Translational diffusion of macromolecule-sized solutes in cytoplasm and nucleus.

Fluorescence recovery after photobleaching (FRAP) was used to quantify the translational diffusion of microinjected FITC-dextrans and Ficolls in the cytoplasm and nucleus of MDCK epithelial cells and Swiss 3T3 fibroblasts. Absolute diffusion coefficients (D) were measured using a microsecond-resolution FRAP apparatus and solution standards. In aqueous media (viscosity 1 cP), D for the FITC-dextrans decreased from 75 to 8.4 x 10(-7) cm2/s with increasing dextran size (4-2,000 kD). D in cytoplasm relative to that in water (D/Do) was 0.26 +/- 0.01 (MDCK) and 0.27 +/- 0.01 (fibroblasts), and independent of FITC-dextran and Ficoll size (gyration radii [RG] 40-300 A). The fraction of mobile FITC-dextran molecules (fmob), determined by the extent of fluorescence recovery after spot photobleaching, was >>0.75 for RG << 200 A, but decreased to <<0.5 for RG >> 300 A. The independence of D/Do on FITC-dextran and Ficoll size does not support the concept of solute "sieving" (size-dependent diffusion) in cytoplasm. Photobleaching measurements using different spot diameters (1.5-4 micron) gave similar D/Do, indicating that microcompartments, if present, are of submicron size. Measurements of D/Do and fmob in concentrated dextran solutions, as well as in swollen and shrunken cells, suggested that the low fmob for very large macromolecules might be related to restrictions imposed by immobile obstacles (such as microcompartments) or to anomalous diffusion (such as percolation). In nucleus, D/Do was 0.25 +/- 0.02 (MDCK) and 0.27 +/- 0.03 (fibroblasts), and independent of solute size (RG 40-300 A). Our results indicate relatively free and rapid diffusion of macromolecule-sized solutes up to approximately 500 kD in cytoplasm and nucleus.

Second messengers regulate endosomal acidification in Swiss 3T3 fibroblasts.

Acidification of the endosomal pathway is important for ligand and receptor sorting, toxin activation, and protein degradation by lysosomal acid hydrolases. Fluorescent probes and imaging methods were developed to measure pH to better than 0.2 U accuracy in individual endocytic vesicles in Swiss 3T3 fibroblasts. Endosomes were pulse labeled with transferrin (Tf), alpha 2-macroglobulin (alpha 2M), or dextran, each conjugated with tetramethylrhodamine and carboxyfluorescein (for pH 5-8) or dichlorocarboxyfluorescein (for pH 4-6); pH in individual labeled vesicles was measured by ratio imaging using a cooled CCD camera and novel image analysis software. Tf-labeled endosomes acidified to pH 6.2 +/- 0.1 with a t1/2 of 4 min at 37 degrees C, and remained small and near the cell periphery. Dextran- and alpha 2M-labeled endosomes acidified to pH 4.7 +/- 0.2, becoming larger and moving toward the nucleus over 30 min; approximately 15% of alpha 2M-labeled endosomes were strongly acidic (pH less than 5.5) at only 1 min after labeling. Replacement of external Cl by NO3 or isethionate strongly and reversibly inhibited acidification. Addition of ouabain (1 mM) at the time of labeling strongly enhanced acidification in the first 5 min; Tf-labeled endosomes acidified to pH 5.3 without a change in morphology. Activation of phospholipase C by vasopressin (50 nM) enhanced acidification of early endosomes; activation of protein kinase C by PMA (100 nM) enhanced acidification strongly, whereas elevation of intracellular Ca by A23187 (1 microM) had no effect on acidification. Activation of protein kinase A by CPT-cAMP (0.5 mM) or forskolin (50 microM) inhibited acidification. Lysosomal pH was not affected by ouabain or the protein kinase activators. These results establish a methodology for quantitative measurement of pH in individual endocytic vesicles, and demonstrate that acidification of endosomes labeled with Tf and alpha 2M (receptor-mediated endocytosis) and dextran (fluid-phase endocytosis) is sensitive to intracellular anion composition, Na/K pump inhibition, and multiple intracellular second messengers.

Fluorescent chloride indicators to assess the efficacy of CFTR cDNA delivery.

Cl(-)-sensitive fluorescent indicators have been used extensively in cell culture systems to measure the Cl(-)-transporting function of the cystic fibrosis transmembrane conductance regulator protein CFTR. These indicators have been used in establishing a surrogate end point to assess the efficacy of CFTR cDNA delivery in human gene therapy trials. The ability to measure Cl- transport with high sensitivity in small and heterogeneous tissue samples makes the use of Cl- indicators potentially attractive in gene delivery studies. In this review article, the important technical aspects of Cl- transport measurements by fluorescent indicators such as SPQ are described, applications of Cl- indicators to assay CFTR function are critically evaluated, and new methodological developments are discussed. The available Cl- indicators have been effective in quantifying Cl- transport rates in cell culture models and in vitro systems such as isolated membrane vesicles and liposomes. However, the imperfect photophysical properties of existing Cl- indicators limit their utility in performing measurements in airway tissues, where gene transfer vectors are delivered in CF gene therapy trials. The low efficiency of gene transfer and the cellular heterogeneity in airway samples pose substantial obstacles to functional measurements of CFTR expression. Significant new developments in generating long-wavelength and dual-wavelength halide indicators are described, and recommendations are proposed for the use of the indicators in gene therapy trials.

Chemical chaperones correct the mutant phenotype of the delta F508 cystic fibrosis transmembrane conductance regulator protein.

Mutations in the cystic fibrosis transmembrane conductance regulator protein (CFTR) often result in a failure of the protein to be properly processed at the level of the endoplasmic reticulum (ER) and subsequently transported to the plasma membrane. The folding defect associated with the most common CFTR mutation (delta F508) has been shown to be temperature sensitive. Incubation of cells expressing delta F508 CFTR at lower growth temperatures results in the proper processing of a portion of the mutant CFTR protein. Under these conditions, the mutant protein can move to the plasma membrane where it functions, similar to the wild-type protein, in mediating chloride transport. We set out to identify other methods, which like temperature treatment, would rescue the folding defect associated with the delta F508 CFTR mutation. Here we show that treatment of cells expressing the delta F508 mutant with a number of low molecular weight compounds, all known to stabilize proteins in their native conformation, results in the correct processing of the mutant CFTR protein and its deposition at the plasma membrane. Such compounds included the cellular osmolytes glycerol and trimethylamine N-oxide, as well as deuterated water. Treatment of the delta F508 CFTR-expressing cells with any one of these compounds, which we now refer to as 'chemical chaperones', restored the ability of the mutant cells to exhibit forskolin-dependent chloride transport, similar to that observed for the cells expressing the wild-type CFTR protein. We suggest that the use of 'chemical chaperones' may prove to be effective for the treatment of cystic fibrosis, as well as other genetic diseases whose underlying basis involves defective protein folding and/or a failure in normal protein trafficking events.

A mouse model to test the in vivo efficacy of chemical chaperones.

In vitro studies in transfected cells have indicated that chemical chaperones including glycerol (0.5-1.2 M) and trimethylamine oxide (TMAO, 50-100 mM) can correct defective trafficking of some proteins, including deltaF508 CFTR in cystic fibrosis and AQP2 mutants in nephrogenic diabetes insipidus. To develop a mouse model to test the efficacy of chemical chaperones in vivo, glycerol and TMAO were administered by intraperitoneal (i.p.), subcutaneous (s.c.), and oral routes. Glycerol and TMAO assays that utilized 1-5 microL of tail vein blood were developed. Administration by the i.p. and s.c. routes gave maximum serum glycerol concentrations of approximately 100 mM, levels that were well below the effective in vitro concentrations. Single i.p. or s.c. doses of TMAO (7 g/kg, 8% solution in water) resulted in serum [TMAO] greater than 50 mM, with a long half-life (t1/2 approximately equal to 18-21 h). Sustained high serum and tissue [TMAO] > 52 mM for 3 days was achieved by s.c. administration of TMAO (7 g/kg) in water every 8 h. Although approximately 50% of the mice died with this multiple-dose regimen, the remaining mice had nearly normal liver, renal, and pancreatic function. A lower dose of TMAO (5 g/kg) given by the s.c. route every 8 h resulted in serum [TMAO] concentration of 22 mM, a level that was well tolerated by all mice for 72 h. These mice also had high [TMAO] in urine, 400 mM. These results demonstrate that potentially therapeutic concentrations of TMAO can be sustained in mice in vivo, permitting the testing of chemical chaperones in transgenic mouse models of diseases caused by defective protein trafficking.

Cell-permeable fluorescent indicator for cytosolic chloride.

A major limitation of quinolinium-based fluorescent indicators for cytosolic Cl- has been the necessity of invasive cell loading because the positively charged ring nitrogen confers high polarity and membrane impermeability. A novel approach to mask the positive nitrogen was developed and evaluated for rapid, noninvasive indicator loading into living cells and effective intracellular trapping. The nonpolar and lipophilic compound 6-methoxy-N-ethyl-1,2-dihydroquinoline (diH-MEQ) was Cl- insensitive but was readily oxidized to the membrane-impermeable and Cl(-)-sensitive fluorescent indicator 6-methoxy-N-ethylquinolium chloride (MEQ), MEQ had 344-nm absorbance and 440-nm emission maxima, 0.70 quantum yield, and 4100 M-1 cm-1 molar extinction coefficient. In aqueous buffers, the fluorescence of MEQ was quenched by Cl- by a collisional mechanism with a Stern-Volmer constant (KCl) of 145 M-1. MEQ fluorescence was quenched by other anions (KBr = 275 M-1, KI = 360 M-1, KSCN = 300 M-1) but not by NO3-, SO4(2-), cations, and pH. Swiss 3T3 fibroblasts and colonic T84 cells were loaded with MEQ by incubation at 37 degrees C with 25-50 microM diH-MEQ for 5-10 min followed by diH-MEQ-free buffer for 15 min. MEQ stained cells brightly and uniformly and was nontoxic in studies of cell growth, cAMP and Ca2+ signaling, and electrophysiological properties. MEQ leaked out of cells by less than 10% in 60 min and was sensitive to cytosolic Cl- with KCl = 19 M-1.(ABSTRACT TRUNCATED AT 250 WORDS)

Functional CFTR in endosomal compartment of CFTR-expressing fibroblasts and T84 cells.

It was proposed that the cystic fibrosis transmembrane conductance regulator (CFTR) functions in the endosomal compartment as a adenosine 3',5'-cyclic monophosphate (cAMP)-regulated Cl channel that regulates endosomal acidification (J. Barasch, B. Kiss, A. Prince, L. Saiman, D. Gruenert, and A. Al-Awqati, Nature Lond. 352: 70-73, 1991). This hypothesis was tested in stably transfected Swiss 3T3 fibroblasts expressing CFTR or delta F508 CFTR and in T84 epithelial cells that normally express CFTR. In fibroblasts, the time course of pH in individual endosomes was measured by quantitative image analysis after 1 min pulse labeling with 2 microM carboxyfluorescein (Cf)-tetramethylrhodamine-transferrin (K. Zen, J. Biwersi, N. Periasamy, and A. S. Verkman. J. Cell Biol. 119: 99-110, 1992). Average endosomal pH reached 6.20 +/- 0.07 (SE) after 15 min in the mock-transfected cells with a half time of approximately 3 min; pH was slightly lower (5.97 +/- 0.06) in the CFTR-expressing fibroblasts. The difference did not result from a subpopulation of highly acidic endosomes. Forskolin (10 microM) increased average pH to 6.62 +/- 0.03 and abolished the difference. For determination of Cl conductance, endosomes in fibroblasts and T84 cells were labeled with Cf-dextran (5 mg/ml); dissipation of the endosomal pH gradient was measured in response to rapid addition of the protonophore carbonyl cyanide m-chlorophenylhydrazone (CCCP; 20 microM). Because the proton flux across the endosomal membrane is limited by the movement of K and Cl, the rate of alkalinization (dpH/dt) after CCCP addition provided a measure of endosomal Cl conductance. In CFTR-expressing fibroblasts, forskolin (10 microM) increased dpH/dt 1.6 +/- 0.2-fold (n = 14).(ABSTRACT TRUNCATED AT 250 WORDS)

Cystic fibrosis transmembrane conductance regulator activation stimulates endosome fusion in vivo.

Previous studies have suggested a role for cystic fibrosis transmembrane conductance regulator (CFTR) in the regulation of intracellular vesicular trafficking. A quantitative fluorescence method was used to test the hypothesis that CFTR expression and activation affects endosome-endosome fusion in intact cells. Endosomes from CFTR-expressing and control (vector-transfected) Swiss 3T3 fibroblasts were labeled by internalization with 4,4-difluoro-5,7-dimethyl-4-bora-3a, 4a-diaza-s-indacene (Bodipy)-avidin, a fluid-phase marker whose fluorescence increases approximately 8-fold upon biotin binding. Cells were washed, chased, and then labeled with biotin-albumin or biotin-transferrin. The fraction of Bodipy-avidin-labeled endosomes that fused with biotin-containing endosomes (f(fusion)) was quantified by ratio imaging microfluorimetry. Endosome fusion in unstimulated CFTR-expressing cells was similar to that in control cells. However, in CFTR-expressing cells activated by forskolin, ffusion was increased by 1.30 +/- 0.18- and 2.65 +/- 0.17-fold for a 0 and 10 min chase time between avidin and biotin-albumin pulses; f(fusion) also increased (1.32 +/- 0.11-fold) when biotin-transferrin replaced biotin-albumin. The stimulation of endosome fusion was not due to differences in rates of endocytosis or endosomal acidification. Endosome fusion was not stimulated by forskolin in Cl--depleted CFTR-expressing cells, suggesting that the increase in endosome fusion is due to the CFTR chloride channel activity. These results provide evidence that CFTR is involved in the regulation of endosome fusion and, thus, a possible basis for the cellular defects associated with cystic fibrosis.

Imaging of endosome fusion in BHK fibroblasts based on a novel fluorimetric avidin-biotin binding assay.

A fluorescence assay of in vivo endosome fusion was developed and applied to define the kinetics of endosome fusion in baby hamster kidney (BHK) fibroblasts. The assay is based on an approximately 10-fold enhancement of the green fluorescence of BODIPY-avidin upon biotin binding. The BODIPY-avidin fluorescence enhancement occurred in < 25 ms, was pH-independent, and involved a BODIPY-tryptophan interaction. For endocytosis in vivo, BHK fibroblasts were pulse-labeled with BODIPY-avidin together with a red (rhodamine) fluorescent fusion-independent chromophore (TMR). After specified chase times in a nonfluorescent medium, a second cohort of endosomes was pulse-labeled with biotin-conjugated albumin, dextran, or transferrin. Fusion of biotin-containing endosomes with avidin-containing endosomes was quantified by ratio imaging of BODIPY-to-TMR fluorescence in individual endosomes, using imaging methods developed for endosome pH studies. Analysis of BODIPY-to-TMR ratio distributions in avidin-labeled endosomes exposed to zero and maximum biotin indicated > 90% sensitivity for detection of endosome fusion. In avidin pulse (10 min) -chase-biotin albumin pulse (10 min) studies, both fused and unfused endosomes were identified; the fractions of avidin-labeled endosomes that fused with biotin-labeled endosomes were 0.48, 0.21, 0.16, and 0.07 for 0-, 5-, 10-, and 20-min chase times. Fitting of fusion data to a mathematical model of in vivo endosome fusion required the existence of an intermediate fusion compartment. Pulse-chase studies performed with biotin-transferrin to label the early/recycling endosomes indicated that after a 10-min chase, avidin-labeled endosomes reached a compartment that was inaccessible to biotin-transferrin. The assay was also applied to determine whether endosome fusion was influenced by temperature, pH (bafilomycin A1), second messengers (cAMP agonists, phorbol 12-myristate 13-acetate, staurosporine), and growth-related factors (platelet-derived growth factor, genistein). The results establish a sensitive fluorescence assay to quantify the fusion of vesicular compartments in living cells.

Direct measurement of trans-Golgi pH in living cells and regulation by second messengers.

In the endocytic compartment, an acidic pH plays a key role in receptor and ligand sorting, vesicular transport, and protein degradation. In the secretory compartment, indirect estimates of trans-Golgi pH based on partitioning of weak bases and following viral infection suggest a mildly acidic pH of > 6.0. We developed a liposome microinjection method to introduce fluorescent indicators into the aqueous compartment of trans-Golgi in living cells. In the presence of ATP and at 37 degrees C, 70-nm diameter liposomes delivered their fluid-phase contents selectively into the trans-Golgi compartment as assessed by colocalization with the trans-Golgi stain N-[6-[(7-nitrobenz-2-oxa-1, 3-diazol-4-yl)amino]caproyl]-sphingosine (C6-NBD-ceramide). Liposome fusion was ATP- and temperature-dependent and blocked by N-ethylmaleimide but not by guanosine 5'-O-(3-thiotriphosphate) (GTP gamma S). trans-Golgi pH in skin fibroblasts was 6.17 +/- 0.02 (S.E., n = 174) as measured by ratio imaging confocal microscopy using fluorescein and rhodamine-based indicators and an in vivo calibration procedure. trans-Golgi pH increased to 6.8 +/- 0.1 by cAMP agonists and to 6.5 +/- 0.1 by protein kinase C activation. These results provide the first direct measurement of trans-Golgi pH in living cells and demonstrate pH regulation by second messengers.

Long-wavelength chloride-sensitive fluorescent indicators.

The goal of this study was to develop long-wavelength fluorescent Cl indicators that have improved optical properties over quinolinium compounds. A series of quaternized tricyclic heterocycles was screened. We found that N,N-dimethyl-9,9-bisacridinium (lucigenin) had very high halide sensitivity with Stern-Volmer constants for collisional quenching of 390 M-1 (Cl), 585 M-1 (Br), 750 M-1 (I), and 590 M-1 (SCN), much higher than those for the reference compound 6-methoxy-N-(3-sulfopropyl)quinolinium (SPQ) (118 M-1 (Cl)); lucigenin fluorescence was insensitive to phosphate, sulfate, and nitrate. Lucigenin fluorescence at 505 nm was excited with molar absorbances of 34,000 M-1 cm-1 (368 nm) and 7500 M-1 cm-1 (455 nm). To examine structure-activity relationships, five 9-substituted N-methylacridinium compounds were synthesized (9-amino, 9-carboxaldehyde, 9-carboxamide, 9-N,N-dimethylcarboxamide, and 9-methylcarboxylate), of which N-methylacridinium-9-carboxamide (MACA) had the best optical properties. MACA had excitation and emission maxima at 424 and 500 nm, and Stern-Volmer constants of 225 M-1 (Cl), 480 M-1 (Br), 550 M-1 (I), and 480 M-1 (SCN). The quantum yields of lucigenin and MACA were 0.6-0.7. The acridinium compounds are useful as Cl indicators in liposomes and membrane vesicles, but are not stable in cell cytoplasm.

[Cyclic ureas. 1. Racemates and enantiomers of hexahydropyrimidine-2-ones: synthesis, configuration and sedative-hypnotic action]. / Zyklische Harnstoffe, 1. Mitt.: Racemate und Enantiomere von Hexahydropyrimidin-2-onen: Synthese, Konfiguration und sedativ-hypnotische Wirkung.

The racemates and the enantiomers of the hexahydropyrimidin-2-ones 2a-2e which can be regarded as cyclic ureas are obtained from the racemates and the enantiomers of the barbiturates 1a-1e by reduction with LiAlH4/AlCl3. The enantiomers of 2a-2d, dextrorotatory in ethanol, possess S-configuration. In a study with rats all cyclic ureas synthesized showed sedative-hypnotic activity. Some of the enantiomers exhibited marked enantioselecive differences in their potency.

[Cyclic ureas. 2. Racemates and enantiomers of imidazolidin-2-ones: synthesis, configuration and sedative-hypnotic action]. / Racemate und Enantiomere von Imidazolidin-2-onen Synthese, Konfiguration und sedativ-hypnotische Wirkung.

The racemates and the enantiomers of the imidazolidin-2-ones 2a and 2b, which can be considered as cyclic ureas, are obtained from the racemates and the enantiomers of the hydantoins 1a and 1b by reduction with LiAlH4/AlCl3. The enantiomers that are dextrorotating in ethanol possess S-configuration. In a study with Wistar-rats, 2a and 2b show sedative-hypnotic activity. The enantiomers exhibit marked enantioselective differences in their potency.

Evidence against defective trans-Golgi acidification in cystic fibrosis.

Defective organelle acidification has been proposed as a unifying hypothesis to explain the pleiotropic cellular abnormalities associated with cystic fibrosis. To test whether cystic fibrosis transmembrane conductance regulator (CFTR) participates in trans-Golgi pH regulation, intraluminal trans-Golgi pH was measured in stably transfected Swiss 3T3 fibroblasts (expressing CFTR or DeltaF508-CFTR) and CFTR-expressing and nonexpressing epithelial cells. trans-Golgi pH was measured by ratio-imaging confocal microscopy using a liposome injection procedure to label the lumen of trans-Golgi with fluid phase fluorescein and rhodamine chromophores (Seksek, O., Biwersi, J., and Verkman, A. S.(1995) J. Biol. Chem. 270, 4967-4970). Selective labeling of trans-Golgi was confirmed by colocalization of the delivered fluid phase fluorophores with N-(6-[(7-nitrobenzo-2-oxa-1, 3-diazol-4-yl)amino]caproyl)-sphingosine. In unstimulated fibroblasts in HCO3--free buffer, trans- Golgi pH was 6.25 +/- 0.04 (mean +/- S.E.; n = 80, vector control), 6.30 +/- 0.03 (n = 74, CFTR) and 6.23 +/- 0.06 (n = 60, DeltaF508) (not significant). After stimulation of plasma membrane Cl- conductance by 8-(4-chlorophenylthio)-cAMP (CPT-cAMP), trans-Golgi pH was 6.42 +/- 0.07 (n = 22, control), 6.47 +/- 0.07 (n = 20, CFTR), and 6.35 +/- 0. 07 (n = 22, DeltaF508) (not significant). Similarly, significant pH differences were not found for control versus CFTR-expressing cells in 25 mM HCO3- buffer. In epithelial cells, which do not express CFTR, trans-Golgi pH was (in 25 mM HCO3-) 6.36 +/- 0.04 (n = 33) and 6.34 +/- 0.08 (n = 23, CPT-cAMP) in MDCK cells and 6.25 +/- 0.04 (n = 18) and 6.24 +/- 0.06 (n = 15, CPT-cAMP) in SK-MES-1 cells. In Calu-3 cells, which natively express CFTR, trans-Golgi pH was (in 25 mM HCO3-) 6.19 +/- 0.05 (n = 25) and 6.17 +/- 0.08 (n = 23, CPT-cAMP). To test whether CFTR expression affects pH in the endosomal compartment in HCO3- buffer, pH was measured by ratio imaging in individual endosomes labeled with fluorescein-rhodamine dextrans. Comparing control and CFTR-expressing fibroblasts, average endosome pH (range, 5.40-5.53 after 10 min; 4.79-4.89, 30 min) differed by <0.13 unit, both before and after cAMP stimulation. These results indicate that CFTR expression and activation do not influence pH in the trans-Golgi and endosomal compartments, providing direct evidence against the defective acidification hypothesis.

Cyclic ureas, III: On the metabolism of the enantiomers of cyclic ureas in rats.

Studies are described on the metabolism of the enantiomers of 1, 2, and 3 in rats. The metabolites were identified in urine after cleavage of conjugates and extraction using gas chromatography-mass spectrometry. Qualitative enantioselective differences of the metabolism of 1, 2, and 3 could not be detected. Because oxidation of 1 and 2 to the corresponding barbiturates could not be found and 3 was oxidized to mephenytoin only in minor amounts, this pathway is not a prerequisite for the sedative-hypnotic effects of 1, 2, and 3.

Synthesis and characterization of dual-wavelength Cl--sensitive fluorescent indicators for ratio imaging.

The fluorescence of quinolinium-based Cl- indicators such as 6-methoxy-N-(3-sulfopropyl)quinolinium (SPQ) is quenched by Cl- by a collisional mechanism without change in spectral shape. A series of "chimeric" dual-wavelength Cl- indicators were synthesized by conjugating Cl--sensitive and -insensitive chromophores with spacers. The SPQ chromophore (N-substituted 6-methoxyquinolinium; MQ) was selected as the Cl--sensitive moiety [excitation wavelength (lambdaex) 350 nm, emission wavelength (lambdaem) 450 nm]. N-substituted 6-aminoquinolinium (AQ) was chosen as the Cl--insensitive moiety because of its different spectral characteristics (lambdaex 380 nm, lambdaem 546 nm), insensitivity to Cl-, positive charge (to minimize quenching by chromophore stacking/electron transfer), and reducibility (for noninvasive cell loading). The dual-wavelength indicators were stable and nontoxic in cells and were distributed uniformly in cytoplasm, with occasional staining of the nucleus. The brightest and most Cl--sensitive indicators were alpha-MQ-alpha'-dimethyl-AQ-xylene dichloride and trans-1, 2-bis(4-[1-alpha'-MQ-1'-alpha'-dimethyl-AQ-xylyl]-pyridinium)ethyl ene (bis-DMXPQ). At 365-nm excitation, emission maxima were at 450 nm (Cl- sensitive; Stern-Volmer constants 82 and 98 M-1) and 565 nm (Cl- insensitive). Cystic fibrosis transmembrane conductance regulator-expressing Swiss 3T3 fibroblasts were labeled with bis-DMXPQ by hypotonic shock or were labeled with its uncharged reduced form (octahydro-bis-DMXPQ) by brief incubation (20 microM, 10 min). Changes in Cl- concentration in response to Cl-/nitrate exchange were recorded by emission ratio imaging (450/565 nm) at 365-nm excitation wavelength. These results establish a first-generation set of chimeric bisquinolinium Cl- indicators for ratiometric measurement of Cl- concentration.

Synthesis of cell-impermeable Cl-sensitive fluorescent indicators with improved sensitivity and optical properties.

Quinolinium compounds have been used as Cl-sensitive fluorescent indicators in cells and cell-free membrane fractions. To improve Cl sensitivity and for conjugation via nucleophilic reaction, the compounds 6-methoxy-N-(n-aminoalkyl)quinolinium bromide hydrochloride (AAQ) with alkyl chain lengths (n) of 2 (AEQ), 3 (APQ), and 4 (ABQ) were synthesized. AAQ was water soluble, fluorescent, and quenched by Cl. The Stern-Volmer constants (KCl) for quenching of protonated AEQ, APQ and ABQ by Cl were 354, 322, and 272 M-1, respectively, higher than KCl for 6-methoxy-N-(3-sulfopropyl)quinolinium (SPQ; 118 M-1). To eliminate pH-dependent fluorescence, 6-methoxy-N-(3-trimethylammoniumpropyl)quinolinium dibromide (TMAPQ) was synthesized (KCl, 310 M-1). To red shift fluorescence excitation and emission spectra, 6-phenyl-N-(3-trimethylammoniumpropyl)quinolinium dibromide (phenyl-TMAPQ) (emission 475 nm) and N-(3-trimethylammoniumpropyl)phenanthridinium dibromide (TMAPP) (excitation 380 nm) were synthesized. AEQ and ABQ were conjugated with neutral dextran activated by cyanogen bromide to give indicator-to-dextran mole ratios of 5 to 20. KCl values at pH 7.4 were 132 (AEQ-dextran) and 237 M-1 (ABQ-dextran). To construct a single molecule with Cl-sensitive and insensitive moieties, the bichromophores 6-methoxy-N-(n- dansylsulfonamidoalkyl)quinolinium with alkyl chains of two and four were synthesized. The new Cl-sensitive indicators were used for measurement of intracellular Cl activity and for the labeling of endocytic vesicles in 3T3 fibroblasts and T84 cells. Our results indicate that N-substitution of quinoline with positively charged moieties gives increased Cl sensitivity, and extension of ring conjugation gives indicators with red-shifted fluorescence spectra.

A point mutation at cysteine 189 blocks the water permeability of rat kidney water channel CHIP28k.

CHIP28k is an important water-transporting protein in the kidney proximal tubule and the thin descending limb of Henle [Zhang, Skach, Hasegawa, Van Hoek, & Verkman (1993) J. Cell Biol. 120, 359-369] that is homologous to human erythrocyte CHIP28 [Preston & Agre (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 11110-11114]. Oligonucleotide-directed mutagenesis was used to identify the cysteine(s) involved in inhibition of the water-transporting function of CHIP28k by the mercurial HgCl2. Each of the four cysteines (at positions 87, 102, 152, and 189) were mutated to serine individually, or in combinations. In vitro transcribed cRNA was expressed in Xenopus oocytes for measurement of osmotic water permeability (Pf) in the absence or presence of 0.3 mM HgCl2. Pf (in cm/s x 10(-4) measured at 10 degrees C) was 7 +/- 1 in water-injected oocytes. In wild-type CHIP28k, Pf was 58 +/- 7 (-HgCl2) and 12 +/- 1 (+HgCl2). Mutation of cysteine 87, 102, or 152, individually or in combinations, had little effect on oocyte Pf or on the inhibition by HgCl2. Mutation of cysteine 189 to serine or glycine gave similar Pf values of 49-56 (-HgCl2); however, Pf was not inhibited up to 1 mM HgCl2. Mutation of cysteine 189 to the larger amino acid tryptophan gave a low Pf of 9 +/- 1; coexpression with wild-type CHIP28k indicated that the tryptophan mutation was not dominant negative. Mutation of the asparagine 42 and 205 glycosylation sites to threonine had little effect on Pf.(ABSTRACT TRUNCATED AT 250 WORDS)

Secondary structure analysis of purified functional CHIP28 water channels by CD and FTIR spectroscopy.

The integral membrane protein CHIP28 is an important water channel in erythrocytes and kidney tubule epithelia and is a member of a family of channel/pore proteins including the lens protein MIP26. The purposes of this study were to purify functional, delipidated CHIP28 to homogeneity and to determine secondary structure by circular dichroism (CD) and Fourier transform infrared spectroscopy (FTIR). CHIP28 was initially purified and delipidated by anion-exchange chromatography following solubilization of N-lauroylsarcosine-stripped erythrocyte membranes with beta-octylglucoside (OG); MIP26 was initially purified and delipidated by anion-exchange chromatography following solubilization of urea-stripped bovine lens membranes by monomyristoylphosphatidylcholine. CHIP28 (glycosylated and nonglycosylated) and MIP26 were purified further by high-performance size-exclusion chromatography, eluting in OG as apparent dimers and tetramers, respectively. Proteoliposomes reconstituted with purified CHIP28 were highly water-permeable, with an osmotic water permeability Pf of 0.04 cm/s at 10 degrees C that was inhibited by 0.1 mM HgCl2. Proteoliposomes reconstituted with MIP26 had a low Pf of 0.005 cm/s. CD spectra of CHIP28 in OG or in reconstituted proteoliposomes gave a maximum at 193 nm and minima at 208 and 222 nm. Spectral decomposition using protein basis spectra gave 40 +/- 5% alpha-helix and 43 +/- 3% beta-sheet and -turn. HgCl2 did not affect the CD spectrum of CHIP28. Attenuated total reflectance FTIR of air-dried, membrane-associated CHIP28 gave 38 +/- 5% alpha-helix and 40 +/- 4% beta-sheet and -turn by spectral decomposition of the amide I resonance. For comparison, CD of MIP26 in OG gave 49 +/- 7% alpha-helix and 32 +/- 12% beta-sheet and -turn; FTIR gave 32 +/- 8% alpha-helix and 45 +/- 6% beta-sheet and -turn. Analysis of CHIP28 and MIP26 sequence data by the generalized hydropathy method of Jähnig [Jähnig, F. (1990) Trends Biochem. Sci. 15, 93-95] predicted 39-47% alpha-helix and 15-20% beta-structures. These results establish procedures to obtain large quantities of pure CHIP28 and MIP26 in functional forms and provide evidence for multiple membrane-spanning alpha-helices or mixed alpha/beta-domains.