Cytolytic peptides induce biphasic permeability changes in mammalian cell membranes

Ž .Journal of Immunological Methods 252 2001 63–71www.elsevier.nlrlocaterjim

Cytolytic peptides induce biphasic permeability changes inmammalian cell membranes

Mei Su, Chufa He, Charles A. West, Steven J. Mentzer)

Laboratory of Immunophysiology, Dana-Farber Cancer Institute, Room G09, JFB, 44 Binney Street, Boston, MA 02115, USAHarÕard Surgical Research Laboratories, HarÕard Medical School, Boston, MA 02115, USA

Received 19 September 2000; received in revised form 13 December 2000; accepted 15 December 2000

Abstract

The cytolytic peptides melittin and gramicidin S are naturally occurring agents that provide a comparative model forstudies of complement, immunotoxin and cell-mediated membrane permeability. Most attempts to characterize cytolyticpeptides have used model membrane systems including phospholipid vesicles or erythrocytes. Membrane vesicles permit theuse of self-quenching concentrations of fluorescent permeability markers, while erythrocytes release measurable hemoglobin.Attempts at measuring early membrane permeability changes in nucleated mammalian cells have been limited. To measurethe kinetics of mammalian cell membrane permeability changes induced by cytolytic peptides, we developed a 96-wellfluorescence cytolysis assay using the cytoplasmic fluorescent dye calcein as the membrane permeability marker. To

Ž .facilitate rapid assessment of membrane permeability, trypan blue was added to the assay solution to quench a releasedŽ .fluorescence and b retained intracellular fluorescence. Trypan blue also provided a complementary visual assessment of

cell viability. Using this assay, a detailed kinetic analysis demonstrated permeability of the cell membranes within secondsof exposure to the cytolytic peptides. The rapid permeabilization of the cell membranes was confirmed by flow cytometryusing the calcium indicator dye fluo-3. The assay also demonstrated a second slower phase of marker release over the nextseveral hours. The fluorescence cytolysis assay was able to reliably detect the biphasic permeability changes associated withthe melittin and gramicidin S peptides suggesting the potential utility of this assay in the assessment of other cytolyticagents. q 2001 Elsevier Science B.V. All rights reserved.

Keywords: Cytolysis; Melittin; Gramicidin; Vital dyes; Calcein; Trypan blue

AbbreÕiations: 7AAD, 7-amino-actinomycin D; DME, Dul-becco’s Modified Eagle’s Medium; PBS, phosphate bufferedsaline; RFU, relative fluorescence units

) Corresponding author. Tel.: q1-617-732-6703; fax: q1-617-730-2898.

Ž .E-mail address: [email protected] S.J. Mentzer .

1. Introduction

Cytolytic peptides are naturally occurring smallmolecules that interact with cell membranes and leadto cell death. In most cases, the peptides are thoughtto contribute to the cytolytic process through thedevelopment of new ion channels in the cell mem-

Ž .brane Vogel and Jahnig, 1986; Laine et al., 1988 ,Žby influencing pre-existing channels Kourie and

.Shorthouse, 2000 , or by creating a disorganizedŽregion of the membrane Benachir and Lafleur,

0022-1759r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.Ž .PII: S0022-1759 01 00334-9

( )M. Su et al.rJournal of Immunological Methods 252 2001 63–7164

.1995 . Gramicidin is cytolytic peptide produced byŽthe bacterium Bacillus breÕis reviewed in Kourie

.and Shorthouse, 2000 . Gramicidin is a dimer of two15 residue peptides that form a b-helix channel inthe cell membrane. Melittin is the polypeptide com-ponent of the venom of the European honey bee

ŽApis mellifera reviewed in Dempsey, 1990; Merri-.field et al., 1994 . Melittin forms an ion channel by

assuming an extended a-helical conformation in thecell membrane. Both melittin and gramicidin havebeen used extensively in studies of transmembraneconductance and pore formation.

Cytolytic peptides, such as gramicidin and melit-tin, provide a useful comparative model for theinvestigation of complement-, immunotoxin- andcell-mediated membrane toxicity. Previous work withmelittin and gramicidin S has focused on modelmembrane systems and erythrocytes. Studies of theaction of these peptides on mammalian cell mem-branes have been limited by the difficulty in measur-ing the rapid effects of these compounds. Flowcytometry has been adapted to the investigation ofthe effects of melittin on a lymphoblastoid cell lineŽ .Weston et al., 1994; Weston and Raison, 1998 .These studies have shown evidence for rapid cell

Ž .death Weston et al., 1994 , as well as membraneŽdisorganization Weston et al., 1994; Weston and

.Raison, 1998 . Although flow cytometry providesuseful early kinetic data, it does not provide theopportunity for multiple parallel comparisons or pro-longed kinetic studies.

In this report, we describe a fluorescence mem-brane permeability assay that uses the fluorescentmarker calcein. Trypan blue was used to quenchreleased extracellular and retained intracellular fluo-rescence in nonviable cells. This assay permitted thekinetic analysis of multiple parallel conditions of thecytolytic peptides, melittin and gramicidin. The as-say demonstrated that the addition of the cytolyticpeptides resulted in a rapid initial permeabilizationof the cells followed by a slower secondary phase.

2. Methods

2.1. Cells and culture medium

ŽThe human lymphoblastoid cell line JY Mentzer.et al., 1985 and the sheep heterohybridoma HT2r4

Ž .Su et al., 2000 cells were cultured in Dulbecco’sŽ .Modified Eagle’s Medium DME with 2000 mgrl

Ž .glucose Sigma, St. Louis, MO , supplemented withŽ .3% heat inactivated fetal calf serum Sigma , 10 mM

Hepes buffer and 2 mM L-glutamine. Human periph-eral blood mononuclear cells were isolated on a

ŽFicollrHypaque gradient Lymphocyte Separation.Medium; Bionetics, Kensington, MD as described

Ž .Mentzer et al., 1987 . The prescapular lymph nodefrom sheep was used for all efferent lymph duct

Ž .cannulations Grau, 1933 . The efferent lymph ductwas cannulated with a heparin-bonded polyurethane

Žcatheter Solo-Cath, CBAS-C35; Setters Life Sci-.ences, San Antonio, TX . The lymph was collected

3 Žin 50 cm sterile centrifuge tubes Falcon, Franklin. 3 ŽLakes, NJ or 250 cm sterile plastic bags Abbott,

.North Chicago, IL . Each collection bag containedŽ200 IU of heparin, 2000 IU of penicillin Cellgro,

.Mediatech, Herndon, VA , and 2000 mg of strepto-Ž .mycin Cellgro .

2.2. Cytolytic peptides

The cytolytic peptides used in this study wereŽ .gramicidin S G-5127, Sigma-Aldrich, St. Louis

Žfrom B. breÕis and melittin M-7391, Sigma-Al-.drich . The lyophyllized peptides were resuspended

Ž .in phosphate buffered saline PBS just prior to use.

2.3. Trypan blue

ŽTrypan blue solution C H N NaO S ; MW34 24 6 14 4.891 was obtained in a 0.4% solution in 0.81% NaCl

Ž .and 0.06% KPO Sigma . The trypan blue was used4Ž .at a final concentration of 0.04% vrv or as indi-

cated.

2.4. Calcein fluorescence loading

The acetoxymethyl ester of the pH stable dyecalcein was used as the cytoplasmic fluorescence

Ž .tracer Molecular Probes, Eugene, OR . Calcein is afluorescein derivative with excitation at 488 nm and

Ž .emmission at 514 nm Kayalar and Duzgunes, 1986 .The calcein was added for a final concentration of 10mM to a cell concentration of 2–5=106 cellsrml.The cells were incubated in 378C water bath for 30min with frequent gentle mixing. The cells were

( )M. Su et al.rJournal of Immunological Methods 252 2001 63–71 65

washed twice in serum containing medium prior tothe assay.

2.5. Fluorescence measurements

The cellular fluorescence was measured from thebottom of the microtiter plate using the CytoFluor

Ž4000 Fluorescence Measurement System Applied.Biosystems, Foster City, CA . The CytoFluor 4000

is a computer-controlled multi-well fluorescencescanning device with a Tungsten halogen lamp andbroadband interference filters: ex 485"20 nm andem 530"25 nm. The kinetic data was serially col-lected at room temperature within the CytoFluorchamber at preset intervals to minimize photo-bleaching. The fluorescence intensity measures were

Ž .exported to the Microsoft Excel Redmond, WAspreadsheet for data analysis.

2.6. Microtiter plate assay

The cells were washed twice in DME and 3%Ž .fetal bovine serum Sigma and labeled with calcein.

After labeling, the cells were washed twice in DMEand 3% fetal bovine serum and counted. The cellswere plated at 0.4–1=105 cellsrwell in 100 ml ofwash medium in a 96-well U-bottom microtiter plateŽ .Falcon . The microtiter plate was centrifuged at 500rpm for 2 min to pellet the cells and scanned as abaseline to insure pipetting accuracy. Trypan blueŽ . Ž .0.16% vrv containing medium 50 ml was addedto each well for a final concentration of 0.04%. Anadditional 50 ml of medium was added to the mini-mum release wells, and 50 ml of 2% Triton X-100Ž .Sigma was added to the maximum release wells for

Ža total volume of 200 ml. The cytolytic peptides 50.ml were added to the experimental wells just prior

Ž .to the scanning period total volume: 200 ml .

2.7. 7AAD staining and flow cytometry

The 7AAD, obtained from Calbiochem–Ž .Novabiochem San Diego, CA as a powder, was

dissolved in absolute methanol, then diluted in PBSto a final concentration of 200 mgrml. The aliquotswere stored at y708C in the dark until use. The

Ž . 67AAD solution 100 ml was added to 10 cells

suspended in 1 ml PBS and mixed thoroughly. Thecells were stained for 20 min at 48C in the dark,pelleted by centrifugation, and resuspended in 500

Ž .ml of 2% paraformaldehyde Sigma . Positive con-trols for apoptosis included treatment of the cells

Ž .with various doses of camptothecin Sigma , a topoi-somerase-I inhibitor, for 4 h. Unstained cells wereused as a negative control. Experimental and control

Žsamples were analyzed using an Epics XL Beckman.Coulter, Miami, FL within 30 min of fixation. Data

on 50,000 cells were acquired and processed usingŽ .WinList 4.0 software Verity, Topsham, ME . Scat-

tergrams were generated by combining forward lightscatter with 7AAD fluorescence.

2.8. Fluo-3 staining and flow cytometry

Ž .The fluo-3 Molecular Probes, Eugene, OR wereprepared as 10 mgrml DMSO stock solutions andaliquoted. The staining method was similar to that

Ž .previously described Novak and Rabinovitch, 1994 .Briefly, the cells were suspended in DME medium atpH 7.4 and containing 5% fetal bovine serumŽ .Sigma . After resuspending the cells in Hanks bal-anced salt solution, the cells were incubated for 30min at 378C in media containing the fluo-3 acetoxy-methyl esters at 2 mgrml. The cells were diluted 1:5

Ž .with 1% fetal bovine serum Sigma and incubatedanother 40 min at 378C, followed by centrifugationand resuspension in the calcium calibration bufferand analyzed at 378C. Flow cytometry gating wasbased on forward and right-angle light scatter. Thecells were analyzed for a variable baseline period at378C. There was a brief break in analysis during theaddition of the cytolytic peptides followed by furtheranalysis at 378C. The data was collected in sequen-tial data file and analyzed using the WinList 4.0

Ž .software Verity .

3. Results

Calcein is a polyanionic fluorescein derivativethat was used in these studies because it demon-strates cellular retention greater than 4 h, does notappear to impair cellular function, and is pH insensi-tive. After cell loading, calcein fluorescence is di-


rectly dependent on cell volume and cell numberŽ .Fig. 1 . Rapid changes in membrane permeabilityare difficult to measure using any fluorescent dye

Ž .because of: 1 difficulty discriminating intracellularŽ .and extracellular fluorescence and 2 residual intra-

cellular fluorescence after membrane permeabiliza-tion. We screened a variety of quenching agents fortheir ability to quench both released and retained

Žcalcein. For nonviable cells, trypan blue 0.04%.vrv provided near total quenching of both released

and retained fluorescence. Further, for nonpermeabi-lized viable cells, this concentration of trypan blue

Žresulted in an acceptable loss average decrease of.16"7% relative fluorescence units of measureable

Ž .fluorescence Fig. 1 . An additional benefit of trypanblue was the visual assessment of cell viability.

The cytolytic peptides melittin and gramicidin Swere used to induce membrane permeability inhuman and sheep cell lines, as well as normal lym-phocytes. The measured permeability of the calcein-labeled cells was markedly enhanced with the addi-

Ž .tion of the trypan blue Fig. 2A . A substantialcomponent of this increase in sensitivity was sup-pression of the cell-free fluorescence released into

Ž .the supernatant Fig. 2B . Parallel comparisons ofthe different cell types demonstrated a reproduciblesusceptibility to cytoltytic peptides: sheep heterohy-

Fig. 1. Microtiter plate fluorescence measurements of calcein-loaded human lymphocytes at various cell concentrations in the

Ž . Ž .presence round or absence diamond of 0.4% trypan blue.Fluorescence was measured from the bottom of U-bottom wells in

Ž . Ž .relative fluorescence units RFU error barss1 SD . Backgroundfluorescence with assay buffer alone was subtracted from the

Ž 2 .triplicate values linear regression: R )0.99 for both conditions .

Fig. 2. Measurement of gramicidin-induced JY cell permeabilityŽ . Ž .in the presence square or absence diamond of 0.4% trypan

Ž . Ž 4 .blue. A Calcein-labeled JY cells 6=10 were treated withvarious concentrations of gramicidin for 30 min at room tempera-

Ž .ture. B In separate experiments, JY cells treated for 30 min withvarying concentrations of gramicidin were pelleted and the super-natants assayed for cell-free fluorescence. Fluorescence in bothŽ . Ž .A and B was measured from the bottom of U-bottom wells in

Ž .relative fluorescence units RFU . Each data point representsŽ .triplicate samples error barss1 SD . Background fluorescence

with assay buffer alone was subtracted from the triplicate values.

Ž .bridoma HT2r4 )human lymphblastoid cell lineŽ .JY )sheep lymphocytes)human lymphocytes.

Although the absolute magnitude of cytolysis var-ied with cell type, a similar time course was obtainedwith human and sheep lymphocytes and cell lines. Inall cell types tested, there was an immediate loss offluorescence upon exposure to the cytolytic peptidesfollowed by a slower phase of fluorescence decreaseŽ .Figs. 3 and 4 . The loss of fluorescence, except forthe highest concentration of melittin, followed expo-

Ž .nential decay kinetics Fig. 3 . In contrast, the slowerphase of gramicidin-associated permeability demon-

Ž .strated linear kinetics over 7–10 h Fig. 4 . Thissecondary phase paralleled the spontaneous loss of


Fig. 3. Dose–response of the melittin-induced sheep lymphocyte membrane permeabilization. The retained intracellular fluorescence of 105

calcein-labeled cells per microtiter well was measured at baseline followed by the addition of various concentrations of the cytolytic peptideŽ . Ž .melittin arrow . The fluorescence was measured in relative fluorescence units RFU with the mean of triplicate samples shown. After the

Ž 2 . Ž 2 .initial decline in fluorescence, exponential regressions were calculated for 35 mM R s0.7177 , 17.5 mM R s0.8542 , 8.75 mMŽ 2 . Ž 2 .R s0.9527 and 4.37 mM R s0.9941 . Background fluorescence with assay buffer alone was subtracted from the triplicate values.

Žfluorescence in the no peptide control conditions not.shown .

To provide a complementary assessment of therapid phase of cytolytic peptide-induced membrane

Fig. 4. Dose–response of gramicidin-induced sheep lymphocyte membrane permeabilization. The retained intracellular fluorescence of 105

calcein-labeled cells per microtiter well was measured at baseline followed by the addition of various concentrations of the cytolytic peptideŽ . Ž .gramicidin arrow . The fluorescence was measured in relative fluorescence units RFU with the mean of triplicate samples shown. After

Ž 2 . Ž 2 .the initial decline in fluorescence, linear regressions were calculated for 160 mM R s0.982 , 80 mM R s0.9987 , 40 mMŽ 2 . Ž 2 .R s0.9997 and 20 mM R s0.9987 . Background fluorescence with assay buffer alone was subtracted from the triplicate values.


Fig. 5. Fluo-3 fluorescence increase associated with the additionŽ . Ž .of melittin A and gramicidin S B to the lymphoblastoid cell

line JY at 378C. A 30–45 s baseline recording period wasobtained followed by addition and mixing of the cytolytic peptidesŽ . Ž . Ž . Ž . Ž .arrow . A Melittin 35 mM and B gramicidin S 160 mM .The data was recorded in a sequential listmode file and processed

Ž .by WinList 4.0 Verity .

permeability, we examined the time course of thecytolytic peptides using flow cytometry. The cyto-plasmic calcium indicator dye fluo-3 was used as thepermeability marker and the cytolytic peptides wereadded during a kinetic analysis. The peptide-inducedpermeability of fluo-3-labeled cells resulted in asharp rise in fluo-3 fluorescence intensity due to theinflux of calcium. For both cytolytic peptides, therewas a rapid increase in fluorescence followed by a

Ž .plateau phase Fig. 5 . Despite using the cell line JY,there was sufficient variability in the absolute fluo-3intensity that a reproducible dose–response relation-ship using flow cytometry could not be established.

The different slow-phase kinetics of melittin andgramicidin S suggested mechanistic differences thatmight distinguish melittin- and gramicidin S-treatedcells. To assess the live, dead and apoptotic fractionsof cells after treatment with cytolytic peptides, JYcells were treated with various concentrations ofcytolytic peptide for 30 min, then labeled with thefluorescent DNA-binding agent 7AAD. Flow cytom-etry of the cells treated with melittin or gramicidin Sshowed no significant difference in the fractions oflive and dead cells. Few apoptotic cells were ob-

Ž .served Fig. 6 . The comparable effects of the cy-tolytic peptides observed by flow cytometry suggeststhat the linear decline in gramicidin-associated fluo-rescence detected by the fluorescence plate assaymay reflect retained subcellular fluorescence thatwas not detectable by flow cytometry.

Ž . Ž . Ž .Fig. 6. Flow cytometry of 7AAD staining of melittin A–D and gramicidin S E–H treated JY cells showing live arrow 1 , deadrnecroticŽ . Ž . Ž . Ž .arrow 2 and apoptotic arrow 3 cells 30 min after the addition of the cytolytic peptides. The melittin concentrations were A 4.4 mM, B

Ž . Ž . Ž . Ž . Ž . Ž .8.75 mM, C 17.5 mM and D 35 mM. The gramicidin S concentrations were E 20 mM, F 40 mM, G 80 mM and H 160 mM.


4. Discussion

Comparative studies of complement-, immuno-toxin- and cell-mediated membrane toxicity havegenerally focused on membrane permeability as ameasure of cell death. In microtiter plate assays, celldeath has been typically defined as the release of a

Ž 51 .permeability marker e.g., Cr or the inability toŽ .exclude a vital dye e.g., trypan blue . Because of

Žtheir relatively slow release Martz and Benacerraf,.1973 , common membrane permeability markers

have not been used to assess the time course ofrapidly cytolytic agents such as melittin and grami-cidin. Similarly, vital dyes are too cumbersome forthe rapid measurement of membrane permeability.

The problems associated with the development ofmore sensitive membrane permeability markers arerelated to the structural complexity of mammaliancells. First, mammalian cells have multiple intra-cellular compartments that can selectively retain cy-toplasmic markers and prevent their release after cellmembrane permeabilization. Second, the marker mustbe retained in cytoplasm without disruptive chemicalmodification. Cytoplasmic modification may be use-

Žful in retaining the marker in the viable cell e.g.,. Ž .acetoxymethyl esters Provinciali et al., 1992 or it

may lead to inappropriate retention of the marker inŽ 51 .the permeabilized cell e.g., Cr radiolabeling

Ž .Henney, 1974 . Third, mammalian cells must becultured in serum and other osmotically activemedium. The diffusion of a permeability marker intothe cell can be a limiting step in the rapid detection

Žof membrane permeabilization Burakoff et al.,.1975 .

To overcome these limitations and facilitate theanalysis of cytolytic peptide-induced membrane toxi-city, we developed a 96-well microtiter plate fluores-cence cytolysis assay that detected rapid changes inmembrane permeability. Released fluorescent dyeand retained intracellular fluorescence from nonvi-able cells was quenched using trypan blue. Rapidquenching obviated the need for elaborate separationprocedures or ongoing manipulation. As a conse-quence, prolonged kinetic studies with multiple par-allel conditions were possible. Using this assay, weobserved a dramatic increase in membrane perme-ability after exposure of the cells to the cytolyticpeptides.

The secondary phase of melittin and gramicidincytotoxicity was a relatively minor component of thecellular toxicity; nonetheless, differences betweenmelittin and gramicidin S were observed. The melit-tin demonstrated exponential decline of the retainedfluorescence consistent with diffusion-dependentquenching of intracellular fluorescence. In contrast,gramicidin S demonstrated a linear decline in cellu-lar fluorescence, which paralleled the spontaneousrelease of calcein in untreated cells. These differ-ences were further investigated by flow cytometry.Flow cytometry using the 7AAD did not demonstrateany difference in live, dead or apoptotic fractions ofthe cells treated with the cytolytic peptides. To-gether, these data suggest that the gramicidin S wascomparably lethal to the treated cells, but gramicidinS treatment may have resulted in retained fluores-cence in a subcellular fraction. This possibility sug-gests a potential limitation in this assay, as mem-brane-enclosed subcellular fractions are inaccessibleto quenching by the added trypan blue. Potentialbiochemical explanations for the observed differencebetween gramicidin S and melittin remain undefined.

Most studies of membrane permeability have usedradionuclide markers. Radionuclide markers are easyto use, are released within 30–60 min of membranepermeability, and are readily measured with com-monly available emission detectors. Despite theircommon usage, it is unclear how radionuclide mark-ers, such as 51Cr are retained in the cytoplasm andwhat molecules they are bound to when they are

Žreleased from the cell Martz and Benacerraf, 1973;.Martz, 1977 . An alternative to radionuclides is fluo-

Žrescence membrane permeability markers Edidin,1970; Horan and Kappler, 1977; Persidsky and Bail-lie, 1977; Bruning et al., 1980; Kolber et al., 1988;

.Wierda et al., 1989 Fluorescence membrane perme-ability markers are generally nontoxic, economicaland provide a hundred-fold increase in sensitivity.Most radionuclide and fluorescence membrane per-meability markers are spontaneously released overtime limiting the length of the assay to less than8–12 h.

Most membrane permeability markers are difficultto use for kinetic analyses. Both radionuclide andfluorescence markers generally require the separation

Žof the released and bound marker Bruning et al.,.1980; Wierda et al., 1989 . In most cases, the assay


must be stopped and a separation procedure usedprior to detection of the released marker. Althoughmultiple conditions are possible, this approach istime-consuming and requires multiple parallel mi-crotiter plates. An alternative to multiple microtiterplates is the use of fluorescence quenching. Quench-ing agents permit the suppression of fluorescencefrom the released dye and prevent the ongoing mea-surement of retained intracellular dye in nonviablecells.

To identify a nontoxic agent capable of quenchingboth intracellular and extracellular fluorescence, wesought a nontoxic agent with an appropriate absorp-tion spectrum. We screened a variety of dyes with anabsorption spectrum appropriate for fluoresceinemissions. The most effective quenching agent was

Žthe aniline dye trypan blue Evans and Schulemann,.1914 . Trypan blue is a common, inexpensive and

nontoxic dye. When trypan blue penetrates the per-meabilized cell membrane, it provides a distinctiveblue appearance to dead cells under light microscopyŽ .Black and Berenbaum, 1964; Tennant, 1964 . With

Žan absorption peak at 595 nm Harrisson et al.,.1981 , trypan effectively quenches the released ex-

Ž .tracellular fluorescence Loike and Silverstein, 1983 .Because trypan blue rapidly diffuses into permeabi-lized cells, it quenches the retained intracellular fluo-rescence as well.

Using a fluorescence plate reader that detectsfluorescence from the bottom of the plate, trypanblue effectively limited the background fluorescencefrom both the released and inappropriately retainedfluorescent dye. The disadvantage of trypan blue wasa modest quenching effect on the intracellular dyesin viable cells. In this study, we observed a 16%reduction in the fluorescence of labeled lymphocytes.

ŽAlso, the emission peak of trypan blue at 670 Har-.risson et al., 1981 , preventing its use with red

fluorescent dyes.

References

Benachir, T., Lafleur, M., 1995. Study of vesicle leakage inducedby melittin. Biochim. Biophys. Acta 1235, 452–460.

Black, L., Berenbaum, M.C., 1964. Factors affecting the dyeexclusion test for cell viability. Exp. Cell Res. 35, 9–13.

Bruning, J.W., Kardol, M.J., Arentzen, R., 1980. Carboxyfluores-cein fluorochromasia assays: I. Non-radioactively labeled cellmediated lympholysis. J. Immunol. Methods 33, 33–44.

Burakoff, S.J., Martz, E., Benacerraf, B., 1975. Is the primarycomplement lesion insufficient for lysis? Failure of cells dam-aged under osmotic protection to lyse in EDTA or at lowtemperature after removal of osmotic protection. Clin. Im-munol. Immunopathol. 4, 108–126.

Dempsey, C.E., 1990. The actions of melittin on membranes.Biochim. Biophys. Acta 1031, 143–161.

Edidin, M., 1970. A rapid, quantitative fluorescence assay for celldamage by cytotoxic antibodies. J. Immunol. 104, 1303–1306.

Evans, H.M., Schulemann, W., 1914. The action of vital stainsbelonging to the benzidine group. Science 39, 443.

ŽGrau, H., 1933. Die lymphgefasse der haut des schafes OÕis.aries . Ztschr. Anat. Entw.-Gesch. 101, 423–448.

Harrisson, F., Callebaut, M., Vakaet, L., 1981. Microspectro-graphic analysis of trypan blue-induced fluorescence in oocytesof the Japanese quail. Histochemistry 72, 563–578.

Henney, C.S., 1974. Estimation of the size of a T-cell-inducedlytic lesion. Nature 249, 456–458.

Horan, P.K., Kappler, J.W., 1977. Automated fluorescent analysisfor cytotoxicity assays. J. Immunol. Methods 18, 309–316.

Kayalar, C., Duzgunes, N., 1986. Membrane action of colicin E1:detection by the release of carboxyfluorescein and calceinfrom liposomes. Biochim. Biophys. Acta 860, 51–56.

Kolber, M.A., Quinones, R.R., Gress, R.E., Henkart, P.A., 1988.Measurement of cytotoxicity by target cell release and reten-tion of the fluorescent dye bis-carboxyethyl-carboxyfluo-

Ž .rescein BCECF . J. Immunol. Methods 108, 255–264.Kourie, J.I., Shorthouse, A.A., 2000. Properties of cytotoxic pep-

tide-formed ion channels. Am. J. Physiol.: Cell Physiol. 278,C1063–C1087.

Laine, R.O., Morgan, B.P., Esser, A.F., 1988. Comparison be-tween complement and melittin hemolysis: anti-melittin anti-bodies inhibit complement lysis. Biochemistry 27, 5308–5314.

Loike, J.D., Silverstein, S.C., 1983. A fluorescence quenchingtechnique using trypan blue to differentiate between attachedand ingested glutaraldehyde-fixed red blood cells in phagocy-tosing murine macrophages. J. Immunol. Methods 57, 373–379.

Martz, E., 1977. Mechanism of specific tumor-cell lysis by al-loimmune T lymphocytes: resolution and characterization ofdiscrete steps in the cellular interaction. Contemp. Top. Im-munobiol. 7, 301–361.

Martz, E., Benacerraf, B., 1973. An effector-cell independent stepin target cell lysis by sensitized mouse lymphocytes. J. Im-munol. 111, 1538–1545.

Mentzer, S.J., Gromkowski, S.H., Krensky, A.M., Burakoff, S.J.,Martz, E., 1985. LFA-1 membrane molecule in the regulationof homotypic adhesions of human B lymphocytes. J. Immunol.135, 9–11.

Mentzer, S.J., Smith, B.R., Barbosa, J.A., Crimmins, M.A., Her-rmann, S.H., Burakoff, S.J., 1987. CTL adhesion and antigenrecognition are discrete steps in the human CTL–target cellinteraction. J. Immunol. 138, 1325–1330.

Merrifield, R.B., Merrifield, E.L., Juvvadi, P., Andreu, D., Bo-


man, H.G., 1994. Design and synthesis of antimicrobial pep-tides. Ciba Found. Symp. 186, 5–20.

Novak, E.J., Rabinovitch, P.S., 1994. Improved sensitivity in flowcytometric intracellular ionized calcium measurement usingfluo-3rFura Red fluorescence ratios. Cytometry 17, 135–141.

Persidsky, M.D., Baillie, G.S., 1977. Fluorometric test of cellmembrane integrity. Cryobiology 14, 322–331.

Provinciali, M., Di Stefano, G., Fabris, N., 1992. Optimization ofcytotoxic assay by target cell retention of the fluorescent dye

Ž .carboxyfluorescein diacetate CFDA and comparison withconventional 51CR release assay. J. Immunol. Methods 155,19–24.

Su, M., He, C., West, C.A., Mentzer, S.J., 2000. Generation ofŽ .sheep= sheep=mouse heterohybridoma cell line expressing

the beta 1 integrin membrane molecule. Hybridoma 19, 81–88.

Tennant, J.R., 1964. Evaluation of the trypan blue technique fordetermination of cell viability. Transplantation 2, 685–694.

Vogel, H., Jahnig, F., 1986. The structure of melittin in mem-branes. Biophys. J. 50, 573–582.

Weston, K.M., Raison, R.L., 1998. Interaction of melittin with ahuman lymphoblastoid cell line, HMy2. J. Cell. Biochem. 68,164–173.

Weston, K.M., Alsalami, M., Raison, R.L., 1994. Cell membranechanges induced by the cytolytic peptide, melittin, are de-tectable by 90 degrees laser scatter. Cytometry 15, 141–147.

Wierda, W.G., Mehr, D.S., Kim, Y.B., 1989. Comparison offluorochrome-labeled and 51Cr-labeled targets for natural killercytotoxicity assay. J. Immunol. Methods 122, 15–24.

Cytolytic peptides induce biphasic permeability changes in mammalian cell membranes

Documents

Transcript of Cytolytic peptides induce biphasic permeability changes in mammalian cell membranes