
1490d
ꢁ58.0
112 ꢂ 6
210
ꢁ98.1
128 ꢂ 5
989e
ꢁ53.0
ꢁ52.5, ꢁ40.6
ꢁ66.7, ꢁ43.8 ꢁ57.9, ꢁ49.0
ꢁ78.1, ꢁ89.2
47 ꢂ 7
89
R1c@ 9.4 T/sꢁ1
R2 @ 9.4 T/sꢁ1
a
116 ꢂ 21
125 ꢂ 6 133 ꢂ 7
162 ꢂ 11
124 ꢂ 10
272
206
355
192
Diamagnetic complexes (with La3+ or Y3+) with the same ligands under the same conditions have R1 B 1 sꢁ1 and R2 B 3 sꢁ1; dF and R1 values given
b
here were independent of complex concentration over the range 0.2 to 2 mM. The share of minor species (at the second chemical shift quoted): 12%
c
d
(Tb), 12% (Ho), 9% (Tm) and 20% (Dy). R2 values estimated as 2p(half-width@half-height). Intermediate exchange regime between two cis–trans
isomers. See text for further details. Intermediate exchange regime between two coordination isomers. See text for further details.
e
the 19F NMR spectra. The calculations highlighted the im-
portant role of the bulky CF3 group in favouring rotamers
that place the CF3 group away from the amide oxygen.
Complexes that belong to a ‘‘slow-exchange’’ class may also
be considered, e.g. [LnꢀL4] (Scheme 1), in which chemical shifts
are pH-independent, and it is relative signal intensities that are
altered.13–15 In aqueous solution, two signals with pH-depen-
dent intensities and 40 ppm chemical shift difference were
observed for [HoꢀL4], corresponding to N-coordinated and
diaqua species (Scheme 3). Fitting the titration data (Fig. 2)
yielded pKa values of 5.71 ꢂ 0.02 (water), 6.88 ꢂ 0.02 (murine
urine) and 6.92 ꢂ 0.02 (murine plasma). The higher pKa values
in biofluids are the result of additional stabilisation of the
acidic form by carbonate coordination (Scheme 3).16,17 The
basic form exists as a mixture of two diastereomers, defined by
the R or S configuration at sulfur (Scheme 3). The experimental
ratio is 6 : 1 (298 K, pH = 8), resonating at ꢁ98 and ꢁ112
ppm, respectively. A slow re-coordination process intercon-
verts these two species: EXSY spectroscopy and a VT 19F
NMR study revealed an exchange process (Tc 323 K at 376
MHz), that probably occurs via a cooperative SQO–Ln bond
cleavage/reformation. DFT calculations confirmed this
hypothesis—the computed energy difference between the two
coordination isomers (via either oxygen at the stereogenic S
centre) is 8.9 kJ molꢁ1l for [LaꢀL4]. A first-order saddle point
was found between the two structures, corresponding to for-
longitudinal relaxation rate is:11,12
ꢂ
ꢃ
ꢀ
ꢁ
2
2
2
2
m0 gFmeff
7trþe
3trþe
R1 ¼
þ
1 þ o2et2rþe 1 þ o2Ft2rþe
15 4p
r6
o2Fm4eff
ꢀ
ꢁ
2
2
m0
3tr
6 1 þ o2Ft2r
þ
2
5 4p
ð3kTÞ r
where all the symbols have their usual meaning11,12 and
tr+e = (1/tr + 1/te)ꢁ1 is the effective dipolar correlation time
(composed of the rotational tr and the electronic te contribu-
tion). Because a sufficiently large data set is available (four
lanthanides at four different fields), we decided to avoid further
approximations (extreme narrowing is sometimes assumed in
one or both terms) and perform the non-linear global fit directly.
Since the geometry does not change significantly between dif-
ferent lanthanides (ionic radius variation is minor: Tb3+ 1.09 A,
Dy3+ 1.08, Ho3+ 1.07 and Tm3+ 1.05 A in 9 coordinate
systems18) the Ln–F distance and the rotational correlation time
can be set to be global between the Tb, Dy, Ho and Tm data sets
(full tables are in the ESIw). The values of meff and tr+e are kept
‘local’ for every lanthanide dataset. While each individual fit is
ambiguous, the global fit (Fig. 3) has a single well-defined
weighted least-squares minimum. The resulting values are
r
Ln–F = 6.9 ꢂ 0.8 A, tr = 280 ꢂ 12 ps (alternative fitting using
fixed values of meff: 7.6 for Tm3+, 10.6 for Ho3+, 9.7 for Tb3+
and 10.6 for Dy3+ results in rLn–F = 6.24 ꢂ 0.01 A, tr = 286 ꢂ
12 ps), compared with the Stokes–Einstein rotational correlation
time estimated from DFT volume data tr = 243 ps (hrmoli =
6.18 A) and the DFT Ln–F distance of hri = 6.95 A. Lower
values of tr have usually been reported by analysing 1H NMRD
ward and backward activation energies of 85 and 76 kJ molꢁ1
.
The fluorine–lanthanide distances may be obtained from the
relaxation analysis,11,12 which was performed for the L3 ligand
complexes. The dominant relaxation mechanisms for the
paramagnetic systems in question are electron–nucleus
dipole–dipole (DD) and Curie processes (CSR). The total
Fig. 1 Variation of 19F chemical shift for [HoꢀL1] with pH (295 K,
1 mM complex in 0.1 M aqueous NaCl). Solid lines: least-squares fits
to the fast-exchange acid–base equilibrium for the chemical shift.
Scheme 2
ꢃc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 2514–2516 | 2515

View Article Online
Fig. 3 Variation of the observed relaxation rate with oF for [LnꢀL3]
(Ln = Tb, Dy, Ho and Tm, 295 K, 1 mM complex, D2O). Solid lines:
global least-squares fits to relaxation theory equation.
relative signal intensities. Such behaviour augurs well for
chemical shift imaging studies using such complexes.
We thank the EPSRC, the Nuffield Foundation (DGS
summer studentship), the Royal Society, Prof. Ross Maxwell
for samples of biofluids and the EC network DIMI for support.
Scheme 3
profiles that implicitly examine the motion of the Gd–water
proton vector.19 Few independent measurements of the overall
Notes and references
tumbling rate of the complex have been made but studies of 17
O
1 H. R. Kalbitzer, G. Rohr, E. Nowak, R. S. Goody, W. Kuhn and
H. Zimmermann, NMR Biomed, 1992, 5, 347.
2 F. Khan, I. Kuprov, T. D. Craggs, P. J. Hore and S. E. Jackson, J.
Am. Chem. Soc., 2006, 128, 10729.
3 G. Papeo, P. Giordano, M. G. Brasca, F. Buzzo, D. Caronni, F.
Ciprandi, N. Mongelli, M. Veronesi, A. Vulpetti and C. Dalvit, J.
Am. Chem. Soc., 2007, 129, 5665.
2
VT NMR (water motion) or ligand 13C or H relaxation time
analysis (2H labelled ligand) to derive tr values have been
reported.20 The values are in the range 160–265 ps at 298 K
for complexes of similar molecular volume, and were system-
atically higher than those derived by examining water proton
motion or relaxation.
4 J. W. Peng, J. Magn. Reson., 2001, 153, 32.
5 L. D. Stegman, A. Rehemtulla, B. Beattie, E. Kievit, T. S.
Lawrence, R. G. Blasberg, J. G. Tjuvajev and B. D. Ross, Proc.
Natl. Acad. Sci. U. S. A., 1999, 96, 9821.
6 J.-x. Yu, V. D. Kodibagkar, W. Cui and R. P. Mason, Curr. Med.
Chem., 2005, 12, 819.
7 Y. G. Gakh, A. A. Gakh and A. M. Gronenborn, Magn. Reson.
Chem., 2000, 38, 551.
8 W. E. Hull and B. D. Sykes, Biochemistry, 1974, 13, 3431.
9 V. D. Kodibagkar, J. X. Yu, L. Liu, H. P. Hetherington and R. P.
Mason, Magn. Reson. Imaging, 2006, 24, 959.
In summary, two new classes of pH-sensitive 19F-labelled
paramagnetic complexes have been defined. With holmium(III),
40 ppm ([HoꢀL4]) and 18.3 ppm ([HoꢀL1]) changes in chemical
shift were observed between acidic and basic forms. [HoꢀL1]
belongs to the ‘fast-exchange’ type and changes its 19F chemi-
cal shift with pH, whereas [HoꢀL4] is a ‘slow-exchange’ ratio-
metric probe, responding to pH changes by variation of
10 J. X. Yu, L. Liu, V. D. Kodibagkar, W. N. Cui and R. P. Mason,
Bioorg. Med. Chem., 2006, 14, 326.
11 I. Bertini, F. Capozzi, C. Luchinat, G. Nicastro and Z. C. Xia, J.
Phys. Chem., 1993, 97, 6351.
12 I. Bertini, C. Luchinat and G. Parigi, Prog. Nucl. Magn. Reson.
Spectrosc., 2002, 40, 249.
13 P. K. Senanayake, A. M. Kenwright, D. Parker and S. K. van der
Hoorn, Chem. Commun., 2007, 2923.
14 M. P. Lowe and D. Parker, Chem. Commun., 2000, 707.
15 M. P. Lowe, D. Parker, O. Reany, S. Aime, M. Botta, G.
Castellano, E. Gianolio and R. Pagliarin, J. Am. Chem. Soc.,
2001, 123, 7601.
16 S. Aime, A. Barge, M. Botta, J. A. K. Howard, R. Kataky, M. P.
Lowe, J. M. Moloney, D. Parker and A. S. de Sousa, Chem.
Commun., 1999, 1047.
17 Y. Bretonniere, M. J. Cann, D. Parker and R. J. Slater, Org.
Biomol. Chem., 2004, 2, 1624.
18 R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr.,
Theor. Gen. Crystallogr., 1976, 32, 751.
19 P. Caravan, J. J. Ellison, T. J. McMurry and R. B. Lauffer, Chem.
Rev., 1999, 99, 2293.
20 H. Lamers, F. Maton, D. Pubanz, M. W. Van Laren, H. Van
Bekkum, A. E. Merbach, R. N. Muller and J. A. Peters, Inorg.
Chem., 1997, 36, 2527.
Fig. 2 Variation of the ratio (smaller/greater) of 19F signal integrals
of [HoꢀL4H] and [HoꢀL4]ꢁ with pH (295 K, 0.5 mM complex in 0.1 M
aq. NaCl). Solid lines: least-squares fits to the slow-exchange
acid–base equilibrium equation for the NMR intensity ratio.
ꢃc
This journal is The Royal Society of Chemistry 2008
2516 | Chem. Commun., 2008, 2514–2516