Bioconjugafe Chem. 1995, 6,77-81
77
Fluorinated o-AminophenolDerivatives for Measurement of Intracellular pH Chung K. Rhee, Louis A. Levy, and Robert E. London* Laboratory of Molecular Biophysics, National Institute of Environmental Health Sciences, Box 12233, Research Triangle Park, North Carolina 27709. Received September 1, 1994@ The simple 2-aminophenol group which serves as a building block for many cationic indicators has been modified to yield a series of pH sensitive probes. This approach is based on the replacement of one of the N-acetate groups of the chelator APTRA (o-aminophenol N,N,O-triacetate) by a n N-ethyl group. The resulting series of (N-ethy1amino)phenol (NEAP) compounds exhibit pK values in the physiological range and negligible affinity for physiological levels of other ions. Three fluorinated analogs have been prepared: N-ethyl-5-fluoro-2-aminophenol N,O-diacetate (5F NEAP), N-ethyl-2((2-fluoro-4-carboxybenzyl)oxy)-4-fluoroaniline-N-acetic acid (5F NEAP-21, and l-(2-(N-ethylamino)5-fluorophenoxy)-2-(2-fluoro-4-aminophenoxy)ethane-N,~,N'-triaceticacid (5F NEAP-3). These derivatives exhibit total titration shifts of -11 ppm. NEAP-2 and NEAP-3 contain a n additional fluorine to serve as an internal chemical shift reference, and NEAP-3, the most highly charged analog prepared, was designed in order to minimize leakage.
Since the pH of biological systems is a n important regulator of many physiological processes considerable effort has been expended on the development of methods for its measurement (1-3). Intracellular pH has been determined using a variety of physical techniques, including direct measurement with microelectrodes, a variety of spectrophotometry methods, fluorescence techniques, and nuclear magnetic resonance. An important advantage of the NMR method has been the ability to carry out determinations based on the observation of endogenous phosphorylated molecules whose chemical shift is pH dependent (I,3-5). This approach tends to be limited by several factors including the low concentration of inorganic phosphate, the most useful pH indicator, in many cells. Interference from extracellular phosphate if present in blood, buffers, or perfusates, as well as the overlap of the phosphate resonance with the resonances from various phosphomonoesters, can also limit the accuracy of the measurement. Finally, the absence in some cell types of a suitable metabolite to serve as a chemical shift reference also serves to limit this technique. Consequently, a number of groups have proposed and developed exogenous NMR probes of intracellular pH (6-15). Such probes can in principle be optimized for pH measurement by consideration of a number of criteria, including: (1)resonances should exhibit a large Ad/ApH and a pK close to the physiological mean; (2) detection sensitivity should be high; (3) it is desirable to utilize indicators which can be loaded into cells and which will not readily leak out once they are loaded; (4) there should be minimal overlap between the resonances of the indicator and those of other endogenous metabolites; (5)if the chemical shift is the parameter of interest, the indicator should ideally contain a n internal reference so that no additional referencing is required. On the basis of the above considerations, the use of fluorinated indicators is particularly attractive since criteria (2) and (4) are completely satisfied, and due to the high chemical shift sensitivity of 19F, the first criterion will be fulfilled as well if the indicator is appropriately designed. Several groups have developed fluorinated pH indicators (9-15). Fquene, a fluorinated analog of the fluorescent indicator quene 1, contains a fluoroquinoline @
Abstract published in Advance ACS Abstracts, December
1, 1994.
moiety and exhibits a total shift of -4.5 ppm (13). This indicator can be loaded into cells as the tetraacetoxymethyl ester (13,14), and the high charge resulting after intracellular hydrolysis minimizes leakage from the cell. However, the Mg2+affinity of quene 1is relatively high (161, and we have found that very similar quinoline compounds have a relatively high affinity for Mg2+ions. Deutsch and co-workers have evaluated a variety of fluorinated compounds to measure intracellular pH (912). The amino acid a-(difluoromethy1)alaninehas a pK of 7.3 and two nonequivalent fluorine nuclei such that the chemical shift difference between the resonances can be used to determine pH without the need for a n additional chemical shift standard (10).However, the shift is relatively small, and the compound leaks out of cells on a fairly short time scale. A series of fluorinated aniline derivatives were also evaluated as potential intracellular pH indicators (12). These compounds show large pH dependent shifts, in some cases sufficient to lead to slow exchange on the chemical shift time scale. However, the molecules considered had only a single carboxyl group, so that leakage is again a potential problem, and there was no additional fluorine t o provide a reference shift (12). RESULTS AND DISCUSSION
Our experience with metallospecific chelators containing a fluorinated aniline unit in which the chemical shift of the para fluorine is sensitive to changes in the electronic environment of the amino group suggested the possibility of developing a useful pH sensitive indicator based on this structure. Chelators such as nF-BAPTA (17)and nF-APTRA (18)used as calcium or magnesium specific indicators and based on a n aniline structure are designed to be pH insensitive in the biologically important range near 7 and, thus, have a pK of about 5.5 (for the 5F derivatives) or 4.0 (for the 4F derivatives). I t was anticipated that the metal binding and pH sensitivities of these compounds could be readily altered by replacing one of the N-carboxymethyl groups with a n alkyl group. This approach exchanges one of the chelating carboxyl functions for an electron-donating group to increase the pK of the amino group. A simple example of this strategy, N-ethyl-2-amino-5-fluorophenol-N,O-diacetic acid, 5F NEAP-1, was prepared as shown in Scheme 1. Monoethylation of 2-(benzyloxy)-4-fluoroanilinewas most
Not subject to US. Copyright. Published 1995 by American Chemical Society
78 Bioconjugate Chem., Vol. 6, No. 1, 1995
Rhee et al.
Scheme 1"
Scheme 2"
@--
f
F
6
1
w--
"NAC~R O V C W
% i
F
Q
f4"
F
P
F
1
f i N
4.R-Bn
3
5R.H
Reagents: (a) NaBH&H3COzH; (b) Hz, PdC, EtOAc; (c) BrCHZCOzBn, Proton Sponge, CH3CN; (d) Hz, PcUC, EtOAc. a
I
9
10.RJllo 11.R.H
a Reagents: (a)&c03, DMF; (b) H2, PcUC, EtOAc; (c) NaBHd CH3COzH; (d) BrCHzCOzMe, Proton Sponge, CH3CN (e)NaOH, EtOWHzO.
changes in pH. The synthesis of 11 (Scheme 2) makes use of the sequence development for the prototype 5. The benzyl protecting group in 1 is now modified to contain a fluorine substituent, whose chemical shift will be pH insensitive, thereby serving as a n internal standard. An additional carboxylic acid function is also present in the benzyl group in order to ensure good aqueous solubility and to retard leakage. The protecting group is therefore not removed, but remains as part of the chelator, 5F NEAP-2. Condensation of 5-fluoro-2-nitrophenol with the benzyl bromide 6 obtained from the bromination of 2-fluoro-4-carbomethoxytoluene led directly to the basic framework of the eventual indicator. Reduction of the nitro group to the aniline 8 and application of the same alkylation used in the synthesis of 5F NEAP-1 yielded the N-ethylaniline derivative 9. Alkylation with methyl bromoacetate to give 10 and subsequent hydrolysis of the two ester functions produced 5F NEAP-2, 11. Upon titration 11 showed a pH and 19Fchemical shift profile similar to 5 (Figure 1). Unfortunately, the 19F shift corresponding to the internal reference fluorine is fairly close to the shift of the pH sensitive resonance, such that there was overlap a t high pH values. Additionally, indicator 11 is larger and more hydrophobic while having only two carboxyl groups. In order to remedy these deficiencies, we prepared compound 19, 5F NEAP-3 (Scheme 31, which, in addition to a pH sensitive 19Fnucleus and pH insensitive 19F to serve as an internal standard, contains three carboxyl groups to further retard the leakage of the indicator out of the cell. The synthesis of 19,5F NEAP3,which contains two unequivalently substituted nitrogens, required a somewhat different strategy than the two preceding cases. The N-ethyl function is introduced in a masked form as a n acetamide in 14. After the coupling of 13 with 2-fluoro-4-nitrophenol to form the backbone of the indicator, the nitro group is reduced to the aniline and the acetamide moiety is reduced to the desired N-ethyl unit. Alkylation with benzyl bromoacetate to 18 and hydrogenolysis yields the desired indicator NEAP-3 (19). The reference fluorine resonance for this compound is ortho to a n oxygen substituent and, hence, exhibits a large upfield chemical shift (Figure 2). As with 5 and 11 this compound also has a n appropriate pK and 19Fchemical shift sensitivity. The compounds described above have been loaded into several cell types and appear to function adequately in the intracellular environment. Another attractive feature of these indicators, illustrated by the derivatives discussed above, is that a wide range of analogs can be prepared by substitution a t the hydroxyl oxygen without
'11 ,L, .
-10
2
4
6
8
10
12
14
PH Figure 1. Fluorine-19 shifts as a function of pH for 5F NEAP-1 (O), 5F NEAP-2 (A), and 5F NEAP-2 (B). The shiR for 5F NEAP-1is referenced to the tetrailuorophthalate l9F resonance, while the data for 5F NEAP-2 and 5F NEAP-3 correspond to the shift differences between the two fluorine resonances. The theoretical curves correspond to 61 = 35.6 ppm, 62 = 24.7 ppm, pK = 6.85 for 5F NEAP-1,61 = 10.3 ppm, 62 = -0.8 ppm, pK = 6.6 for 5F NEAP-2, and 61 = 24.1,62 = 13.4, pK = 6.8 for 5F NEAP-3. As is apparent from the data, the two 19Fresonances of NEAP-2 cross at lower pH, and the resonance corresponding to the fluorine on the benzoate ring shows a small additional shift perturbation at low pH due t o titration of the benzoic acid.
easily effected using the alkylation reagent NaBHdCHSCOzH (19).The removal of the benzyl protecting group and alkylation of both the amino and phenolic hydroxyl moieties yielded the benzyl ester of 5F NEAP-1 (N-ethyl2-amino-5-fluorophenolN,O-diacetic acid). Hydrogenolysis of the benzyl esters then yielded the desired acid form of the indicator. Titration of the acid 5 (Figure 11, as followed by the shift of the 19Fresonance (tetrafluoroterphthalic acid as standard), showed our conception to be sound, yielding a pK of 6.8 and with a large, 11ppm chemical shift. As noted above, it is desirable that indicators for which the chemical shift serves as the parameter of interest contain a n additional fluorine nucleus to serve as a chemical shift standard. In addition to the obvious advantage of convenience, this ensures that the shift reference will be in the same region of the magnetic field, so that local field variations will not affect the pH determination. Further, it is now clear that the intracellular environment can significantly perturb the shift of fluorinated compounds (20).We thus prepared compound 11, 5F NEAP-2 (Scheme 2), which incorporated an additional fluorine subsitituent designed to serve as a n internal reference; i.e, it is insensitive or nearly so to
Sioconjugate Chem., Vol. 6,No. 1, 1995 79
Measurement of Intracellular pH
Scheme sa
building block means that fluorescent pH indicators with structures and fluorescent properties closely analogous to existing cationic indicators can be readily prepared.
&--
EXPERIMENTAL PROCEDURES
F
1
NHX OR
"&.,e N4
f
14
12. X = Ac, R =En 13. X =Ac. R = H
13
+
o*CHzC!Hfo~
14
NRz
15. X = Ac. 16. Ac,R = 0 H 17. X = Et, R = H
F
P
a Reagents: (a) H2, W C , EtOAc; (b) Ac20, EtsN, (c) H2, Pd/ C, EtOAc; (d) BrCHzCHzBr, K&O3, DMF; (e) KzC03, DMF; (0 H2, W C , EtOAc; ( g ) LiAlH4, THF; (h)BrCHzCOzMe, Proton Sponge, DMF; (i) KOH, MeOH.
NEAP4
1
7.70
6.50
Figure 2. Fluorine-19 NMR spectra as a function of pH for 5F NEAP-3. The sample also contained NaF as an external standard.
exerting a major perturbation on the pK and fluorine shift parameters. Finally, the presence of the o-aminophenol
Fluorine-19 NMR spectra of the indicators were measured as a function of pH a t 37 "C on a n NT-360 spectrometer using a 5 mm probe retuned to the 19FNMR frequency of 340 MHz. Studies were carried out in a buffer containing 120 mM KC1,20 mM NaC1, and 20 mM Tris-HEPES to model the intracellular milieu. In some studies, a n external capillary with 20 mM NaF was used as a chemical shift standard. Unless otherwise noted, commercially available reagents and dry solvents were used as received. Reactions were carried out under a n atmosphere of argon, and reaction temperatures refer to the bath. Unless stated otherwise, all reported compounds were homogeneous as judged by the thin layer chromatography (TLC) analysis and their NMR spectra. Flash column chromatography (FCC) was performed according to Still et al. (22)with Merck silica gel 60 (4063 pm). Synthetic intermediates were characterized using 'H nuclear magnetic resonance (lH NMR) and 19F nuclear magnetic resonance (I9FNMR) measured a t 500 and 470 MHz, respectively, on a General Electric GN500 spectrometer. Unless otherwise noted, NMR spectra were obtained in CDCl3 solution. For 'H NMR, the residual CHCl3 in CDCl3 was employed as the internal standard and assigned as 7.24 ppm downfield from tetramethylsilane (TMS). For 19FNMR, hexafluorobenzene was employed as the internal standard and assigned as 0 ppm. Melting points were determined on a FisherJohns melting point apparatus and are uncorrected. N-Ethyl-2-(benzyloxy)4fluoroaniline (2). Two pellets of NaBH4 (311 mg) were added to a solution of 3.18 g (14.7 mmol) of 2-(benzyloxy)-4-fluoroanilinein 80 mL of glacial acetic acid. Another two pellets (320 mg) were added 0.5 h d e r the first addition. The reaction mixture was then stirred overnight a t room temperature. The reaction mixture was neutralized with 3 N NaOH and extracted with ether, and the combined extracts were washed with H20 and dried (MgS04). Removal of the solvent yielded 3.06 g of crude product. Flash chromatography (9:l hexandethyl acetate) of this material yielded 1.22 g (35%)of white crystals, mp 58-59 "C. lH NMR: 1.24 (t, J = 7.1 Hz, 3H), 3.12 (9, J = 7.1, 2H), 5.03 (s, 2H), 6.5 (m, lH), 6.6 (m, 2H), 7.4 (m, 5H). N-Ethyl-2-hydroxy-4-fluoroaniline (3). A solution of 847 mg (3.46 mmol) of 2 in 30 mL of ethyl acetate was hydrogenolyzed, using 109 mg of 10% PdC. The reaction proceeded rapidly until 1 equiv of hydrogen had been taken up. The catalyst was then filtered through a pad of Celite and the solvent removed to yield 484 mg of product (go%), mp 92-97 "C dec. This material was homogeneous by TLC, Rf= 0.2 (9:l hexane/ethylacetate) but discolored rapidly. lH NMR: 1.22 (t, J = 7 Hz, 3H), 3.07 (9, J = 7 Hz, 2H), 6.5 (m, 2H), 6.7 (m, 1H).
N-Ethyl-2-hydroxy-4-fluoroaniline-N,O-diace tic Acid Dibenzyl Ester (4). A mixture of 475 mg (3.06 mmol) of 3, 1.60 g of Proton Sponge, 1.74 g of benzyl bromoacetate, and 20 mL of acetonitrile was heated a t reflux for 48 h under a n argon atmosphere. The cool solution was filtered, and ether was then added to the filtrate. The resulting precipitate was filtered off and the filtrate washed with pH 2 buffer, saturated NaCl solution, and water and dried (MgSOJ. After removal of the solvent, the resulting crude product was purified by flash chromatography (85:15 hexane/ethyl acetate) to yield 303 mg (23%)of clear colorless oil. lH NMR: 1.05 (t, J = 7.1
Rhee et al.
80 Bioconjugate Chem., Vol. 6,No. 1, 1995
(10). A mixture of 624 mg (1.94 mmol) of 9, 465 mg of Proton Sponge, 360 mg of methyl bromoacetate, and 10 mL of dry acetonitrile was refluxed under argon for 54 N-Ethyl-2-hydroxy-4-fluoroaniline-N,O-diacetic h. After 24 h of reflux, a n additional 65 mg each of methyl bromoacetate and Proton Sponge were added. The Acid (5). Hydrogenolysis of the diester 4 (54 mg, 0.1 cool solution was diluted with ether and the amine salt mmol) in 15 mL of ethyl acetate with 10 mg of 10% PcUC filtered off. The filtrate was washed with pH 2 buffer yielded the free acid, 27 mg, mp 54-56 "C. lH NMR and water and dried (MgS04). Removal of the solvent (DzO): 0.99 (t, J = 7 Hz, 3H), 3.23 (9, J = 7 Hz, 2H), yielded 670 mg of crude product which after flash 3.89 (br s, 2H), 4.64 (br s, 2H), 6.7 (m, 2H), 7.1 (m, 1H). chromatography (8515 hexandethyl acetate) yielded 419 19FNMR(D20): 43.03 (m). mg (55%)of crystalline product, mp 61-62 "C. lH NMR Methyl 3-fluoro-4-(bromomethyl)benzoate(6). A (CDCl3) 1.07 (t, J = 7 Hz, 3H), 3.26 (9, J = 7 Hz, 2H), mixture of 7.32 g (43.6 mmol) of methyl 3-fluoro-43.58 (s, 2H), 3.91 (s, 3H), 3.94 (s, 3H), 5.16 (s, 2H), 6.1 methylbenzoate, 4.92 g, N-bromosuccinimide (NBS), and (m, 2H), 7.01 (dd, J = 2 and 6 Hz, lH), 7.57 (dd, J = 7.5 a few mg of 2,2-azobisisobutyronitrile(AIBN) in 80 mL Hz, lH), 7.73 (dd, J = 1 and 10 Hz, lH), 7.84 (dd, J = 1 of CCL4 was brought to reflux and illuminated with a and 7.5, Hz, 1H). 19FNMR (DMF-&): 43.2 (m, lF), 45.7 high intensity lamp. The reaction was rapid, and within (m, 1F). 2 h, all of the NBS had been converted to succinimide. N-Ethyl-2-((2-fluoro-4-carboxybenzyl)oxy~-4-fluoTLC and NMR analysis indicated a 1:l mixture of roaniline-N-acetic Acid (11). A sample of 10 (42 mg, product and starting material as judged by the appear0.11 mm) was hydrolyzed with 1 M NaOH (3 mL) in ance of a new signal at 4.49 corresponding to bromination aqueous methanol and after acidification (0.6 N HC1) of the methyl group. The mixture was worked up by yielded 33 mg (84%) of pure acid 11, mp 73-77 "C. lH filtering the succinimide and washing the CC14solution, NMR (DzO): 0.61 (t, J = 7 Hz, 3H), 2.76 (9, J 7 Hz), first with aqueous NazS203 and then with aqueous 4.54 (s, 2H), 4.84 (s, 2H), 6.38 (ddd, J = 7.5, 7.5, and 2.5 NaHC03. The product obtained after drying and removal Hz, lH), 6.50 (dd, J = 10.5 and 2.5 Hz, lH), 6.70 (dd, J of the solvent was recycled with additional NBS and = 8.5 and 6.5 Hz, lH), 7.28 (dd, J = 7.5 and 7.5 Hz, lH), treated as above to yield 7.46 g of yellow oil which was 7.36 (br d, J = 10.5 Hz, lH), 7.44 (br d, J = 8.5 Hz, 1H). an 8:l mixture of product and starting material and was 19F(DMF-d7): 43.2 (m, lF), 45.7 (m, 1F). used as such in the next step. (12). Ace24(2-Fluoro-4-(methoxycarbonyl)benzyl)oxy)-4- N-Acetyl-2-(benzyloxy)-4-fluoroaniline tic anhydride (10 mL) was added to a solution of 2-(benfluoronitrobenzene (7). A mixture of 7.37 g of the zyloxy)-5-fluoroaniline(700 mg, 3.2 mmol) in pyridine (20 previously described impure 6 , 30 mL of dimethylformamL), and the reaction mixture was stirred for 1h. Icemide, 4.5 g of KzC03, and 4.29 g (30 mmol) of 2-nitro-5water (10 mL) was added to the mixture, and the crude fluorophenol was heated at 70 "C in a n oil bath. After 2 product was filtered and washed with water. Recrystalh, the reaction mixture was cooled and poured into icelization from benzene afforded benzyl ether 12 as a white water and the product filtered off. The crude product was solid (801 mg, 96%), mp 136-138 "C. lH NMR (CDCl3): extracted with warm hexane to remove the methyl 2.00 (s, 3H), 4.94 (br s, 2H), 6.8-6.7 (m, 2H), 7.2-7.1 (br 3-fluoro-4-methylbenzoate impurity carried along from s, 5H), 8.16 (dd, J = 8.6 Hz, 1H). the preparation of 6. The resulting cream colored N-Acetyl-2-hydroxy-5-fluoroaniline (13). A mixproduct, 5.74 g, after recrystallization from ethyl acetate ture of benzyl ether 12 (800 mg, 3.1 mmol) in ethyl had mp 163-165 "C. lH NMR: 3.92 (s, 3H), 5.29 (s, 2H), acetate (15 mL) and Pd/C (100 mg) was stirred overnight 6.78 (m, 1H), 6.86 (dd, J = 3.4 and 9.5 Hz, xH), 7.75 (m, under Hz (1 atm). The reaction mixture was filtered 2H), 7.89 (dd, J = 1 and 6.4 Hz, lH), 8.00 (dd, J = 6.9 through Celite and concentrated under reduced pressure. and 9.1 Hz, 1H). 2-((2-Fluoro-4-(methoxycarbonyl)benzyl)oxy)-4- The crude product was recrystallized from ethyl acetate to give phenol 13 as a white solid (512 mg, 98%), mp 185fluoroaniline (8). A solution of 1.28 g (3.96 mmol) of 7 186 "C. 'H NMR (CDCl3): 6.57 (dd, J = 10.5, 2.5 Hz, in 25 mL of ethyl acetate and 125 mg of 5% PtK was lH), 6.75(dd,J = 9.5, 2.5 Hz, lH), 6.88 ( d d , J = 9.5, 5.5 reduced with hydrogen at atmospheric pressure. After Hz, lH), 7.35 (br s, lH), 9.05 (br s, 1H). the uptake of 3 equiv of hydrogen the catalyst was filtered 1-Bromo-2-(2-fluoro-6-nitrophenoxy)ethane (14). off and the solvent removed to yield 1.14 g (98%) of Potassium carbonate (2.3 g, 16.7 mmol) was added to a product, mp 92-95 "C. lH NMR: 3.92 (s,3H), 5.16 (9, solution of 2-fluoro-4-nitrophenol (1.5 g, 10 mmol) in 2H), 6.54 (ddd, J = 2.5, 5.9 and 5.9 Hz, lH), 6.65 (m, dimethylformamide (10 mL), and the mixture was stirred 2H), 7.55 (dd, J = 7.5 and 7.5 Hz, lH), 7.74 (dd, J = 1.2 for 10 min. Dibromoethane (2.5 mL, 29 mmol) was added and 10.3 Hz, lH), 7.83 (dd, J = 1.2 and 8 Hz, 1H). to the reaction mixture and was heated a t 90 "C overN-Ethyl-2((2-fluoro-4(methoxycarbony1)benzyl)night. The reaction mixture was diluted with ether, and (9).To a solution of 850 mg (2.90 oxy)-4-fluoroaniline the ether layer was washed with water and brine and mmol) of 8 in 17 mL of glacial acetic acid was added 4 then dried (MgS04). The solvent was removed under pellets (600 mg) of NaBH4 over a period of 2 h. TLC reduced pressure, and the crude product was fractionated indicated the slow appearance of a product at Rf = 0.8 by FCC with 2.5%ethyl acetate in hexane to yield 14 as (7:3 hexane/ethyl acetate). Stirring a t room temperature a yellow oil (1.9 g, 75%). 'H NMR (CDC13): 3.69 (t, J = was continued for a total of 6 h. The reaction was 6 Hz, 2H), 4.44 (t, J = 6 Hz, 2H), 7.03 (dd, J = 8.5, 8.5 neutralized with aqueous NaHC03. The product preHz, lH), 8.1-7.9 (m, 2H). cipitated, after filtration was obtained as buff colored 1 (2-Acetamido-S-fluorophenoxy )-2(2-fluoro-4-nicrystals (654 mg, 70%), and after recrystallization from trophen0xy)ethane (15). A mixture of phenol 13 (450 hexane had mp 109-111 "C. lH NMR (CDCl3): 1.25 (t, mg, 2.7 mmol) and bromoethane 14 (850 mg, 3.2 mmol) J = 7.1 Hz, 3H), 3.11 (q, J = 7.1 Hz, 2H), 3.92 (s, 3H), in dimethylformamide (10 mL) containing K2C03 (500 5.15 (s, 2H), 6.5 (m, lH), 6.6 (m, 2H), 7.5 (dd, J = 7.5 mg, 3.6 mmol) was heated a t 70 "C overnight. The and 7.5 Hz, lH), 7.75 (dd, J =1.5 and 10.3 Hz, lH), 7.84 reaction mixture was quenched with ice-water, filtered, (dd, J = 1.5 and 7.9 Hz, 1H). N-Ethyl-2-( (2-fluoro-4-(methoxycarbonyl)benzyl)- and washed with cold water. The crude product was fractionated by FCC (dry application with 50% EtOAc/ 0xy)-4-flu0roaniline-N-aceticAcid, Methyl Ester Hz, 3H), 3.27 (q, J = 7.1 Hz, 2H), 4.03 (s, 2H), 4.67 (s, 2H), 5.05 (s, 2H), 5.17 (s, 2H), 6.47 (dd, J = 2.7 and 9.8 Hz, lH), 6.60 (ddd, J = 8.3, 2.4, and 2.4 Hz, 1H).
-
Bioconjugate Chem., Vol. 6,No. 1, 1995 81
Measurement of Intracellular pH
(2) Nuccitelli, R. and Deamer, D. W., Eds. (1982)Intracellular hexane) to provide 15 as a pale tan solid (1.0 g, 88%), p H Its measurement, Regulation and Utilization in Cellular mp 192-194 “C. ‘H NMR (CDC13): 2.15 (s, 3H), 4.6Functions Liss, New York. 4.4 (m, 2H), 7.59 (br s, lH), 8.03 (dd, J = 10.5, 2.5 Hz, (3) Jacobson, L., and Cohen, J. S. (1982) Intracellular pH lH), 8.08 (dd, J = 9.5, 2.5 Hz, lH), 8.29 (dd, J = 9.5, 7 Measurements by NMR Methods. In Noninvasive Probes of Hz, 1H). 1-(2-Acetamido-5-fluorophenoxy)-2-(2-fluom-4-~- Tissue Metabolism (J.S . Cohen, Ed.) Wiley, New York. (4) Moon, R. B., and Richards, J. H. (1973) Determination of nophenoxy)ethane (16). A mixture of nitro compound intracellular pH by 31PMagnetic Resonance. J . Biol. Chem. 15 (1.0 g, 3.1 mmol) in ethyl acetate (20 mL) and W C 248, 7276-7278. (200 mg) was stirred for 2 h under Hz (1 atm). The (5) Roberts, J. K. M., Wade-Jardetzky, N., and Jardetzky, 0. reaction mixture was filtered through Celite and concen(1981) Intracellular pH Measurements by 31PNuclear Magtrated under reduced pressure. The crude product was netic Resonance. Influence of Factors Other Than pH on 31P recrystallized with benzene, providing amino ethane 16 Chemical Shifts. Biochemistry 20, 5389-5394. as a white solid (709 mg, 78%),mp 134-136 “C. ‘H NMR (6) Thoma, J. W., Steiert, J. G., Crawford, R. L., and Ugurbil, (CDC13): 2.12 (s, 3H), 4.29 (s,4H), 6.4-6.3 (m, lH), 6.46 K. (1986) pH Measurements by 3lP NMR in Bacterial (dd, J = 12.5,2.5 Hz, lH), 6.7-6.6 (m, 2H), 6.85 (dd, J = Suspensions Using Phenyl Phosphonate as a Probe. Bwchem. 9, 9 Hz, lH), 8.30 (dd, J = 9, 6 Hz, 1H). Biophys. Res. Commun. 138, 1106-1109. 1-(2-(N-Ethylamino)-5-fluorophenoxy)-2-(2-fluoro- (7) DeFronzo, M., and Gillies, R. J. (1987) Characterization of 4-aminophenoxy)ethane(17). Lithium aluminum hyMethylphosphonate as a 31P NMR pH Indicator. J . Biol. dride (ca. 50 mg) was added to 16 (500 mg, 1.5 mmol) in Chem. 262, 11032-11037. THF (15 mL), and the reaction mixture was stirred (8) Szwergold, B. S., Brown, T. R., and Freed, J. (1989) overnight. The reaction mixture was quenched with Bicarbonate abolishes Intracellular Alkalinization in MitogenMeOH (1 mL) and filtered through Celite. After being Stimulated 3T3 Cells. J . Cell. Physiol. 138, 227-235. concentrated under reduced pressure, the crude product (9) Taylor, J. S., Deutsch, C. J., McDonald, G. G., and Wilson, was fractionated by FCC with 50% EtOAc in hexane to D. F. (1981) Measurement of Transmembrane pH Gradients in Human Erythrocytes Using 19FNMR. Anal. Biochem. 114, yield 17 as a white solid (300 mg, 63%) and recovered 415-418. starting material 16 (100 mg, 20%). After recrystalliza(10) Taylor, J. S., and Deutsch, C. J. (1983) Fluorinated tion from benzene, the product had mp 111-114 “C. ‘H a-Methylamino Acids as ‘9F NMR Indicators of Intracellular NMR (CDC13): 1.23 (t, J = 7 Hz, 3H), 3.10 (9, J = 7 Hz, pH. Biophys. J . 43, 261-267. 2H), 4.4-4.2 (m, 2H), 6.37 (br d, J = 8.5 Hz, lH), 6.5(11) Deutsch, C. J., and Taylor, J. S. (1987) Intracellular pH 6.5 (m, 2H), 6.57 (br d, J = 9 Hz, lH), 6.6-6.5 (m, lH), as Measured by ”F NMR. Ann. N.Y. Acad. Sci. 508,33-47. 6.85 (dd, J = 9, 9 Hz, 1H). (12) Deutsch, C. J., and Taylor, J. S. (1989) New class of 19F 1-(2-(N-Ethylamino)-5-fluorophenoxy~-2-(2-fluoropH indicators: fluoroanilines. Biophys. J . 55, 799-804. 4-aminophenoxy)ethane-N,”,”-triacetic Acid,”ri(13) Metcalfe, J. C., Hesketh, T. R., and Smith, G. A. (1985) methyl Ester (18). Methyl bromoacetate (0.34 mL, 3.6 Free cytosolic Ca2+ Measurements with Fluorine Labeled mmol) was added to the solution of amine 17 (200 mg, Indicators Using 19FNMR. Cell Calcium 6, 183-195. 0.6 mmol) and Proton Sponge (660 mg) in CH&N (10 (14) Beech, J. S., and Iles, R. A. (1987) 19Fn.m.r. indicators of mL), and the reaction mixture was heated under reflux hepatic intracellular pH in vivo. Biochem. SOC. Trans. 15, for 3 days. The cool reaction mixture was diluted with 871-872. ether (40 mL) and washed with pH 2 solution (20 mL), (15) Mehta, V. D., Kulkami, P. V., Mason, R. P., and Antich, water (20 mL), and then brine (20 mL). After being dried P. P. (1993) Fluorinated macromolecular probes for nonover MgS04, the ether solution was concentrated under invasive assessment of pH by magnetic resonance spectrosreduced pressure and fractionated by FCC with 50% copy. Bioorg. Med. Chem. Lett. 3, 187-192. EtOAc in hexane yielding triacetate 18 (270 mg, 79%), (16) Rogers, J., Hesketh, T. R., Smith, G. A., and Metcalfe, J. mp 62-64 “C. lH NMR (CDCl3): 1.87 (t,J = 7 Hz, 3H), C. (1983) Intracellular pH of Stimulated Thymocytes Mea3.65 (s, 3H), 3.75 (s,6H), 4.15 (s, 2H), 4.25 (x, 4H), 4.3sured with a New Fluorescent Indicator. J . Biol. Chem. 258, 4.2 (m, 2H), 4.4-4.3 (9, J = 7 Hz, 2H), 4.5-4.4 (m, 2H), 5994-5997. 6.76 (dd, J = 9, 2.5 Hz, lH), 6.85 (dd, J = 13.5, 3 Hz, (17) Smith, G. A., Hesketh, R. T., Metcalfe, J. C., Feeney, J., 1H), 7.2-7.1(m,2H),7.46(dd, J = 9 . 9 H z , lH), 7.83(dd, and Morris, P. G. (1983) Intracellular calcium measurements J = 8.5, 6 Hz, 1H). by 19FNMR of fluorine-labeled chelators. Proc. Natl. Acad. l-(2-(N-Ethylamino)-5-fluorophenoxy~-2-(2-fluoro-Sci. U S A . 80, 7178-7182. 4-aminophenoxy)ethane-N,”,”-triacetic Acid (19). (18) Levy, L. A., Murphy, E., Raju, B., and London, R. E. (1988) Potassium hydroxide (50 mg, 0.9 mmol) was added to Measurement of Cytosolic Free Magnesium Ion Concentration ester 18 (100 mg, 0.2 mmol) in MeOH (3 mL) and stirred by 19F NMR. Biochemistry 27, 4041-4048. overnight. The reaction mixture was acidified with 1M (19) Gribble, G. W., Lord, P. D., Skotnicki, J., Dietz, S. E., HC1, filtered, and washed with water to provide acid 19 Eaton, J. T., and Johnson, J. L. (1974) Reactions of sodium borohydride in acidic media. I. Reduction of Indoles and as a white solid (73 mg, 80%), mp 110-115 “C. lH NMR alkylation of aromatic amines with carboxylic acids. J . Am. (DzO): 0.78 (t, J = 7 Hz, 3H), 2.92 (9, J = 7 Hz, 2H), Chem. Soc. 96, 7812-7814. 3.68 (br s, 4H), 4.5-4.3 (m, 2H), 4.8-4.6 (m, 2H), 6.08 (20) London, R. E., and Gabel, S. A. (1988) Determination of (br d , J = 8.5 H z , l H ) , 6 . 2 0 ( d , J = 13.5 Hz,lH),6.6-6.5 Membrane Potential and Cell Volume by 19F NMR Using (m, lH), 6.71 (d, J = 9 Hz, lH), 7.0-6.8 (m, 2H). 19F Trifluoroacetate and Trifluoroacetamide Probes. Biochem(DMSO-&): 42.4 (m, lF), 30.3 (m, 1F).
LITERATURE CITED (1) Cohen, R. D., and Iles, R. A. (1975) Intracellular pH: Measurement, Control, and Metabolic Interrelationships. In Crit. Rev. Clin. Lab. Sci. 6 , 101-143.
istry 28,2378-2382. (21) Still, W. C., Kahn, M., and Mitra, A. (1978) Rapid Chromatographic techniques for preparative separations with moderate resolution. J . Org. Chem. 43, 2923-2925. BC940087B
