Abstract
The yeast Saccharomyces cerevisiae was engineered to express the ρ1 subunit of the human γ-aminobutyric acid ρ1 (GABAρ1) receptor. RNA that was isolated from several transformed yeast strains produced fully functional GABA receptors in Xenopus oocytes. The GABA currents elicited in the oocytes were fast, nondesensitizing chloride currents; and the order of agonist potency was GABA > β-alanine > glycine. Moreover, the receptors were resistant to bicuculline, strongly antagonized by (1,2,5,6 tetrahydropyridine-4-yl)methylphosphinic acid, and modulated by zinc and lanthanum. Thus, the GABA receptors expressed by the yeast mRNA retained all of the principal characteristics of receptors expressed by cRNA or native retina mRNAs. Western blot assays showed immunoreactivity in yeast plasma membrane preparations, and a ρ1-GFP fusion gene showed mostly intracellular distribution with a faint fluorescence toward the plasma membrane. In situ immunodetection of ρ1 in yeast demonstrated that some receptors reach the plasma membrane. Furthermore, microtransplantation of yeast plasma membranes to frog oocytes resulted in the incorporation of a small number of functional yeast ρ1 receptors into the oocyte plasma membrane. These results show that yeast may be useful to produce complete functional ionotropic receptors suitable for structural analysis.
Keywords: GABAC, recombinant expression, yeast
Fast neuronal inhibition in vertebrates is mediated mainly through γ-aminobutyric acid (GABA)-activated chloride receptor channels, which are genetically related to the nicotinic acetylcholine (nAChR), glycine, and serotonin 3 receptors. Most of the structural knowledge on ionotropic neurotransmitter receptors is based on data obtained from the Torpedo electric organ nAChR (1), and more recently from crystals of the soluble acetylcholine-binding protein (AchBP) (2). Thus, it is mainly by analogy to the nAChRs that the stoichiometry of ionotropic GABA receptors is assumed to be pentameric. Some electron microscope images of native GABAA receptors suggested a pentameric array with a central pore (3). However, efforts to obtain more detailed structural information of ionotropic GABA receptors have been greatly hampered by the present inability to overproduce, and crystallize, fully functional membrane proteins.
GABAC receptors are expressed most abundantly in the retina and are formed by the homomeric assembly of a single class of subunit (the ρ subunits). To date, three cDNAs have been cloned (ρ1–ρ3) with at least two splicing variants (4–6). Whether homomeric assembly of GABAC receptors occurs in vivo is still not clear. Nevertheless, when expressed in heterologous systems, such as frog oocytes or mammalian cells in culture, each of the ρ subunits produces receptors that gate low-desensitizing, bicuculline-resistant chloride currents that have properties similar to those of GABAC receptors from the vertebrate retina or of receptors expressed in frog oocytes by retina poly(A)+ RNA (7, 8).
Although a great deal of functional and pharmacological information exists on ionotropic GABA receptors, there is still a big gap concerning their structure. In view of the relative simplicity of GABAC receptors, we have taken the first steps toward producing the human ρ1 subunit in a recombinant system, aiming at expressing enough protein suitable for structural analysis. We opted for the baker's yeast Saccharomyces cerevisiae, because this is a simple eukaryote, easy to manipulate, and relatively inexpensive to grow, thus offering an attractive road toward the production of large amounts of recombinant receptor proteins.
To determine whether the yeast recombinant RNAs and proteins retain the properties of the original retina receptor we injected either the mRNA or plasma membranes from ρ1-transformed yeast into frog oocytes, followed by electrophysiological recordings of the receptors “expressed” in the oocyte's plasma membrane.
Materials and Methods
Plasmid Manipulations. Expression plasmids for the production of ρ1 or a fusion ρ1-GFP were based on the plasmid pYEX-BX (Clontech), which carries the Cu++ inducible transcriptional promoter of the S. cerevisiae metallothionein gene driving the expression of the gene of interest. The ρ1-GFP construct is known to produce fully functional GABA receptors when expressed in frog oocytes (9), thus offering the possibility of following up the expression of the chimeric receptor by simple fluorescence microscopy analysis.
ρ1 was shuttled as a BamHI–XhoI fragment into the BamHI–SalI sites of pYEX-BX, yielding the plasmid pYEXρ1. A three-way ligation was used to assemble the plasmid pYEXρ1-GFP. pYEX-BX was cut with BamHI and SalI and ligated to a BamHI–HindIII fragment carrying part of the 5′ end of ρ1 and a HindIII–XhoI fragment carrying the 3′ end of ρ1 fused to GFP. For ρ1 cRNA synthesis we used the plasmid pAV111 linearized with SalI as template, and in vitro transcription was done as described (10).
Transformation and Induction of Expression in S. cerevisiae. E. coli (DH5α) was transformed with pYEX-BX, pYEXρ1, or pYEXρ1-GFP by the CaCl2 protocol (11), and, after selection, a single colony was used to isolate the plasmids for yeast transformation. S. cerevisiae (DY150 MATa, ura3-52, leu 2-3, 112, trp1-1, ade2-1, his3-11 can1-100 from Clontech) was transformed with 10 μg of plasmid by using the LiCl2 method (PT3081-1 from Clontech). Transformed cells were plated on yeast nitrogen base (YNB) uracil-deficient selective agar. A second selection consisted in propagating colonies in YNB uracil-, leucine-, adenine-, tryptophan-, and histidine-deficient plates. Positively selected colonies were grown in liquid YNB-GM (growth media). After 18 h, CuSO4 was added to 0.5 mM. One hour after induction, cells were harvested by centrifugation at 700 × g for 5 min, washed three times with cold PBS, and stored at –70°C until used.
The strain TD4 (MATa, his4-519, ura3-52, leu2-2, leu2-112, trp1-1, gal-2, can1 [cir+]) has been reported to express nAChRs (12). Therefore, we also used this strain to see whether the yield of functional ρ1 receptors could be increased. TD4 was transformed and selected as described above for DY150, grown in yeast extract/peptone/dextrose broth supplemented with 1.0 mM CuSO4. Cells were harvested by centrifugation, frozen immediately in liquid nitrogen, and stored at –70°C until used.
RNA Extraction, Plasma Membrane Preparation, and Oocyte Expression. RNAs were extracted from nontransformed yeast or yeast expressing ρ1 or the ρ1-GFP fusion receptor. RNA was isolated from 48 h-induced yeast and was done basically as described (13). Poly(A)+ RNA was isolated by oligo(dT) affinity chromatography, and 50–100 ng of poly(A)+ were injected into Xenopus oocytes in a volume of 56 nl. As control for GABAC receptor expression we also injected oocytes with in vitro synthesized ρ1 cRNA (50 ng in 50 nl). Voltage-clamp recordings were performed (14) at –60 mV for RNA injected oocytes and at –80 mV for oocytes injected with yeast membranes, which were selected by using a sucrose gradient as described (15). Membranes were suspended at a protein concentration of 2 mg/ml. Fifty nanoliters of plasma membranes from ρ1 or ρ1-GFP-transformed yeast were injected per oocyte, and currents were recorded from 1 h to several days after injection.
RNA Assays. Ribonuclease protection assays (RPAs) were performed to determine the level of expression of ρ1 mRNA in transformed yeast at different stages of induction. RNA was extracted, as described above, from yeast 0, 1, 2, and 3 h and1–2 days after induction. The ρ1 probe was prepared by using the plasmid pAV111 as template (10). After PstI restriction, the 3′ overhangs were blunt-ended. An antisense RNA probe of the 3′ noncoding sequence was generated with the T3 RNA polymerase (Fisher), yielding a probe of 398 bases. The radiolabeled probe was prepared by using 50 μCi (1 Ci = 37 GBq) of [α-32P]UTP (Amersham Biosciences). RPA was performed with the RPA III kit (Ambion, Austin, TX) by using 15 μg of total RNA and 7 × 105 cpm of probe per sample. Controls using yeast total RNA were set: one was digested with RNase, whereas the other was not digested to assess the integrity of the probe and to detect any RNase contamination during the procedure.
Western Blots, Immunodetection, and Confocal Microscopy. Immunoblots of extracted yeast membranes were performed by using either a rabbit antiserum against a synthetic 15-mer peptide of the N terminus of ρ1 (HEMSKKGRPQRQRRE) made on request (Sigma-Genosys) at 1:250 or a rabbit serum anti-GFP (BioReagents) 1:2,000. In both cases, the second antibody was a goat IgG against rabbit IgG (Sigma) coupled to alkaline phosphatase 1:100,000.
Living yeast cells were processed for confocal microscopy as described (16). For cells expressing ρ1-GFP, a double filter unit was used, exciting at 580 nm. Immunofluorescence labeling with Texas red was done on nonpermeabilized yeast by using a 1:100 dilution of the rabbit anti ρ1 and goat anti-rabbit IgG-Texas red (Molecular Probes).
Results
RNA from ρ1-Transformed Yeast Produces Functional GABA Receptors in Frog Oocytes. After repeated trials, using several conditions for expression of ρ1 (temperature shifts from 25 to 30°C; CuSO4 concentration 0.01–1.0 mM), the yeast did not produce large amounts of the receptor, as judged by SDS/PAGE (data not shown). To investigate whether the low level of expression was due to problems in either transcription or translation, we decided to determine whether the ρ1 mRNA was correctly expressed in the yeast. To monitor yeast transcription of ρ1 mRNA after Cu++ induction we used RPAs. After induction, we detected increasing levels of the ρ1 transcript, although some hybridization was detected even before Cu++ was added (Fig. 1A). This spurious expression is attested by low levels of fluorescence in noninduced ρ1-GFP-transformed yeast (see below). The probe was annealed to the noncoding 3′ end of ρ1, which secured that the transcript was properly extended up to the transcriptional stop signal. The functional properties of the receptors encoded by the transformed yeast mRNA were determined by using frog oocytes as an expression system.
Fig. 1.
Coding potential of mRNA isolated from yeast expressing ρ1. (A) RPA of the 3′ UTR of poly(A)+ ρ1-mRNA. A faint protected band is observed even before induction (0) and is expressed increasingly after induction (1–48 h). (B) Currents generated by glycine, β-Alanine, and GABA (all 1 mM) in a representative oocyte injected with RNA isolated from ρ1-expressing yeast (holding potential –60 mV). (C) Amplitude of the currents elicited by GABA, β-alanine, and glycine (1 mM, n = 10) in oocytes from one frog. The percent ratio for β-alanine/GABA and glycine/GABA was 39% and 4%, respectively. (D) GABA current–voltage relation in an oocyte injected with mRNA from ρ1-transformed yeast.
For example, when exposed to GABA (1 mM), oocytes injected with poly(A)+ RNA isolated from ρ1-transformed yeast or with in vitro transcribed ρ1 cRNA generated currents of 400 ± 50 and 2,000 ± 100 nA (each eight oocytes), respectively. In both cases, GABA elicited nondesensitizing inward currents. Because the amino acids glycine and β-alanine are agonists of ρ1 receptors (17), it was of interest to determine whether they exert a similar action on receptors encoded by the yeast mRNA. Glycine and β-alanine also elicited currents in oocytes injected with mRNA isolated from ρ1-transformed yeast (Fig. 1B). Mean amplitudes of the currents generated by GABA, β-alanine, or glycine (all 1 mM) in oocytes injected with yeast poly(A)+ RNA are illustrated in Fig. 1C. The agonist potencies were similar to those previously described for cloned ρ1 receptors, i.e., GABA (100%) > β-alanine (39%) > glycine (4%) (cf. 17).
To identify the ion permeating through the receptor channel, the current reversal potential was determined while activating the receptors with GABA. The voltage–current relation indicates a nonrectifying channel with a reversal potential around –20 mV (Fig. 1D), as expected for chloride ions in Xenopus oocytes using Ringer as external solution (18). The GABAC-specific antagonist 1,2,5,6 tetrahydropyridine-4-yl)methylphosphinic acid (TPMPA) (19) potently, and reversibly, reduced the GABA currents elicited in oocytes expressing the translated yeast ρ1 mRNA with an IC50 of 3.2 μM (Fig. 2). Moreover, it is known that zinc2+ and lanthanum3+ modulate the human ρ1 receptor, and Fig. 3A shows the current blocking effect of zinc (IC50 28 μM) on receptors expressed by the yeast mRNA. The blockage by zinc was not voltage-dependent (not shown), indicating that the zinc-binding site is not near or inside the channel pore. Although lanthanum alone did not induce significant currents when superfused onto oocytes expressing ρ1 receptors, it clearly acted as a positive modulator of the GABA receptors (Fig. 3B).
Fig. 2.
TPMPA blocks GABA currents in oocytes expressing yeast poly(A)+ mRNA. Increasing concentrations of TPMPA (0.3–10 μM) were added in the presence of GABA (1 μM), as indicated in Inset from an oocyte injected with mRNA from ρ1-expressing yeast. The TPMPA IC50 was 3.2 μM.
Fig. 3.
Zinc and lanthanum modulation of ρ1 receptors. (A) Zinc dose–current response relation in oocytes injected with poly(A)+ from yeast transformed with ρ1. The zinc IC50 was 28 μM. (B). Sample record illustrating the GABA-current potentiation by lanthanum.
Repeated injections of oocytes with poly(A)+ RNAs isolated from untransformed yeast failed to make the oocytes acquire the ability of generating GABA currents (data not shown). This indicates that the expressed ρ1 receptors derived exclusively from translation of the mRNA of yeast transformed with pYEXρ1.
Although we could not clearly detect a distinct protein band corresponding to the ρ1 receptor in SDS/PAGE of whole yeast extracts, it was evident that the ρ1 gene was successfully transcribed because ρ1 RNA was effectively produced. This suggests that a limiting step during, or after, translation of the ρ1 mRNA might be affecting the synthesis of the receptor in yeast. Therefore, we used other approaches to determine whether functional receptors were present in the plasma membrane of the transformed yeast.
The ρ1-GFP Fusion Is Translated in Yeast. To determine the cellular distribution of the receptors expressed, we used the pYEXρ1-GFP construct to study in vivo the expression of the receptors. Fig. 4A shows a Western blot of ρ1-GFP, using an antibody against GFP. Yeast subcellular compartments were fractionated by ultracentrifugation (see Materials and Methods), and a protein of 90 kDa was observed in all of the fractions analyzed, as expected for the fusion of the molecular masses of ρ1 (473 aa, 57–60 kDa) and GFP (238 aa, 28–30 kDa). Confocal laser microscopy imaging of yeast expressing ρ1GFP showed that the fluorescence occurs mainly as intracellular puncta, with some clusters close to the plasma membrane (Fig. 4B). In a few cells, a faint evenly distributed fluorescence was located near or at the plasma membrane (Fig. 4C). Oocytes injected with poly(A)+ RNA isolated from these yeast induced GABA currents with properties similar to those of oocytes injected with RNA of yeast expressing wtρ1, i.e., currents that were nondesensitizing, bicuculline-resistant, and strongly antagonized by TPMPA (Fig. 5). A faint green fluorescence was also detected in transformed but noninduced cultures. Because no fluorescence was detected in nontransformed yeast, or in those expressing ρ1, this indicates that the CUP1 promoter is not tightly regulated or that the presence of Cu++ or another metal in the growth media is constitutively activating the promoter.
Fig. 4.
ρ1 and ρ1-GFP are in plasma membrane fractions. (A) Western blot using anti-GFP antibody applied to sucrose gradients fractions during purification of plasma membranes from yeast expressing ρ1-GFP. Lane 1, crude extract; lane 2, crude membranes after differential centrifugation; lane 3, plasma membranes (43% sucrose); lane 4, flow-through (53% sucrose). Each lane was loaded with 15 μg of protein. (B) Confocal image of living yeast expressing ρ1-GFP superposing fluorescent and phase-contrast images. (C) Yeast from the same culture showing a faint evenly distributed fluorescence located at, or near, the plasma membrane. (D) GABAρ1 receptors detected by Western blots of whole yeast extracts (WY) and plasma membranes (M). (E) GABAρ1 receptors in the yeast plasma membrane. Nonpermeabilized yeast expressing GABAρ1 receptors were exposed to a rabbit antibody against the N-terminal of ρ1 followed by a second antibody labeled with Texas red.
Fig. 5.
TPMPA block of GABAρ1-GFP receptors expressed in an oocyte by transformed yeast mRNA. The holding potential was –60 mV.
ρ1 Is Sorted to the Yeast Plasma Membrane. Some plasma membrane localization of ρ1-GFP protein is suggested by Western blots of fractionated yeast extracts, as well as by the fluorescence of living cells. Again, Western blot assays using whole ρ1-yeast extracts did not show an obvious immunoreaction, whereas the fraction corresponding to the plasma membrane showed an immunoreacting protein band of ≈66 kDa (Fig. 4D). Moreover, using our antibody against ρ1 and IgG-Texas red as second antibody in nonpermeabilized yeast expressing ρ1 we found that some receptors are in the plasma membrane and that they have an extracellular antigenic site that is recognized by the ρ1 antibody (Fig. 4E). Although these results suggest strongly that ρ1 receptors are reaching the yeast plasma membrane, they do not demonstrate that the receptors are functional.
To determine whether the ρ1 receptors in the yeast membrane are functional, we injected oocytes with plasma membranes isolated from ρ1-expressing yeast, expecting to “transplant” to the oocyte plasma membrane the receptors already anchored in the yeast plasma membrane. This approach has been used successfully to microtransplant receptors from the electric organ of Torpedo and from the human brain (20, 21) to the oocyte plasma membrane. Membranes isolated form ρ1-expressing yeast induced the oocytes to acquire the ability to generate nondesensitizing bicuculline-resistant GABA currents. However, the amplitude of these currents was only 5–7 nA (GABA 1 mM), suggesting that the ρ1 receptors are not very abundant in the yeast plasma membrane, that they are not properly assembled, or that the oocyte plasma membrane cannot incorporate the yeast membranes as efficiently as it does other eukaryotic membranes.
Discussion
Gathering information on ionotropic receptors at the atomic level is a very difficult task because it has not been possible to crystallize the complete receptors. Therefore, as a first step toward establishing the ρ1 receptor as a model for detailed biophysical and structural studies we have expressed the ρ1 receptor in yeast. We have focused on ρ1, because this is a comparatively simple homo-oligomeric receptor that desensitizes very little and may thus help in obtaining more detailed structural/functional information. So far, S. cerevisiae has not efficiently overproduced functional ρ1 receptors. The results suggest that the expression problems may involve steps down-stream, or during, translation of the ρ1 mRNA. It has been argued that high concentrations of a cDNA or mRNA poorly predict the abundance of their encoded protein in yeast (22–24). However, we found abundant expression of both ρ1 and ρ1-GFP mRNAs, indicating an efficient transcription. Furthermore, the mRNA carries the entire ORF, as indicated by the RPA and by the substantial GABA currents generated by oocytes injected with mRNAs isolated from yeast induced to express either ρ1 or ρ1-GFP. Moreover, those receptors conserved the functional and pharmacological characteristics of typical GABAρ1 receptors, i.e., slow desensitization, insensitivity to bicuculline, TPMPA antagonism, and zinc and lanthanum modulation. Although the ρ1 RNA is abundant and translated by the yeast, the low yields of receptors indicates translational or posttranslational problems, such as protein truncation, read-through products, inclusion-aggregation, or hyperglycosylation.
The CUP1 promoter in pYEX-BX is putatively tightly regulated and independent of culture conditions (24). However, we observed a clear fluorescence (for ρ1-GFP) and spurious transcription (as detected by RPA for ρ1) even before Cu++ was added to the media, indicating that there is some transcriptional leakage. An explanation for this basal expression of RNAs could be the presence of traces of Cu++ or other inducing metals in the yeast nitrogen base (YNB) and yeast extract/peptone/dextrose broths.
The ρ1 and ρ1-GFP receptors expressed in yeast carry their own native-signal peptide and several lines of evidence indicate that the receptor is reaching the plasma membrane, most probably directed by this signal. Firstly, immunostaining of ρ1 in nonpermeabilized yeast locates the receptors at the plasma membrane. Second, Western blot analysis of cell fractions, selected by sucrose gradient and ultracentrifugation, disclosed ρ1 and GFP immunoreactive bands in the plasma membrane fraction. Finally, because noninjected oocytes, and oocytes injected with membranes from nontransformed yeast, did not respond to GABA, the GABA currents elicited in the oocytes injected with plasma membranes from transformed yeast are definitive proof that some GABA receptors are incorporated in the yeast plasma membrane and that these receptors are functional.
It has been shown previously that the assembly of nicotinic AchR subunits requires a proper N-glycosylation for efficient insertion of functional receptors into the plasma membrane of frog oocytes (25, 26). Hyperglycosylation is a problem encountered frequently during the expression of heterologous proteins in S. cerevisiae (24), and this could hinder the incorporation of functional GABAρ1 receptors into the yeast plasma membrane. On the other hand, Western blot assays of ρ1-expressing yeast occasionally showed anti-ρ1 immunoreactive bands of molecular weights higher than the one predicted by the receptor's primary sequence. Whether this discrepancy is due to differential glycosylation, clustering, or other modifications still remains to be elucidated.
In conclusion, our results demonstrate that the plasma membrane of ρ1-expressing yeast contains some GABAρ1 receptors that are functional when “microtransplanted” to Xenopus oocytes. Although the number of transplanted receptors is still rather small, these experiments point the way toward designing strategies aimed at increasing the number of fully functional neurotransmitter receptors in yeast. In addition, transfer of yeast plasma membranes or poly(A)+ RNA into frog oocytes, combined with electrophysiological techniques, provides a powerful method for studying the properties of yeast native ion channels and receptors as well as of heterologously expressed proteins.
Acknowledgments
We thank the University of California, Irvine, Imaging Facility for use of the confocal microscope and H. Nguyen and L. V. Mendez for technical assistance. A.M.-M. thanks the PEW Charitable Trusts for a Latin American Fellowship, and J.M.R.-R. thanks the Universidad Autónoma de Nuevo León and Consejo Nacional de Ciencia y Tecnología (Mexico) (CONACYT) for support. This work was supported by the National Science Foundation (Neural and Glial Mechanisms) and a University of California–Mexus grant to R.M. and A.M.-T. The cost of publication was partially defrayed by Universidad Nacional Autónoma de México (212702) and CONACYT (41309Q).
Abbreviations: GABA, γ-aminobutyric acid; RPA, ribonuclease protection assay; TPMPA, (1,2,5,6 tetrahydropyridine-4-yl)methylphosphinic acid.
References
- 1.Miyazawa, A., Fujiyoshi, Y., Stowell, M. & Unwin, N. (1999) J. Mol. Biol. 288, 765–786. [DOI] [PubMed] [Google Scholar]
- 2.Cromer, B. A., Morton, C. J. & Parker, M. W. (2002) Trends Biochem. Sci. 27280–27287. [DOI] [PubMed]
- 3.Nayeem, N., Green, T. P., Martin, I. L. & Barnard, E. A. (1994) J. Neurochem. 62, 815–818. [DOI] [PubMed] [Google Scholar]
- 4.Martinez-Torres, A., Vazquez, A. E., Panicker, M. M. & Miledi, R. (1998) Proc. Natl. Acad. Sci. USA 95, 4019–4022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Demuro, A., Martinez-Torres, A. & Miledi, R. (2000) Neurosci. Res. 36, 141–146. [DOI] [PubMed] [Google Scholar]
- 6.Enz, R. (2001) Biol. Chem. 382, 1111–1122. [DOI] [PubMed] [Google Scholar]
- 7.Cutting, G. R., Lu, L., O'Hara, B. F, Kasch, L. M., Montrose-Rafizadeh, C., Donovan, D. M., Shimada, S., Antonarakis, S. E., Guggino, W. B., Uhl, G. R., et al. (1991) Proc. Natl. Acad. Sci. USA 88, 2673–2677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Polenzani, L., Woodward, R. M. & Miledi, R. (1991) Proc. Natl. Acad. Sci. USA 88, 4318–4322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Martínez-Torres, A. & Miledi, R. (2001) Proc. Natl. Acad. Sci. USA 98, 1947–1951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Calvo, D. J., Vazquez, A. E. & Miledi, R. (1994) Proc. Natl. Acad. Sci. USA, 91, 12725–12729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Lab. Press, Plainview, NY), 2nd Ed., pp. 1.82–1.84.
- 12.Jansen, K. U., Conroy, W. G., Claudio, T., Fox, T. D., Fujita, N., Hamill, O., Lindstrom, J. M., Luther, M., Nelson, N., Ryan, K. A., et al. (1989) J. Biol. Chem. 264, 15022–15027. [PubMed] [Google Scholar]
- 13.Sherman, F. (1991) Methods Enzymol. 194, 3–21. [DOI] [PubMed] [Google Scholar]
- 14.Miledi, R. (1982) Proc. R. Soc. London 215, 491–497. [DOI] [PubMed] [Google Scholar]
- 15.Serrano, R. (1988) Methods Enzymol. 157, 533–544. [DOI] [PubMed] [Google Scholar]
- 16.Pringle, J. R., Adams, A. E. M., Drubin, D. G. & Haarer, B. K. (1991) Methods Enzymol. 194, 567–608. [DOI] [PubMed] [Google Scholar]
- 17.Calvo, D. J. & Miledi, R. (1995) NeuroReport 6, 1118–1120. [DOI] [PubMed] [Google Scholar]
- 18.Kusano, K., Miledi, R. & Stinnakre, J. (1982) J. Physiol. 328, 143–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ragozzino, D., Woodward, R. M., Murata, Y., Eusebi, F., Overman, L. E. & Miledi, R. (1996) Mol. Pharmacol. 50, 1024–1030. [PubMed] [Google Scholar]
- 20.Marsal, J., Tigyi, G. & Miledi, R. (1995) Proc. Natl. Acad. Sci. USA 92, 5224–5228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Miledi, R., Eusebi, F., Martinez-Torres, A., Palma, E. & Trettel, F. (2002) Proc. Natl. Acad. Sci. USA 99, 13238–13242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Blackstock, W. P. & Weir, W. P. (1999) Trends Biotechnol. 17, 121–127. [DOI] [PubMed] [Google Scholar]
- 23.Rudd, K. E., Humphery-Smith, I., Wasinger, V. C. & Bairoch, A. (1998) Electrophoresis 19, 536–544. [DOI] [PubMed] [Google Scholar]
- 24.Wasinger, V. C. & Humphery-Smith, I. (1998) FEMS Microbiol. Lett. 169, 375–382. [DOI] [PubMed] [Google Scholar]
- 25.Sumikawa, K., Parker, I. & Miledi, R. (1988) Brain Res. 464, 191–199. [DOI] [PubMed] [Google Scholar]
- 26.Sumikawa, K. & Miledi, R. (1989) Brain Res. Mol. Brain Res. 5, 183–192. [DOI] [PubMed] [Google Scholar]