- Email: [email protected]
Measurement of Intracellular Calcium in Cell Populations Loaded with Aequorin: Neurokinin-1 Responses in U373MG Cells
ANALYTICAL BIOCHEMISTRY ARTICLE NO.
236, 270–274 (1996)
0166
Measurement of Intracellular Calcium in Cell Populations Loaded with Aequorin: Neurokinin-1 Responses in U373MG Cells Lynne Hedley, Stephen B. Phagoo, and Iain F. James1 Sandoz Institute for Medical Research, 5 Gower Place, London WC1E 6BN, England
Received October 31, 1995
Changes in intracellular calcium concentration are important in mediating a wide variety of physiological responses. Recently there has been renewed interest in the use of aequorin, a protein from jellyfish that emits light when calcium is bound, to measure calcium levels in cells. We have loaded populations of cells from the human glioma line, U373MG, with aequorin. Lysis of aequorin-loaded but not control cells with detergent resulted in a luminescence signal that was dependent on extracellular calcium. Aequorin-loaded cells responded to substance P, histamine, or the calcium ionophore, ionomycin, with an increase in luminescence. Signals in response to detergent, ionomycin, or substance P could be detected up to 48 h after cells were loaded with aequorin. Other neurokinin-1 agonists but not agonists at neurokinin-2 or neurokinin3 receptors produced luminescence signals. Neurokinin-1 antagonists inhibited the substance P-induced signal. The aequorin-loading procedure worked well with U373MG cells but not with AR42J, CHO, IMR-90, or WI-38 cells. q 1996 Academic Press, Inc.
considering transfection of apoaequorin, we have investigated methods of loading cell populations with the aequorin protein. There are several reports of loading by various methods, including scrape loading (9), gravity loading (10), liposome fusion (11), permeabilization with EGTA (12, 13) or DMSO2 (14), hypoosmotic shock (15–17), cell hybridization (18), and release from micropinocytotic vesicles (19). The simplest of these methods is gravity loading. The procedure involves incubating cells with aequorin in calcium-free buffer, centrifuging, resuspending in medium, and plating out. The centrifugation step appears to be essential (10). We show here that the human glioma cell line, U373MG, loaded with aequorin in this way responded to substance P (SP), ionomycin, or histamine with an increase in luminescence. The responses could be obtained up to 48 h after loading with aequorin. The SP signal was mediated by neurokinin1 (NK1) receptors. MATERIALS AND METHODS
Materials The use of aequorin to measure intracellular calcium in mammalian cells is well established, especially for work with single cells after microinjection (for a review see Ref. 1). More recently, transfection of cells with recombinant apoaequorin has allowed measurement of calcium responses in cell populations. The ability to direct the expressed apoaequorin selectively to cytoplasm (2–4), mitochondria (5, 6), endoplasmic reticulum (7), or nucleus (8) is now providing new tools for the study of calcium changes in subcellular organelles. We have a requirement for measuring calcium levels in populations of cells on microtiter plates. As well as 1 To whom correspondence should be addressed. Fax: /44-171-3874116. E-mail: [email protected].
All cell lines were obtained from the European Collection of Animal Cell Cultures, Porton Down. HBSS, cell culture media, and supplements were from Gibco (Paisley, Scotland). Aequorin was from Friday Harbor Photoproteins (Friday Harbor, WA). SP, SP methyl ester, and [Sar9, Met(O2)11]SP were from Bachem (Bubendorf, Switzerland). Spantide II was from Neosytem Laboratoire (Strasbourg, France). [b-Ala8]neurokinin A and [Phe(Me)7]neurokinin B were from Novabiochem 2 Abbreviations used: DMSO, dimethyl sulfoxide; DMEM, Dulbecco’s modified Eagle’s medium; EC50 , the concentration of agonist required to produce 50% of the maximal response; FCS, fetal calf serum; HBSS, Hanks’ balanced salt solution; IC50 , the concentration of antagonist required to inhibit the agonist response by 50%; NK, neurokinin; PBS, phosphate-buffered saline; RLU, relative light units; SE, standard error of the mean; SP, substance P.
270
0003-2697/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
/ 6m14$$9507
03-27-96 14:32:40
abal
AP-Anal Bio
MEASURING NEUROKININ RESPONSES IN AEQUORIN-LOADED CELLS
271
(Nottingham, UK). NK1 antagonists were synthesized at the Sandoz Institute. Cell Culture U373MG cells were grown in minimum essential medium with Earle’s salts supplemented with 1% nonessential amino acids, 1 mM pyruvate, 2 mM Gln, and 10% FCS. WI-38 and IMR-90 cells were grown in DMEM with 1% nonessential amino acids, 2 mM Gln, and 10% FCS. AR42J cells were grown in Ham F-12 with 2 mM Gln and 15% FCS. CHO cells were grown in a 1:1 mixture of DMEM and Ham F-12 with 2 mM Gln and 10% FCS. Media for all the cell lines were also supplemented with 100 international units/ml penicillin and 100 mg/ml streptomycin. Loading with Aequorin Cells were loaded with aequorin by the method of Borle et al. (10). Confluent cells in four 800-ml flasks were washed twice with 25 ml of Ca-free PBS (in mM: 135 NaCl, 4 KCl, 0.15 KH2PO4 , 0.51 K2HPO4 , 11 glucose, pH 7.4) and then detached by incubating with Cafree PBS containing 1 mM EGTA. Cells were spun down and resuspended in 0.5 ml of ice-cold loading buffer (140 mM NaCl, 1 mM EGTA, 3 mM Hepes, pH 7.4) containing 50 mg aequorin. The suspension was incubated on ice for 10 min and then centrifuged at 500g for 5 min. The supernatant fluid was removed and the cells were gently resuspended in 40 ml of growth medium. This suspension was pipetted into wells in whitewalled microtiter plates (Packard viewplates) at a density of 20,000–60,000 cells/well. Cells were placed in the incubator and left until assayed. For some experiments cells were placed on 6-mm coverslips after loading. In this case cell density was about 100,000 cells/ coverslip. Measurement of Intracellular Calcium For cells in microtiter plates, the medium was removed and replaced with 100 ml HBSS buffered to pH 7.4 with 10 mM Hepes, containing antagonists when appropriate. The plate was incubated for 10 min at room temperature and then placed in the chamber of a Lab Systems Luminoskan plate reader. HBSS (20 ml) with or without agonist was injected into each well and the luminescence signal was integrated for 20 s after the injection. Cells on coverslips were removed from the medium, washed by dipping in a beaker of HBSS, and then placed in a holder and into a cuvette containing 0.5 ml HBSS at room temperature. The cuvette was placed in the chamber of a Bio-Orbit 1250 luminometer. Drugs were applied by injecting 250 ml of appropriate solution in HBSS into the cuvette. Luminescence was followed for up to 1 min.
/ 6m14$$9507
03-27-96 14:32:40
abal
FIG. 1. U373MG cells were loaded with aequorin as described under Materials and Methods, plated onto coverslips, and cultured overnight before assay (aequorin-loaded). An identical sample of cells was treated in the same way, but not exposed to aequorin (control cells). Cuvettes containing cells on coverslips were placed in the BioOrbit luminometer. Luminescence was recorded and Triton X-100 dissolved in either HBSS or Ca-free HBSS containing 6 mM EGTA was injected after 3 s. Data are means and SE from four runs. RLU is the abbreviation for relative light units.
RESULTS
Validation of the Loading Procedure Aequorin-loaded cells on coverslips gave a luminescence signal when treated with 0.3% Triton X-100 in HBSS. The detergent breaks open the cell and exposes the aequorin to extracellular calcium. There was no signal from control cells (not treated with aequorin), or from aequorin-loaded cells when the Triton was dissolved in Ca-free HBSS containing EGTA (Fig. 1). These results confirm that the luminescence signal comes from aequorin and is dependent on calcium. Aequorin-loaded cells on coverslips also responded to SP and ionomycin, but not to buffer injections. Control cells not exposed to aequorin gave no signal to SP or ionomycin (Fig. 2). Figure 3 shows a time course of responses to ionomycin and SP after loading with aequorin. These cells were loaded, placed in a microtiter plate, and assayed at different times using the Luminoskan luminometer. Immediately after loading there was a relatively large response to ionomycin and no response to SP. After 30 min, the response to ionomycin was reduced and the cells had started to respond to SP. After 60 min the response to ionomycin had stabilized at around four times lower than the initial response. The response to SP had reached a maximum. In this experiment, responses to both ionomycin and SP were followed up to 18 h after loading, and there was no decay. The pattern for luminescence induced with Triton X-100 was the same as that for ionomycin, but the signals were about 10-fold higher (see Fig. 5A). The responses
AP-Anal Bio
272
HEDLEY, PHAGOO, AND JAMES
FIG. 2. Responses of aequorin-loaded U373MG cells to ionomycin and SP. U373MG cells were loaded with aequorin, placed on coverslips, left in culture overnight, and assayed as described under Materials and Methods. Control cells were treated in the same way as loaded cells, but there was no aequorin in the loading buffer. Data are means and SE from four runs. (A) Response of aequorin-loaded and control cells to 10 mM ionomycin. (B) Response of aequorin-loaded cells to 100 nM SP or buffer.
to ionomycin, SP, and Triton X-100 could still be measured 48 h after loading. In some experiments there was no loss of signal; in others there was a slight loss. Overall responses to SP after 48 h were 95 { 35% (mean { SE, N Å 3) of the response after 18–22 h. The signal obtained from aequorin-loaded cells depended on the concentration of aequorin used in the loading procedure and on the density of cells. Good signals were obtained at between 20,000 and 60,000 cells/well.
FIG. 4. Log concentration–response curves for agonists in the aequorin assay. U373MG cells were loaded with aequorin and assayed as described under Materials and Methods. Agonist responses were normalized to the maximum aequorin signal, obtained by breaking open cells with 0.25% Triton X-100. (A) Curves for SP (filled squares), substance P-methyl ester (open triangles), and [Sar9, Met(O2)11]substance P (filled circles). (B) Curve for histamine. Note the different X axes.
Agonists Three NK1 agonists all produced concentration-related responses in the aequorin assay (Fig. 4A). There was little or no response to high concentrations of [bAla8]neurokinin A or [Phe(Me)7]neurokinin B, selective NK2 and NK3 receptor agonists, respectively (data not shown). Histamine also caused an increase in luminescence (Fig. 4B). The curves shown in Fig. 4 imply that histamine gave a bigger maximal response than the NK1 agonists. This was not always the case. In other experiments, where histamine and SP were tested on the same cells, we have seen smaller responses to histamine than SP. Estimates of EC50 values for the agonists are reported in Table 1. For SP, our value is higher than that obtained by measuring calcium changes with Fura-2 (20). For histamine, our estimate agrees well with the EC50 value reported for histamine-induced release of interleukin-1b in U373MG cells (21).
TABLE 1
Agonist Potencies in the Aequorin-Based Assay with U373MG Cells Agonist
FIG. 3. Time course of responses after loading U373MG cells with aequorin. Cells were loaded with aequorin and pipetted into microtiter plates and responses were measured as described under Materials and Methods. The concentration of ionomycin was 3 mM and SP was 100 nM. Data are means and SE from 12 replicates.
/ 6m14$$9507
03-27-96 14:32:40
abal
Substance P Substance P–methyl ester [Sar9, Met(O2)11]substance P Histamine
EC50 (nM) 31 37 96 7300
{ { { {
4 (10) 10 (2) 10 (3) 1400 (3)
Note. Data are means { SE from the number of independent experiments given in parentheses.
AP-Anal Bio
MEASURING NEUROKININ RESPONSES IN AEQUORIN-LOADED CELLS TABLE 2
Potencies of NK1 Antagonists in the Aequorin-Based Assay Compound CP 99994 RPR 100893 SR 140333 Spantide II
IC50 (nM) 3.0 14 1.2 5.0
{ { { {
0.49 (6) 2.7 (5) 0.35 (5) 2.0 (5)
Note. IC50 is the concentration of antagonist that gave 50% inhibition of the response to 30 nM SP. Data are means { SE from the number of independent experiments given in parentheses.
Neurokinin Antagonists The effects of an EC50 concentration of SP (30 nM) in aequorin-loaded U373MG cells were blocked by the NK1 antagonists CP99994 (22), RPR100893 (23), SR 140 333 (24), and Spantide II (25). IC50 values are reported in Table 2. All four NK1 antagonists were able to block the SP response completely with IC50 values of less than 15 nM. Loading Other Cell Lines We have investigated the use of the gravity loading procedure with other cell lines and receptor types. Our results indicate that U373MG cells were a fortunate choice for our first experiments. All the cell lines we tested took up the aequorin. There then appeared to be two patterns. In some cells (CHO and AR42J) aequorin activity declined rapidly to very low levels within 15 min; in others (U373MG, WI-38, and IMR-90) the decline was much slower (Fig. 5A). We tested agonist responses in cells that retained significant activity. For U373MG cells the agonist was SP; for WI-38 and IMR90 cells we used bradykinin to activate the B2 receptors on these cells. Only the U373MG cells gave consistent responses to agonist (Fig. 5B).
273
another agonist and suggest that the technique is applicable to other receptor systems. We loaded cells with the aequorin protein as a complementary approach to transfecting cells with aequorin cDNA. The advantage of using aequorin rather than calcium-sensitive fluorescent dyes is that the assay can be run conveniently on microtiter plates using luminometers that allow injection of reagent and immediate measurement of the signal. This protocol is essential because the calcium concentration is elevated for a relatively short time. At present, commercially available fluorimeters have no facility for reagent injection. The major disadvantage is that aequorin is less sensitive to calcium than the dyes. Practically, this means that it is difficult to measure background levels of calcium and the signal does not increase until the calcium concentration reaches probably several hundred nanomolar. The rather high EC50 value of about 30 nM that we report here for substance P, compared to those of 1–2 nM found with Fura2 measurements (20), may reflect the difference in assay sensitivity. In spite of these differences, our estimates of antagonist potencies agree well with published values (22, 25, 28). Transfection of apoaequorin into cells is likely to be a very useful technique both for routine screening assays and for investigating calcium changes in subcellular organelles (as discussed in the introduction) and is probably the method of choice in many circumstances. The loading technique may have some advantages, however. First, it is a relatively simple procedure that requires no skill in the techniques of molecular biology.
DISCUSSION
Activation of NK1 receptors in U373MG cells causes a variety of responses which include increases in levels of phophoinositides and intracellular calcium (20, 26– 28), translocation of protein kinase C from the cytoplasm to the membrane (26), release of taurine (26, 28), activation of c-fos and c-jun synthesis (27, 28), activation of an outward current in voltage-clamped cells (28), and release of interleukin-6 (21, 29, 30). Histamine also activates the phospholipase C system, probably through H1 receptors and causes release of interleukin-6 in these cells (21, 31). We have used U373MG cells to develop a convenient relatively high throughput assay for activation of NK1 receptors based on measurements of intracellular calcium with aequorin. The data with histamine simply show that we can activate the same pathways with
/ 6m14$$9507
03-27-96 14:32:40
abal
FIG. 5. Aequorin loading in other cell lines. Cells were loaded with aequorin as described under Materials and Methods. (A) Total aequorin activity in each cell line was estimated by breaking cells open with 0.25% Triton X-100 and exposing the aequorin to the extracellular calcium. (B) Responses to agonists; 100 nM SP for U373MG cells, 100 nM bradykinin for WI-38 and IMR-90 cells. Data are means of 12 determinations. Cell lines were U373MG (filled squares), WI-38 (filled circles), IMR-90 (open triangles), CHO (closed triangles), and AR42J (open circles).
AP-Anal Bio
274
HEDLEY, PHAGOO, AND JAMES
Second, there may well be advantages in loading other reagents along with the aequorin. For example, it may be possible to load cells efficiently with antisense oligonucleotides with the gravity loading technique. If the antisense is loaded along with the aequorin and is taken up by the same cells, then most of the signal should come from cells that have received antisense. This may allow the use of relatively low concentrations of oligonucleotides and at the same time ensure that there is no (or only a very low) background from untreated cells. The same could apply to other reagents that act inside cells but are not taken up efficiently, e.g., peptides. Application of the gravity loading procedure in these kinds of experiments may be limited. We have successfully used gravity loading only with U373MG cells. Of the cell lines we tested, two (CHO and AR42J) lost the aequorin very quickly. Two cell lines which retained the aequorin (WI-38 or IMR-90 cells) gave very little or no response to agonist. We do not know the reason for the differences between cell lines. Perhaps bradykinin receptors on WI-38 and IMR-90 cells were damaged during the loading procedure and did not recover over the period of the experiment. It is also possible that in some cells aequorin is sequestered in intracellular organelles and hence protected from changes in cytoplasmic calcium concentration. In summary, we have shown that U373MG cells can be loaded with aequorin by a simple method, and that the luminescence signal in loaded cells depends on calcium. Aequorin-loaded U373MG cells can be used to measure SP-stimulated increases in intracellular calcium. The SP effect is mediated by NK1 and not NK2 or NK3 receptors, since (a) NK1 antagonists inhibit the SP signal and (b) there is no or very little response to NK2- or NK3-selective agonists. This system provides a relatively rapid and convenient assay for NK1 antagonists. The loading method is not applicable to all cell lines. REFERENCES 1. Cobbold, P. H., and Rink, T. J. (1987) Biochem. J. 248, 313–328. 2. Sheu, Y-A., Kricka, L. J., and Pritchett, D. B. (1993) Anal. Biochem. 209, 343–347. 3. Button, D., and Brownstein, M. (1993) Cell Calcium 14, 663– 671. 4. Brini, M., Maesault, R., Bastianutto, C., Alvarez, J., Pozzan, T., and Rizutto, R. (1995) J. Biol. Chem. 270, 9896–9903. 5. Rutter, G. A., Theler, J-M., Murgia, M., Wollheim, C. B., Pozzan, T., and Rizutto, R. (1993) J. Biol. Chem. 268, 22385–22390.
/ 6m14$$9507
03-27-96 14:32:40
abal
6. Rizzuto, R., Simpson, A. W. M., Brini, M., and Pozzan, T. (1992) Nature 358, 325–327. 7. Kendall, J. M., Dormer, R. L., and Campbell, A. K. (1992) Biochem. Biophys. Res. Commun. 189, 1008–1016. 8. Badminton, M. N., Kendall, J. M., Sala-Newby, G., and Campbell, A. K. (1995) Exp. Cell Res. 216, 236–243. 9. McNeil, P. L., and Taylor, D. L. (1985) Cell Calcium 6, 83–93. 10. Borle, A. B., Freundrich, C. C., and Snowdowne, K. W. (1986) Am. J. Physiol. 251, C323–C326. 11. Hallet, M. B., and Campbell, A. K. (1980) Biochem. J. 192, 585– 596. 12. Morgan, J. P., DeFeo, T., and Morgan, K. G. (1984) Pflu¨gers Arch. 400, 338–340. 13. Johnson, P. C., Ware, J. A., Clivenden, P. B., Smith, M., Dvorak, A. M., and Salzman, E. W. (1985) J. Biol. Chem. 260, 2069– 2076. 14. Yamaguchi, A., Suzuki, H., Tanoue, K., and Yamazaki, H. (1986) Thromb. Res. 44, 165–174. 15. Borle, A. B., and Snowdowne, K. W. (1982) Science 217, 252– 254. 16. Dormer, R. L. (1983) Biosci. Rep. 3, 233–240. 17. Malmgren, R., Grunfelt, S., and Joseph, R. (1992) Thromb. Haemostasis 68, 352–356. 18. Campbell, A. K., and Hallet, M. B. (1983) J. Physiol. (London) 338, 537–550. 19. Hallet, M. B., and Campbell, A. K. (1983) Immunology 50, 487– 495. 20. Bordley, A., Feltz, P., and Trouslard (1994) Glia 11, 277–283. 21. Cadman, E. D., Witte, D. G., and Lee, C. M. (1994) J. Neurochem. 63, 980–987. 22. McLean, S., Ganong, A., Seymour, P. A., Snider, R. M., Desai, M. C., Rosen, T., Bryce, D. K., Longo, K. P., Reynolds, L. S., Robinson, G., and Scmidt, A. W. (1993) J. Pharmacol. Exp. Ther. 267, 472–479. 23. Tabart, M., and Peyronel, J. F. (1994) Bioorg. Med. Chem. Lett. 4, 673–676. 24. Emonds-Alt, X., Doutremepuich, J. D., Heaulme, M., Neliat, G., Santucci, V., Steinberg, R., Vilain, P., Bichon, D., Ducoux, J. P., and Proietto, V. (1993) Eur. J. Pharmacol. 250, 403–413. 25. Hakanson, R., Leander, S., Asano, N., Feng, D.-M., and Folkers, K. (1990) Regul. Peptides 31, 75–82. 26. Lee, C. M., Tung, W. L., and Young, J. D. (1992) J. Neurochem. 59, 406–414. 27. Eistetter, H. R., Mills, A., Brewster, R., Alouani, S., Rambosson, C., and Kawashima, E. (1992) Glia 6, 89–95. 28. Oury-Donat, F., Lefevre, I. A., Thurneyssen, O., Gauthier, T., Bordey, A., Feltz, P., Emonds-Alt, X., Le Fur, G., and Soubrie, P. (1994) J. Neurochem. 62, 1399–1407. 29. Palma, C., Goso, C., and Manzini, S. (1994) Neurosci. Lett. 171, 221–224. 30. Palma, C., Urbani, F., and Manzini, S. (1995) J. Neuroimmunol. 59, 155–163. 31. Arias-Montano, J. A., Berger, V., and Young, J. M. (1994) Br. J. Pharmacol. 111, 598–608.
AP-Anal Bio
236, 270–274 (1996)
0166
Measurement of Intracellular Calcium in Cell Populations Loaded with Aequorin: Neurokinin-1 Responses in U373MG Cells Lynne Hedley, Stephen B. Phagoo, and Iain F. James1 Sandoz Institute for Medical Research, 5 Gower Place, London WC1E 6BN, England
Received October 31, 1995
Changes in intracellular calcium concentration are important in mediating a wide variety of physiological responses. Recently there has been renewed interest in the use of aequorin, a protein from jellyfish that emits light when calcium is bound, to measure calcium levels in cells. We have loaded populations of cells from the human glioma line, U373MG, with aequorin. Lysis of aequorin-loaded but not control cells with detergent resulted in a luminescence signal that was dependent on extracellular calcium. Aequorin-loaded cells responded to substance P, histamine, or the calcium ionophore, ionomycin, with an increase in luminescence. Signals in response to detergent, ionomycin, or substance P could be detected up to 48 h after cells were loaded with aequorin. Other neurokinin-1 agonists but not agonists at neurokinin-2 or neurokinin3 receptors produced luminescence signals. Neurokinin-1 antagonists inhibited the substance P-induced signal. The aequorin-loading procedure worked well with U373MG cells but not with AR42J, CHO, IMR-90, or WI-38 cells. q 1996 Academic Press, Inc.
considering transfection of apoaequorin, we have investigated methods of loading cell populations with the aequorin protein. There are several reports of loading by various methods, including scrape loading (9), gravity loading (10), liposome fusion (11), permeabilization with EGTA (12, 13) or DMSO2 (14), hypoosmotic shock (15–17), cell hybridization (18), and release from micropinocytotic vesicles (19). The simplest of these methods is gravity loading. The procedure involves incubating cells with aequorin in calcium-free buffer, centrifuging, resuspending in medium, and plating out. The centrifugation step appears to be essential (10). We show here that the human glioma cell line, U373MG, loaded with aequorin in this way responded to substance P (SP), ionomycin, or histamine with an increase in luminescence. The responses could be obtained up to 48 h after loading with aequorin. The SP signal was mediated by neurokinin1 (NK1) receptors. MATERIALS AND METHODS
Materials The use of aequorin to measure intracellular calcium in mammalian cells is well established, especially for work with single cells after microinjection (for a review see Ref. 1). More recently, transfection of cells with recombinant apoaequorin has allowed measurement of calcium responses in cell populations. The ability to direct the expressed apoaequorin selectively to cytoplasm (2–4), mitochondria (5, 6), endoplasmic reticulum (7), or nucleus (8) is now providing new tools for the study of calcium changes in subcellular organelles. We have a requirement for measuring calcium levels in populations of cells on microtiter plates. As well as 1 To whom correspondence should be addressed. Fax: /44-171-3874116. E-mail: [email protected].
All cell lines were obtained from the European Collection of Animal Cell Cultures, Porton Down. HBSS, cell culture media, and supplements were from Gibco (Paisley, Scotland). Aequorin was from Friday Harbor Photoproteins (Friday Harbor, WA). SP, SP methyl ester, and [Sar9, Met(O2)11]SP were from Bachem (Bubendorf, Switzerland). Spantide II was from Neosytem Laboratoire (Strasbourg, France). [b-Ala8]neurokinin A and [Phe(Me)7]neurokinin B were from Novabiochem 2 Abbreviations used: DMSO, dimethyl sulfoxide; DMEM, Dulbecco’s modified Eagle’s medium; EC50 , the concentration of agonist required to produce 50% of the maximal response; FCS, fetal calf serum; HBSS, Hanks’ balanced salt solution; IC50 , the concentration of antagonist required to inhibit the agonist response by 50%; NK, neurokinin; PBS, phosphate-buffered saline; RLU, relative light units; SE, standard error of the mean; SP, substance P.
270
0003-2697/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
/ 6m14$$9507
03-27-96 14:32:40
abal
AP-Anal Bio
MEASURING NEUROKININ RESPONSES IN AEQUORIN-LOADED CELLS
271
(Nottingham, UK). NK1 antagonists were synthesized at the Sandoz Institute. Cell Culture U373MG cells were grown in minimum essential medium with Earle’s salts supplemented with 1% nonessential amino acids, 1 mM pyruvate, 2 mM Gln, and 10% FCS. WI-38 and IMR-90 cells were grown in DMEM with 1% nonessential amino acids, 2 mM Gln, and 10% FCS. AR42J cells were grown in Ham F-12 with 2 mM Gln and 15% FCS. CHO cells were grown in a 1:1 mixture of DMEM and Ham F-12 with 2 mM Gln and 10% FCS. Media for all the cell lines were also supplemented with 100 international units/ml penicillin and 100 mg/ml streptomycin. Loading with Aequorin Cells were loaded with aequorin by the method of Borle et al. (10). Confluent cells in four 800-ml flasks were washed twice with 25 ml of Ca-free PBS (in mM: 135 NaCl, 4 KCl, 0.15 KH2PO4 , 0.51 K2HPO4 , 11 glucose, pH 7.4) and then detached by incubating with Cafree PBS containing 1 mM EGTA. Cells were spun down and resuspended in 0.5 ml of ice-cold loading buffer (140 mM NaCl, 1 mM EGTA, 3 mM Hepes, pH 7.4) containing 50 mg aequorin. The suspension was incubated on ice for 10 min and then centrifuged at 500g for 5 min. The supernatant fluid was removed and the cells were gently resuspended in 40 ml of growth medium. This suspension was pipetted into wells in whitewalled microtiter plates (Packard viewplates) at a density of 20,000–60,000 cells/well. Cells were placed in the incubator and left until assayed. For some experiments cells were placed on 6-mm coverslips after loading. In this case cell density was about 100,000 cells/ coverslip. Measurement of Intracellular Calcium For cells in microtiter plates, the medium was removed and replaced with 100 ml HBSS buffered to pH 7.4 with 10 mM Hepes, containing antagonists when appropriate. The plate was incubated for 10 min at room temperature and then placed in the chamber of a Lab Systems Luminoskan plate reader. HBSS (20 ml) with or without agonist was injected into each well and the luminescence signal was integrated for 20 s after the injection. Cells on coverslips were removed from the medium, washed by dipping in a beaker of HBSS, and then placed in a holder and into a cuvette containing 0.5 ml HBSS at room temperature. The cuvette was placed in the chamber of a Bio-Orbit 1250 luminometer. Drugs were applied by injecting 250 ml of appropriate solution in HBSS into the cuvette. Luminescence was followed for up to 1 min.
/ 6m14$$9507
03-27-96 14:32:40
abal
FIG. 1. U373MG cells were loaded with aequorin as described under Materials and Methods, plated onto coverslips, and cultured overnight before assay (aequorin-loaded). An identical sample of cells was treated in the same way, but not exposed to aequorin (control cells). Cuvettes containing cells on coverslips were placed in the BioOrbit luminometer. Luminescence was recorded and Triton X-100 dissolved in either HBSS or Ca-free HBSS containing 6 mM EGTA was injected after 3 s. Data are means and SE from four runs. RLU is the abbreviation for relative light units.
RESULTS
Validation of the Loading Procedure Aequorin-loaded cells on coverslips gave a luminescence signal when treated with 0.3% Triton X-100 in HBSS. The detergent breaks open the cell and exposes the aequorin to extracellular calcium. There was no signal from control cells (not treated with aequorin), or from aequorin-loaded cells when the Triton was dissolved in Ca-free HBSS containing EGTA (Fig. 1). These results confirm that the luminescence signal comes from aequorin and is dependent on calcium. Aequorin-loaded cells on coverslips also responded to SP and ionomycin, but not to buffer injections. Control cells not exposed to aequorin gave no signal to SP or ionomycin (Fig. 2). Figure 3 shows a time course of responses to ionomycin and SP after loading with aequorin. These cells were loaded, placed in a microtiter plate, and assayed at different times using the Luminoskan luminometer. Immediately after loading there was a relatively large response to ionomycin and no response to SP. After 30 min, the response to ionomycin was reduced and the cells had started to respond to SP. After 60 min the response to ionomycin had stabilized at around four times lower than the initial response. The response to SP had reached a maximum. In this experiment, responses to both ionomycin and SP were followed up to 18 h after loading, and there was no decay. The pattern for luminescence induced with Triton X-100 was the same as that for ionomycin, but the signals were about 10-fold higher (see Fig. 5A). The responses
AP-Anal Bio
272
HEDLEY, PHAGOO, AND JAMES
FIG. 2. Responses of aequorin-loaded U373MG cells to ionomycin and SP. U373MG cells were loaded with aequorin, placed on coverslips, left in culture overnight, and assayed as described under Materials and Methods. Control cells were treated in the same way as loaded cells, but there was no aequorin in the loading buffer. Data are means and SE from four runs. (A) Response of aequorin-loaded and control cells to 10 mM ionomycin. (B) Response of aequorin-loaded cells to 100 nM SP or buffer.
to ionomycin, SP, and Triton X-100 could still be measured 48 h after loading. In some experiments there was no loss of signal; in others there was a slight loss. Overall responses to SP after 48 h were 95 { 35% (mean { SE, N Å 3) of the response after 18–22 h. The signal obtained from aequorin-loaded cells depended on the concentration of aequorin used in the loading procedure and on the density of cells. Good signals were obtained at between 20,000 and 60,000 cells/well.
FIG. 4. Log concentration–response curves for agonists in the aequorin assay. U373MG cells were loaded with aequorin and assayed as described under Materials and Methods. Agonist responses were normalized to the maximum aequorin signal, obtained by breaking open cells with 0.25% Triton X-100. (A) Curves for SP (filled squares), substance P-methyl ester (open triangles), and [Sar9, Met(O2)11]substance P (filled circles). (B) Curve for histamine. Note the different X axes.
Agonists Three NK1 agonists all produced concentration-related responses in the aequorin assay (Fig. 4A). There was little or no response to high concentrations of [bAla8]neurokinin A or [Phe(Me)7]neurokinin B, selective NK2 and NK3 receptor agonists, respectively (data not shown). Histamine also caused an increase in luminescence (Fig. 4B). The curves shown in Fig. 4 imply that histamine gave a bigger maximal response than the NK1 agonists. This was not always the case. In other experiments, where histamine and SP were tested on the same cells, we have seen smaller responses to histamine than SP. Estimates of EC50 values for the agonists are reported in Table 1. For SP, our value is higher than that obtained by measuring calcium changes with Fura-2 (20). For histamine, our estimate agrees well with the EC50 value reported for histamine-induced release of interleukin-1b in U373MG cells (21).
TABLE 1
Agonist Potencies in the Aequorin-Based Assay with U373MG Cells Agonist
FIG. 3. Time course of responses after loading U373MG cells with aequorin. Cells were loaded with aequorin and pipetted into microtiter plates and responses were measured as described under Materials and Methods. The concentration of ionomycin was 3 mM and SP was 100 nM. Data are means and SE from 12 replicates.
/ 6m14$$9507
03-27-96 14:32:40
abal
Substance P Substance P–methyl ester [Sar9, Met(O2)11]substance P Histamine
EC50 (nM) 31 37 96 7300
{ { { {
4 (10) 10 (2) 10 (3) 1400 (3)
Note. Data are means { SE from the number of independent experiments given in parentheses.
AP-Anal Bio
MEASURING NEUROKININ RESPONSES IN AEQUORIN-LOADED CELLS TABLE 2
Potencies of NK1 Antagonists in the Aequorin-Based Assay Compound CP 99994 RPR 100893 SR 140333 Spantide II
IC50 (nM) 3.0 14 1.2 5.0
{ { { {
0.49 (6) 2.7 (5) 0.35 (5) 2.0 (5)
Note. IC50 is the concentration of antagonist that gave 50% inhibition of the response to 30 nM SP. Data are means { SE from the number of independent experiments given in parentheses.
Neurokinin Antagonists The effects of an EC50 concentration of SP (30 nM) in aequorin-loaded U373MG cells were blocked by the NK1 antagonists CP99994 (22), RPR100893 (23), SR 140 333 (24), and Spantide II (25). IC50 values are reported in Table 2. All four NK1 antagonists were able to block the SP response completely with IC50 values of less than 15 nM. Loading Other Cell Lines We have investigated the use of the gravity loading procedure with other cell lines and receptor types. Our results indicate that U373MG cells were a fortunate choice for our first experiments. All the cell lines we tested took up the aequorin. There then appeared to be two patterns. In some cells (CHO and AR42J) aequorin activity declined rapidly to very low levels within 15 min; in others (U373MG, WI-38, and IMR-90) the decline was much slower (Fig. 5A). We tested agonist responses in cells that retained significant activity. For U373MG cells the agonist was SP; for WI-38 and IMR90 cells we used bradykinin to activate the B2 receptors on these cells. Only the U373MG cells gave consistent responses to agonist (Fig. 5B).
273
another agonist and suggest that the technique is applicable to other receptor systems. We loaded cells with the aequorin protein as a complementary approach to transfecting cells with aequorin cDNA. The advantage of using aequorin rather than calcium-sensitive fluorescent dyes is that the assay can be run conveniently on microtiter plates using luminometers that allow injection of reagent and immediate measurement of the signal. This protocol is essential because the calcium concentration is elevated for a relatively short time. At present, commercially available fluorimeters have no facility for reagent injection. The major disadvantage is that aequorin is less sensitive to calcium than the dyes. Practically, this means that it is difficult to measure background levels of calcium and the signal does not increase until the calcium concentration reaches probably several hundred nanomolar. The rather high EC50 value of about 30 nM that we report here for substance P, compared to those of 1–2 nM found with Fura2 measurements (20), may reflect the difference in assay sensitivity. In spite of these differences, our estimates of antagonist potencies agree well with published values (22, 25, 28). Transfection of apoaequorin into cells is likely to be a very useful technique both for routine screening assays and for investigating calcium changes in subcellular organelles (as discussed in the introduction) and is probably the method of choice in many circumstances. The loading technique may have some advantages, however. First, it is a relatively simple procedure that requires no skill in the techniques of molecular biology.
DISCUSSION
Activation of NK1 receptors in U373MG cells causes a variety of responses which include increases in levels of phophoinositides and intracellular calcium (20, 26– 28), translocation of protein kinase C from the cytoplasm to the membrane (26), release of taurine (26, 28), activation of c-fos and c-jun synthesis (27, 28), activation of an outward current in voltage-clamped cells (28), and release of interleukin-6 (21, 29, 30). Histamine also activates the phospholipase C system, probably through H1 receptors and causes release of interleukin-6 in these cells (21, 31). We have used U373MG cells to develop a convenient relatively high throughput assay for activation of NK1 receptors based on measurements of intracellular calcium with aequorin. The data with histamine simply show that we can activate the same pathways with
/ 6m14$$9507
03-27-96 14:32:40
abal
FIG. 5. Aequorin loading in other cell lines. Cells were loaded with aequorin as described under Materials and Methods. (A) Total aequorin activity in each cell line was estimated by breaking cells open with 0.25% Triton X-100 and exposing the aequorin to the extracellular calcium. (B) Responses to agonists; 100 nM SP for U373MG cells, 100 nM bradykinin for WI-38 and IMR-90 cells. Data are means of 12 determinations. Cell lines were U373MG (filled squares), WI-38 (filled circles), IMR-90 (open triangles), CHO (closed triangles), and AR42J (open circles).
AP-Anal Bio
274
HEDLEY, PHAGOO, AND JAMES
Second, there may well be advantages in loading other reagents along with the aequorin. For example, it may be possible to load cells efficiently with antisense oligonucleotides with the gravity loading technique. If the antisense is loaded along with the aequorin and is taken up by the same cells, then most of the signal should come from cells that have received antisense. This may allow the use of relatively low concentrations of oligonucleotides and at the same time ensure that there is no (or only a very low) background from untreated cells. The same could apply to other reagents that act inside cells but are not taken up efficiently, e.g., peptides. Application of the gravity loading procedure in these kinds of experiments may be limited. We have successfully used gravity loading only with U373MG cells. Of the cell lines we tested, two (CHO and AR42J) lost the aequorin very quickly. Two cell lines which retained the aequorin (WI-38 or IMR-90 cells) gave very little or no response to agonist. We do not know the reason for the differences between cell lines. Perhaps bradykinin receptors on WI-38 and IMR-90 cells were damaged during the loading procedure and did not recover over the period of the experiment. It is also possible that in some cells aequorin is sequestered in intracellular organelles and hence protected from changes in cytoplasmic calcium concentration. In summary, we have shown that U373MG cells can be loaded with aequorin by a simple method, and that the luminescence signal in loaded cells depends on calcium. Aequorin-loaded U373MG cells can be used to measure SP-stimulated increases in intracellular calcium. The SP effect is mediated by NK1 and not NK2 or NK3 receptors, since (a) NK1 antagonists inhibit the SP signal and (b) there is no or very little response to NK2- or NK3-selective agonists. This system provides a relatively rapid and convenient assay for NK1 antagonists. The loading method is not applicable to all cell lines. REFERENCES 1. Cobbold, P. H., and Rink, T. J. (1987) Biochem. J. 248, 313–328. 2. Sheu, Y-A., Kricka, L. J., and Pritchett, D. B. (1993) Anal. Biochem. 209, 343–347. 3. Button, D., and Brownstein, M. (1993) Cell Calcium 14, 663– 671. 4. Brini, M., Maesault, R., Bastianutto, C., Alvarez, J., Pozzan, T., and Rizutto, R. (1995) J. Biol. Chem. 270, 9896–9903. 5. Rutter, G. A., Theler, J-M., Murgia, M., Wollheim, C. B., Pozzan, T., and Rizutto, R. (1993) J. Biol. Chem. 268, 22385–22390.
/ 6m14$$9507
03-27-96 14:32:40
abal
6. Rizzuto, R., Simpson, A. W. M., Brini, M., and Pozzan, T. (1992) Nature 358, 325–327. 7. Kendall, J. M., Dormer, R. L., and Campbell, A. K. (1992) Biochem. Biophys. Res. Commun. 189, 1008–1016. 8. Badminton, M. N., Kendall, J. M., Sala-Newby, G., and Campbell, A. K. (1995) Exp. Cell Res. 216, 236–243. 9. McNeil, P. L., and Taylor, D. L. (1985) Cell Calcium 6, 83–93. 10. Borle, A. B., Freundrich, C. C., and Snowdowne, K. W. (1986) Am. J. Physiol. 251, C323–C326. 11. Hallet, M. B., and Campbell, A. K. (1980) Biochem. J. 192, 585– 596. 12. Morgan, J. P., DeFeo, T., and Morgan, K. G. (1984) Pflu¨gers Arch. 400, 338–340. 13. Johnson, P. C., Ware, J. A., Clivenden, P. B., Smith, M., Dvorak, A. M., and Salzman, E. W. (1985) J. Biol. Chem. 260, 2069– 2076. 14. Yamaguchi, A., Suzuki, H., Tanoue, K., and Yamazaki, H. (1986) Thromb. Res. 44, 165–174. 15. Borle, A. B., and Snowdowne, K. W. (1982) Science 217, 252– 254. 16. Dormer, R. L. (1983) Biosci. Rep. 3, 233–240. 17. Malmgren, R., Grunfelt, S., and Joseph, R. (1992) Thromb. Haemostasis 68, 352–356. 18. Campbell, A. K., and Hallet, M. B. (1983) J. Physiol. (London) 338, 537–550. 19. Hallet, M. B., and Campbell, A. K. (1983) Immunology 50, 487– 495. 20. Bordley, A., Feltz, P., and Trouslard (1994) Glia 11, 277–283. 21. Cadman, E. D., Witte, D. G., and Lee, C. M. (1994) J. Neurochem. 63, 980–987. 22. McLean, S., Ganong, A., Seymour, P. A., Snider, R. M., Desai, M. C., Rosen, T., Bryce, D. K., Longo, K. P., Reynolds, L. S., Robinson, G., and Scmidt, A. W. (1993) J. Pharmacol. Exp. Ther. 267, 472–479. 23. Tabart, M., and Peyronel, J. F. (1994) Bioorg. Med. Chem. Lett. 4, 673–676. 24. Emonds-Alt, X., Doutremepuich, J. D., Heaulme, M., Neliat, G., Santucci, V., Steinberg, R., Vilain, P., Bichon, D., Ducoux, J. P., and Proietto, V. (1993) Eur. J. Pharmacol. 250, 403–413. 25. Hakanson, R., Leander, S., Asano, N., Feng, D.-M., and Folkers, K. (1990) Regul. Peptides 31, 75–82. 26. Lee, C. M., Tung, W. L., and Young, J. D. (1992) J. Neurochem. 59, 406–414. 27. Eistetter, H. R., Mills, A., Brewster, R., Alouani, S., Rambosson, C., and Kawashima, E. (1992) Glia 6, 89–95. 28. Oury-Donat, F., Lefevre, I. A., Thurneyssen, O., Gauthier, T., Bordey, A., Feltz, P., Emonds-Alt, X., Le Fur, G., and Soubrie, P. (1994) J. Neurochem. 62, 1399–1407. 29. Palma, C., Goso, C., and Manzini, S. (1994) Neurosci. Lett. 171, 221–224. 30. Palma, C., Urbani, F., and Manzini, S. (1995) J. Neuroimmunol. 59, 155–163. 31. Arias-Montano, J. A., Berger, V., and Young, J. M. (1994) Br. J. Pharmacol. 111, 598–608.
AP-Anal Bio