Accepted Manuscript Interaction analysis of a ladder-shaped polycyclic ether and model transmembrane peptides in lipid bilayers by using Förster resonance energy transfer and polarized attenuated total reflection infrared spectroscopy Kazuya Yamada, Haruki Kuriyama, Toshiaki Hara, Michio Murata, Raku Irie, Yanit Harntaweesup, Masayuki Satake, Seketsu Fukuzawa, Kazuo Tachibana PII: DOI: Reference:
S0968-0896(14)00309-5 http://dx.doi.org/10.1016/j.bmc.2014.04.044 BMC 11539
To appear in:
Bioorganic & Medicinal Chemistry
Received Date: Revised Date: Accepted Date:
25 February 2014 21 April 2014 22 April 2014
Please cite this article as: Yamada, K., Kuriyama, H., Hara, T., Murata, M., Irie, R., Harntaweesup, Y., Satake, M., Fukuzawa, S., Tachibana, K., Interaction analysis of a ladder-shaped polycyclic ether and model transmembrane peptides in lipid bilayers by using Förster resonance energy transfer and polarized attenuated total reflection infrared spectroscopy, Bioorganic & Medicinal Chemistry (2014), doi: http://dx.doi.org/10.1016/j.bmc.2014.04.044
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Title Interaction analysis of a ladder-shaped polycyclic ether and model transmembrane peptides in lipid bilayers by using Förster resonance energy transfer and polarized attenuated total reflection infrared spectroscopy
Authors Kazuya Yamadaa, Haruki Kuriyamaa, Toshiaki Harab,c, Michio Muratab,c, Raku Iriea, Yanit Harntaweesupa, Masayuki Satakea, Seketsu Fukuzawaa∗, Kazuo Tachibanaa,b ∗
Affiliations a
Department of Chemistry, School of Science, The University of Tokyo,
Hongo, Bunkyo-ku, Tokyo 113-0033, Japan b
JST ERATO Lipid Active Structure, 1-1 Machikaneyama, Toyonaka, Osaka 560-0043,
Japan c
Department of Chemistry, Graduate School of Science, Osaka University,
1-1 Machikaneyama, Toyonaka, Osaka 560-0043, Japan
*Tel/Fax +81-3-5841-4366 e-mail:
[email protected],
[email protected] 1
Abstract Ladder-shaped polycyclic ethers (LSPs) are predicted to interact with membrane proteins; however, the underlying mechanism has not been satisfactorily elucidated.
It has been hypothesized that LSPs possess non-specific affinity to
α-helical segments of transmembrane proteins.
To verify this hypothesis, we
constructed a model LSP interaction system in a lipid bilayer. α-helical peptides and reconstituted them in liposomes.
We prepared 5 types of The reconstitution and
orientation of these peptides in the liposomes were examined using polarized attenuated total reflection infrared (ATR-IR) spectroscopy and gel filtration.
The results revealed
that 4 peptides were retained in liposomes, and 3 of them formed stable transmembrane structures. The interaction between the LSP and the peptides was investigated using Förster resonance energy transfer (FRET).
In the lipid bilayer, the LSP strongly
recognized the peptides that possessed aligned hydrogen donating groups with leucine caps.
We propose that this leucine-capped 16-amino acid sequence is a potential LPS
binding motif.
2
1. Introduction∗ Many secondary metabolites produced by marine organisms have unique molecular structures and possess remarkable bioactivities.
Ladder-shaped
polycyclic ethers (LSPs) among them, such as the brevetoxins1,2 and ciguatoxins3,4 produced by marine dinoflagellates, exhibit potent toxicity.
These toxins are a
known cause of fish or shellfish poisoning, and therefore, the molecular mechanism of toxicity onset needs to be clarified for further physiological studies and thus to prevent poisoning.
A proposed target of these LSP toxins is channel proteins.
Brevetoxin B, a red tide toxin, binds to voltage-sensitive sodium channel (VSSC),5 preventing their inactivation.
Similarly, ciguatoxin, the causative toxin in ciguatera
fish poisoning, binds to VSSC and inhibits their inactivation.
The dissociation
constants (KD) of these toxin to VSSC are in the nanomolar/sub-nanomolar range.6,7 In addition to brevetoxins and ciguatoxins that bind to VSSCs, gambierol and its truncated analogues bind to voltage-gated potassium channels (VGPCs),8-10 and maitotoxin causes calcium ion influx through an unidentified membrane-bound protein, but not on liposomes.11
Interestingly, brevetoxins behave as maitotoxin
antagonists,12 and their binding to VSSC is inhibited by brevenal, a shorter LSP.13 Recently, an artificial ladder-shaped heptacyclic polyether was found to inhibit maitotoxin-induced calcium ion influx.14
These works suggest that the primary
target molecules of LSPs are membrane proteins including ion channels.
∗
However,
Abbreviations: ATR-IR, attenuated total reflection infrared; CD, circular dichroism; DMEQ-TAD,
4-[2-(3,4-Dihydro-6,7-dimethoxy-4-methyl-3-oxo-2-quinoxalinyl)ethyl]-3H-1,2,4-triazole-3,5(4H)-dione; FRET, Förster resonance energy transfer; GpA, glycophorin A; LSP, ladder-shaped polycyclic ether; TFE, 2,2,2-trifluoroethanol; YTX, yessotoxin.
3
with the exception of yessotoxin (YTX), 15,16 their precise mode of interaction at the molecular level is yet to be satisfactorily elucidated. According to surface plasmon resonance (SPR) and saturation transfer difference NMR measurements, YTX interacts with a transmembrane domain of human glycophorin A (GpA).17,18 Hydrogen bonds between α-protons of glycine residues, forming a glycine zipper,19 and the vicinal ether oxygen atoms of YTX are thought to be the key mediators of this interaction.
On the basis that nearly identical distance between one pitch of an
α-helix (5 Å) and the interval of the vicinal ether oxygen atoms at one side of LSP molecules (~5 Å), it is hypothesized that sequential hydrogen bonds exist between the vicinal ether oxygen atoms of LSPs and the protons of amino acid residues in the α-helical domains (Figure 1).20 These interaction analyses thus far described were performed in a solution phase with21 or without18,22 micelles.
Therefore,
investigation in a lipid bilayer system was desired to simulate real events in the biological systems.
Here, we used YTX as a model LSP to construct an interaction
model between YTX and the α-helical transmembrane peptides in liposomes to further elucidate the general recognition mechanism mediating LSP-membrane protein interactions.
Although the rotational-echo double-resonance (REDOR)
solid-phase NMR technique is one of the most powerful tools to observe intermolecular or distant intermolecular interaction in lipid bilayers,23 stable isotope labelling of LSPs and peptides is laborious. and ATR-IR techniques.
For our analysis, we opted for FRET
The model interaction this study here reported establishes
a method that can be used to elucidate the interaction mechanism of LSP and ion channel and other membrane proteins.
4
2. Results and discussion 2.1. Design and preparation of peptides Initially, we predicted that hydrogen bonding and hydrophobicity are two of the most essential driving forces of the LSP-membrane protein interaction.
On the
basis of this speculation, we designed 5 α-helical peptides, as shown in Table 1. The leucine-lysine (peptide 1) and leucine-serine (peptide 2) repeat helical peptides are laterally amphiphilic, and their hydrophilic side chains with hydrogen donors (cyan in Table 1) aligned on one side of the helix, are predicted to interact with ether oxygen atoms of LSPs.
Because the amino groups of the lysine repeats in peptide
1 make the α-helix too hydrophilic to be reconstituted in the lipid bilayer, peptide 1 was designed to have 2 helical segments.
Similar to ion channels, this tandem
helical structure was expected to segregate the hydrophilic side inside of the dimmer and thus to stabilize the α-helices in the lipid bilayer.24
In contrast, peptide 2, more
hydrophobic and free of charged residues, was expected to be embedded in the lipid bilayer more comfortably.
An additional 3 lysine residues were introduced at the
both termini of the peptide in order hopefully to lock the helical segment between two sides of bilayer. hydrogen donors.
Peptide 3 is a leucine-alanine repeat peptide, which has no This hydrophobic peptide is known to form a monomeric
transmembrane structure.25 attached to both termini.
As with peptide 2, lysine and arginine residues were
The rest borrowed the sequence from a natural membrane
proteins, namely the human GpA transmembrane domain (GpA-TM, peptide 4), is known to interact with YTX in solution,18 and as previously described, its recognition site is predicted to be the α-protons of the glycine residues within its sequence.
Furthermore 3 glycine residues were replaced with serine in peptide 4,
5
with the expectation that it would have more efficient hydrogen bonding to ether oxygen atoms of YTX. 3G/3S.
The resultant peptide, peptide 5, was named GpA-TM
GpA-TM 2G/2I (peptide 6), which was subjected to SPR experiments in
previous work,18 was not used in this study. solid-phase peptide synthesis (SPPS).
All peptides were prepared using
In addition, native chemical ligation was
used to connect the linker part at the middle of the sequence in the synthesis of peptide 1.26
2.2. Reconstitution and secondary structural analysis The secondary structures of these peptides were analysed using circular dichroism (CD) spectra attained in 2,2,2-trifluoroethanol (TFE) (Figure S1).
All
spectra showed local minimums at 207 and 222 nm, indicating typical α-helical profiles.
Therefore, the peptides all met the requirement at least for α-helix models.
Next, we examined the reconstitution of the peptides in liposomes using egg-yolk phosphatidyl choline. of the peptide.
The reconstitution method used depended on the behaviour
A detergent removal method using dialysis27 was suitable for the
reconstitution of the hydrophobic peptides 3, 4, and 5.
We found that
3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulphonate
(CHAPS)
and
n-octyl-β-D-glucopyranoside (OG) yielded the highest α-helix population in detergent screens for peptide 3, and peptides 4 and 5, respectively.
In contrast, the
thin-layer hydration method28 was suitable for the reconstitution of the amphiphilic peptides 1 and 2, because they were lost during a detergent removal step in the dialysis procedure.
6
2.3. Peptide behaviours in liposomes To determine whether the α-helical peptides stably form transmembrane structures in the liposomes, we conducted polarized ATR-IR measurements.29-31 Using these measurements, we evaluate the orientation of the peptide α-helix and the phospholipid acyl chain axis relative to the membrane surface, by comparing the absorbance at 0 and 90 degree polarization. measurements are assembled in Table 2.
The results of the polarized ATR-IR
The dichroic absorbance ratios (R values,
Table 2) of amide I (1670-1640 cm-1) in peptides 3, 4, and 5 were 3–4, which suggested that their orientation angle is approximately 30° in transmembrane structures.
In the case of peptides 1 and 2, the R values were approximately 2,
corresponding to the magic angle of 54°. whether
these
peptides
formed
54°
Therefore, we could not distinguish transmembrane
structures
or
were
non-orientated. To acquire more detailed information regarding peptides 1 and 2 in liposomes, bound/free (B/F) separation was performed using gel filtration.
When a peptide
bonds stably to the membrane, the liposomes and the peptide should elute simultaneously.
In contrast, in an aqueous medium, the peptide should elute after
the liposomes.
First, the peptides were labelled with a fluorescent tag to improve
their detection limit (Scheme S1).
Peptide 3, which was shown to form
transmembrane structures in ATR-IR experiments,25 was used as a positive control. In the first gel filtration, the peptides reconstituted in liposomes and free peptides were eluted in fractions 4 and 13, respectively (Figure 2).
To examine the stability
of the membrane association of the peptides, fractions 4–6 were combined and a second gel filtration was performed.
In the case of peptide 1, the free peptide was
7
eluted, once again, in fraction 13.
However, the reconstituted peptides 2 and 3
were eluted again in fraction 4, suggesting that they were largely retained in the liposomes.
These results imply that peptide 2 was stably incorporated in the lipid
bilayer and its conformation in the lipid bilayer was not as rigid as that of peptides 3, 4, and 5.
2.4. Interaction analysis The FRET strategy we employed for our interaction analysis of the model transmembrane peptides and YTX is presented in Figure 3.32
In these experiments,
we labelled the peptides (Scheme S2) and YTX (Scheme S3) with fluorescein and 6,7-dimethoxy-1-methyl-2(H)-quinoxalinone
(DMEQ),
respectively.33
The
excitation/emission wavelengths of DMEQ and fluorescein are 370 nm/440 nm and 493 nm/515 nm, respectively.
Once YTX and the peptides interact, FRET from
DMEQ to fluorescein occurs.
According to FRET theory, distance information
ranging from 10–100 Å can be obtained.32
Because the typical bilayer thickness is
approximately 30 Å, we cannot determine the orientation (parallel or antiparallel) of both chromophores by using this technique. FRET experiments were conducted in both TFE solution and in liposome suspensions.
The results are assembled in Figure 4.
FRET efficiency was defined
as the difference in the fluorescence intensity of the fluorescein-labelled peptides in the presence versus absence of DMEQ-dsYTX.
In TFE solution, the FRET
efficiencies of peptides 1, 2, and 4 were considered moderate, whereas peptide 5 showed the highest FRET efficiency of all the peptides.
Peptide 3, which lacks a
hydrogen donating group, had a low FRET efficiency in TFE solution.
8
In contrast,
the human GpA-derived peptides 4 and 5 had high FRET efficiency in liposomes. Although the simple repeat-containing peptides (peptides 1 and 2) exhibited moderate FRET efficiency in TFE solution, the lysine-rich double-helical peptide (peptide 1) showed low FRET efficiency in liposomes. The experimental results pertaining to the relationship between the peptide sequence and FRET efficiency are summarized as follows.
In the case of the
artificially repeated sequences (peptides 1, 2, and 3), the leucine-alanine repeats assist in the formation of α-helical structures, and the aligned residues (Figure 5a; X: K, S, or L/A, cyan) control the affinity of the peptides towards LSP. YTX-peptide interactions in TFE solution have hydrophilicity dependence.
The
side chain amino and hydroxyl groups enhance the affinity of peptides to YTX (peptides 1 and 2; Figures 6a and 6b), whereas the absence of these functional groups reduces this affinity (peptide 3; Figure 6c).
These data implied the
presence of complementary hydrogen bonds between YTX and the peptide. However, the affinity tendency of peptides towards YTX differed in liposomes.
In
liposomes, the hydrophobicity or membrane association ability dictated the affinity of peptides towards YTX.
Peptide 1 is too hydrophilic to stay stably in the lipid
bilayer, as we described previously (Figure 6a).
Peptide 2 is considered
moderately hydrophobic and possesses several hydrogen donating groups (Figure 6b).
The transmembrane structure of peptide 2 is not as stable as that of peptide 3,
as shown in the polarized ATR-IR results.
Both peptide 3 and YTX stay in the
lipid bilayer, where they inadvertently come into close proximity (Figure 6c). This may account for the near equal interaction ability of peptides 2 and 3 towards YTX in liposomes.
Thus, moderate hydrophobicity, in combination with aligned
9
hydrogen donating groups, might be important for YTX-peptide interactions. In contrast, the natural membrane protein-derived sequences peptides 4 and 5 showed higher affinity to YTX than did peptides 1, 2, and 3.
Particularly in
liposomes, the enhanced FRET efficiency is likely due to the stable transmembrane structure of the peptides.
In comparison with other peptides in the helical wheel,
the leucine residues (L4 and L19; Figure 5b, magenta) in peptides 4 and 5 sandwiched the key aligned hydrogen donating groups (Figure 5b, cyan).
In the
primary sequences (Table 1), there are leucine-isoleucine triplet repeats (L4I5I6 and I17L18L19) at both ends of the core sequence.
Both the hydrophobic caps and the
aligned hydrogen donating groups in peptides 4 and 5 can stabilize their intermolecular interaction with YTX (Figures 6d and 6e).
In peptide 2, S4A5L6
and L17A18S19 correspond to the leucine-isoleucine triplet repeats.
The serine
residues at position 4 and 19 might destabilize the YTX-peptide interaction.
This
might account for the observed lower FRET efficiency of peptide 2 than that of peptide 5 in TFE solution (Figures 6b and 6e). Generally, the hydrogen bonding ability of glycine α-protons is lesser than that of hydroxyl groups.
The FRET
efficiency of peptide 4 is almost equal to that of peptide 2 in TFE solution. Presumably, both of the leucine caps assist YTX recognition through the glycine α-protons in peptide 4 (Figure 6d).
In peptide 5, the hydrogen bonding ability of
the serine hydroxyl groups and the hydrophobic leucine caps might co-operatively contribute to the YTX-peptide interaction (Figure 6e).
This might explain why
peptide 5 exhibited the highest affinity towards YTX. This combined driving force also enhanced the FRET efficiency in liposomes. Therefore, peptides 4 and 5, which have stable transmembrane structures and
10
aligned hydrogen donating groups with leucine caps, exhibit high affinity towards LSP.
In addition to glycine and serine residues, threonine, asparagine, and
glutamine residues can replace in the sequence.
Thus far, there is only one report
that describes the intermolecular interaction between an LSP and a protein through hydrogen bonds.34,35 Using X-ray crystal structural analysis, this study showed an intermolecular hydrogen bond between the oxygen atom of ciguatoxin fragments and the asparagine side chain amide group of the ciguatoxin-specific antibody 10C9. However, we have an alternative interpretation of LSP-peptide interactions. According to the kinetic analysis performed using the SPR technique, the GpA-TM mutant GpA-TM 2G/2I (peptide 6, Table 1) showed higher affinity towards YTX than the wild-type version (peptide 4).18 The characteristics of the peptide were changed by replacing glycine residues with isoleucine.
These data imply that high
hydrophobicity, such as that introduced by a leucine zipper motif, is an important factor for LSP recognition,36 and it suggests diversity might exist in the mechanism by which recognition occurs.
3. Conclusion We investigated the mechanism underlying the interaction between YTX and the α-helical peptides in the lipid bilayers using FRET and polarized ATR-IR experiments.
The aligned hydrogen donating groups with the leucine caps are
predicted to be one of the possible LSP recognition motifs.
Further peptide
screening will enable us to identify a recognition sequence for other LSPs. Isotope-labelled peptides in combination with REDOR experiments, is a promising approach to elucidate the mechanism of ion channel function.
11
4. Experimental Section 4.1. Peptide synthesis Peptides were synthesized using a Model 433A synthesizer (Applied Biosystems, Foster City, CA, USA) or by performing manual solid-phase synthesis at the 0.1 mmol scale.
Deprotection and cleavage from the resin was performed
using trifluoroacetic acid/water/1,2-ethanedithiol (9.0/0.5/0.5, v/v) or trifluoroacetic acid/water/1,2-ethanedithiol/thioanisole/phenol
(8.25/0.5/0.5/0.25/0.5,
Peptide 1 was obtained using native chemical ligation.26
v/v).
The N-terminal thioester
and the C-terminal cysteine capped peptide was dissolved in phosphate buffer (pH 7), and the ligation proceeded spontaneously. The resulting crude peptides were purified by reversed-phase HPLC (JASCO PU980 HPLC System, Tokyo, Japan) using the COSMOSIL 5C18-AR II column (20 ϕ × 250 mm, Nacalai Tesque, Kyoto, Japan) for peptides 1, 2, and 3, and the COSMOSIL 5C4-AR 300 column (20 ϕ × 250 mm; Nacalai Tesque) for peptides 4 and 5.
Peptides were eluted using a liner gradient of aqueous methanol (peptides 1,
2, and 3) or aqueous acetonitrile/isopropanol (peptides 1, 2, and 3) containing 0.1% TFA.
The purified peptides were validated by MALDI-MS (AXIMA-CFR,
Shimadzu, Kyoto, Japan) (Figures S2–S6).
4.2. Preparation of DMEQ-dsYTX YTX was isolated from the cultured extracts of the dinoflagellate Protoceratium reticulatum as previously reported.37 12
The cells were harvested from
67 L of the cultured medium and were extracted with methanol.
The resultant
extracts were washed with ethyl acetate, followed by extraction with n-butanol. The butanolic extracts were chromatographed over alumina and passed through a Sep-Pak C18 cartridge (Waters, Milford, MA, USA).
The YTX-containing
fractions were validated using MALDI-MS and purified using ODS-HPLC, generating 7.803 mg of YTX. The desulphation of YTX was performed by adding 8 mg of p-toluenesulphonic acid monohydrate to a solution of YTX (4 mg) in 3 mL of dioxane at room temperature.
After stirring for 11 h, a saturated solution of
sodium bicarbonate was added to quench the reaction.
The mixture was extracted
with 3 mL of ethyl acetate 3 times, washed with brine, and the combined organic layer was dried over sodium sulfate. temperature using a vacuum.
The solvent was removed at room
This mixture was used in subsequent reactions
without further purification.38
dsYTX was labelled by adding 1 mg of
4-[2-(3,4-dihydro-6,7-dimethoxy-4-methyl-3-oxo-2-quinoxalinyl)ethyl]-3H-1,2,4-tri azole-3,5(4H)-dione (DMEQ-TAD) to a stirred solution of dsYTX in 2 mL of dichloromethane at room temperature, in the dark. was added to quench the reaction. using a vacuum.
After stirring for 1 h, methanol
The solvent was removed at room temperature
Lastly, the reaction mixture was purified by reversed-phase
HPLC using a COSMOSIL 5C18-AR II column.33,39
4.3. Preparation of fluorescently labelled peptides Peptides 1 and 3 were fluorescently labelled, and gel filtration was performed by
dissolving
equal
amounts
13
(0.1
mg)
of
N-(7-dimethylamino-4-methyl-3-coumarinyl)-maleimide and peptide in DMSO. The reaction mixture was vortexed for 3 h in the dark and then lyophilized.
The
crude reaction products were purified by reversed-phase HPLC using a COSMOSIL 5C18-AR II column. (Figures S7–S8).
The purified products were validated by MALDI-MS
For the FRET experiments, peptides 1–5 were fluorescently
labelled by adding 0.5 mg of fluorescein maleimide to DMSO/TFE (1/1, v/v) peptide-containing (1.0 mg) solutions. in the dark and then lyophilized.
The reaction mixture was vortexed for 5 h
The crude reaction products were purified by
reversed-phase HPLC using a COSMOSIL 5C18-AR II column for peptides 1–3, and using a COSMOSIL 5C4-AR 300 column for peptides 4 and 5.
The purified
products were validated by MALDI-MS (Figures S9–S13).
4.4. CD measurements CD measurements were performed on a JASCO J-725 CD spectrometer with a JASCO Model 121-1 cylindrical quartz cell (20 ϕ × 1 mm).
Each peptide was
dissolved in TFE to a concentration of 0.2 mM, and its spectrum from 190–260 nm was obtained by carrying out 10 scans.
4.5. Peptide Reconstitution Peptides 1 and 2 were reconstituted using the thin-layer hydration method. Equal volumes (1 mL) of the peptide in TFE (0.5 mg/mL for peptide 1 and 0.25 mg/mL for peptide 2) and egg yolk phosphatidycholine (2.25 mg/mL) were combined and evaporated in a 50 mL flask, and dried under high vacuum overnight. 14
To the peptide-lipid film, 600 µL of 0.1 M phosphate buffer (pH 7.0) was added. The suspension was vortexed for 10 min. Peptides 3, 4, and 5 were reconstituted using the detergent removal method. The peptides in TFE (2.5 mg/mL, 200 µL), were each combined with egg yolk phosphatidylcholine (2.25 mg) and 1 mg of detergent (CHAPS for peptide 3, and OG for peptides 4 and 5).
Next, the resultant mixture was concentrated in a 50 mL
flask using a rotary evaporator and dried under high vacuum overnight to generate the peptide-lipid-detergent film.
The film was suspended in 300 µL of 0.1 M
phosphate buffer (pH 7.0), vortexed for 10 min, and dialyzed using a semipermeable membrane (MWCO, 5 kDa) for 3 days.
4.6. Polarized ATR-IR measurements Samples for the polarized ATR-IR measurement were prepared by exchanging the buffer of the reconstituted peptide liposomes for D2O (150 mL) by dialyzing through a semipermeable membrane (MWCO, 3 kDa) for 1 day.
The resultant
solution was loaded onto a germanium optical plate, and dried by a gentle stream of nitrogen to generate a lipid film.
The film thickness (>8 µm), estimated from the
amount of liquid applied amount, was much greater than the penetration depth of the polarized light (0.2–0.8 µm) in the range of 3000–800 cm-1.
ATR-IR spectra were
obtained using a JASCO FT/IR-6100 spectrometer equipped with an ATR-500/M attachment, HgCdTe detector, and KRS-5 polarizer.
The total reflection number
was 5 on the film side.
The spectra were measured at a resolution of 4 cm-1 and
45° angle of incidence.
The peak areas of the symmetric CH2 stretching band
15
(2843–2863 cm-1) and amide I peak (1645–1665 cm-1) were calculated from the spectra when the angle of a polarization plate was 0° and 90° against the optical plate.
Subsequently, the absorbance ratios (R) of 0° to 90° polarization were
calculated using equation 1.40
S m and St are the order parameters of the molecular axis around the z-axis and the transition moment around the molecular axis, respectively. amplitude.
E is electric field
These order parameters are calculated using the following equations:
θ is molecular axis and α is transition moment, which is set to 35° or 39.5° in the case of amide I, and 90° in the case of the symmetric CH2 stretching band, respectively.28 absorption.
The R value is calculated first from the peak areas of the
Sm and θ are calculated from equations 1 and 2.
4.7. Gel filtration B/F separation of the reconstituted peptides in the liposomes was carried out using gel filtration.
Before separation, the column (sepharose 4B, 10 ϕ × 550 mm,
GE Healthcare Life Sciences, Piscataway, NJ, USA) was washed with 0.5 M
16
sodium hydroxide for 3 h, and equilibrated with 0.1 M phosphate buffer (pH 7.0) for 6 h.
The liposome suspensions (500 µL) were loaded into the column and eluted
using phosphate buffer.
The elution was performed at a flow rate of 0.5 mL/min
and a fraction was collected every 5 min. used.
A detection wavelength of 208 nm was
A 250 µL aliquot of each fraction was subjected to UV absorbance and
fluorescence measurements using a Varioskan Flash microplate reader (Thermo Fisher Scientific, Waltham, MA, USA).
The absorbance wavelength for the
detection of the liposomes was 400 nm, and the excitation/emission wavelengths for fluorescence detection were 400 nm/480 nm (peptide 1 and 3) and 494 nm/521 nm (peptide 2), respectively.
The lipid quantification was performed by the enzymatic
method,41 using a phospholipid C-test (Wako, Tokyo, Japan).
To an aliquot of
each fraction (250 µL), 50 µL of the phospholipid C-test enzyme solution was added and the reaction mixture was incubated at 37°C for 5 min.
The lipid quantification
was calculated by comparing the absorbance at 600 nm.
4.8. FRET measurements Samples were prepared for FRET experiments as follows.
YTX was
desulphated under acidic conditions to enhance its affinity towards the lipid bilayer, and the resultant desulphated YTX (dsYTX) was reacted with DMEQ-TAD to generate DMEQ-dsYTX (Scheme S2).33
Fluorescein-labelled peptides were
produced by the nucleophilic addition of fluorescein maleimide to the thiol group of a cysteine residue (Scheme S3).
The peptides reconstituted in the liposomes with
and without YTX were prepared as described previously.
17
In the thin-layer
hydration method, the peptide (0.1 mg) and DMEQ-dsYTX (0.05 mg) were dissolved in TFE (200 µL) in a 50 mL flask, and egg yolk phosphatidylcholine (0.6 mg) was added.
The solvent was removed using a rotary evaporator and the
resultant mixture was dried under high vacuum overnight to produce the lipid film. The lipid film, suspended in 300 µL of 0.1 M phosphate buffer (pH 7.0), was vortexed for 10 min to produce liposomes.
In the detergent removal method, to the
TFE solution (200 µL) containing the peptide (0.1 mg) and DMEQ-dsYTX (0.05 mg), egg yolk phosphatidylcholine (0.6 mg), and CHAPS or OG (0.2 mg) were added. The resultant mixture was concentrated using a rotary evaporator followed by drying under high vacuum overnight.
The lipid film, suspended in 300 µL of
0.1 M phosphate buffer (pH 7.0), was vortexed for 10 min.
Subsequently, the
buffer was dialyzed using a semipermeable membrane (MWCO, 5 kDa) for 3 days to produce the peptide reconstituted liposomes. An aliquot of each liposome sample (250 µL) was subjected to fluorescence measurements using a Varioskan Flash microplate reader.
The excitation
wavelength used was 360 nm, and the detection was scanned from 380–600 nm. The self-fluorescence intensity of each fluorescein-labelled peptide was measured (excitation wavelength, 493 nm; emission wavelength, 505–599 nm).
The relative
FRET efficiency was calculated as the fluorescence intensity (515 nm) excited at 360 nm, divided by the intensity of fluorescence itself (515 nm), excited at 493 nm.
Acknowledgements We are very grateful to our laboratory colleagues Dr. Ai Yoshinaka-Niitsu and
18
Dr. Kohtaro Sugahara for their helpful discussions and technical advice regarding isolation and purification of YTX, respectively.
Thanks are also due to Jun
Yamaguchi, Mayuri Tashiro, and Yuhei Takenishi of our laboratory for the initial studies, and Prof. Masayuki Inoue of the University of Tokyo for help in solid-phase peptide synthesis.
Polarized ATR-IR analyses were performed with
the generous help of Mr. Mitstuo Ohama and Mr. Kazushi Kawamura at Osaka University.
Part of this work was supported by a Grant-in Aid for Scientific
Research (A16201044) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
References 1. Lin, Y.-Y.; Risk, M.; Ray, S. M.; Engen, D. V.; Clardy, J.; Golik, J.; James, J. C.; Nakanishi, K. J. Am. Chem. Soc. 1981, 103, 6773. 2. Shimizu, Y.; Chou, H.-N.; Bando, H.; van Duyne, G.; Clardy, J. C. J. Am. Chem. Soc. 1986, 108, 514. 3. Murata, M.; Legrand, A.-M.; Ishibashi, Y.; Fukui, M.; Yasumoto. T. J. Am. Chem. Soc. 1990, 112, 4380. 4. Satake, M.; Morohashi, A.; Oguri, H.; Oishi, T.; Hirama, M.; Harada, N.; Yasumoto, T. J. Am. Chem. Soc. 1997, 119, 11325. 5. Poli, M. A.; Mende, T. J.; Baden, D. G. Mol. Pharmacol. 1986, 30, 129. 6. Trainer, V. L.; Thomsen, W. J.; Catterall, W. A.; Baden, D. G. Mol. Pharmacol. 1991, 40, 988. 7. Gawley, R. E.; Rein, K. S.; Kinoshita, M.; Baden, D. G. Toxicon 1992, 30, 780.
19
8. Cuypers, E.; Abdel-Mottaleb, Y.; Kopljar, I.; Rainier, J. D.; Raes, A. L.; Snyders, D. J.; Tytgat, J. Toxicon 2008, 51, 974. 9. Alonso, E.; Fuwa, H.; Vale, C.; Suga, Y.; Goto, T.; Konno, Y.; Sasaki, M.; LaFerla, F. M.; Vieytes, M. R.; Giménez-Llort, L.; Botana, L. M. J. Am. Chem. Soc. 2012, 134, 7467. 10. Kopljar, I.; Labro, A. J.; de Block, T.; Rainier, J. D.; Tytgat, J.; Snyders D. J. J. Gen. Physiol. 2013, 141, 359. 11. Konoki. K.; Hashimoto, M.; Murata, M.; Tachibana, K. Chem. Res. Toxicol. 1999, 12, 993. 12. Konoki. K.; Hashimoto, M.; Nonomura, T.; Sasaki, M.; Murata, M.; Tachibana, K. J. Neurochem. 1998, 70, 409. 13. Bourdelais, A. J.; Campbell, S.; Jacocks, H.; Naar, J.; Wright, J. L. C.; Carsi, J.; Baden D. G. Cell. Mol. Neurobiol. 2004, 24, 553. 14. Oishi, T.; Konoki, K.; Tamate, R.; Torikai, K.; Hasegawa, F.; Matsumori, N.; Murata, M. Bioorg. Med. Chem. Lett. 2012, 22, 3619. 15. Murata, M.; Kumagai, M.; Lee, J. S.; Yasumoto, T. Tetrahedron Lett. 1987, 28, 5869. 16. Takahashi, H.; Kusumi, T.; Kan, Y.; Satake, M.; Yasumoto, T. Tetrahedron Lett. 1996, 37, 7087. 17. Mori, M.; Oishi, T.; Matsuoka, S.; Ujihara, S.; Matsumori, N.; Murata, M.; Satake, M.; Oshima, Y.; Matsushita, N.; Aimoto, S. Bioorg. Med. Chem. 2005, 13, 5099. 18. Ujihara, S.; Oishi, T.; Torikai, K.; Konoki, K.; Matsumori, N.; Murata, M.; Oshima, Y.; Aimoto, S. Bioorg. Med. Chem. Lett. 2008, 18, 6115.
20
19. Acharya, R.; Gupta, M.; Ramakumar, S.; Ramagopal, U. A.; Chauhan, V. S. BMC Struct. Biol. 2007, 7, 51. 20. Oguri, H.; Oomura, A.; Tanabe, S.; Hirama, M. Tetrahedron Lett. 2005, 46, 2179. 21. Torikai, K.; Oishi, T.; Ujihara, S.; Matsumori, N.; Konoki, K.; Murata, M.; Aimoto, S. J. Am. Chem. Soc. 2008, 130, 10217. 22. Sasaki, M.; Tachibana, K. Tetrahedron Lett. 2007, 48, 3181. 23. Murata, M.; Kasai, Y.; Umegawa, Y.; Matsushita, N.; Tsuchikawa, H.; Matsumori, N.; Oishi, T. Pure Appl. Chem. 2009, 81, 1123. 24. Zhou, Y.; Morais-Cabral, J. H.; Kaufman, A.; Mackinnon, R. Nature 2001, 414, 43. 25. Lorin, A.; Charloteaux, B.; Fridmann-Sirkis, Y.; Thomas, A.; Shai, Y.; Brasseur, R. J. Biol.Chem. 2007, 282, 18388. 26. Dawson, P. E.; Muir, T. W.; Clark-Lewis, I; Kent, S. B. Science 1994, 266, 776. 27. Rigaud, J.-L.; Lévy, D. Method Enzymol. 2003, 372, 65. 28. Hu, M.; Konoki, K.; Tachibana, K. Biochem. Biophys. Acta 1996, 1299, 252. 29. Tamm, L. K.; Tatulian, S. A. Quart. Rev. Biophys. 1997, 30, 365. 30. Castano, S.; Desbat, B. Biochem. Biophys. Acta. 2005, 1715, 81. 31. Goormaghtigh, E.; Raussens, V.; Ruysschaert, J. M. Biochem. Biophys. Acta. 1999, 1422, 105. 32. Wu, P. G.; Brand, L. Anal. Biochem. 1994, 218, 1. 33. Yasumoto, T.; Takizawa, A. Biosci. Biotech. Biochem. 1997, 61, 1775. 34. Ui, M.; Tanaka, Y.; Tsumuraya, T.; Fujii, I.; Inoue, M.; Hirama, M.; Tsumoto, K. J. Biol. Chem. 2008, 283, 19440. 35. Ui, M.; Tanaka, Y.; Tsumuraya, T.; Fujii, I.; Inoue, M.; Hirama, M.; Tsumoto, K. Mol. BioSyst. 2011, 7, 793.
21
36. Landschulz, W. H.; Johnson, P. F.; Mcknight, S. L. Science 1988, 240, 1759. 37. Satake, M.; Terasawa, K.; Kadowaki, Y.; Yasumoto, T. Tetrahedron Lett. 1996, 37, 5955. 38. Ujihara, S.; Oishi, T.; Mouri, R.; Tamate, R.; Konoki, K.; Matsumori, N.; Murata, M.; Oshima, Y.; Sugiyama, N.; Tomita, M.; Ishihama, Y. Bioorg. Med. Chem. Lett. 2010, 20, 6443. 39. Draisci, R.; Ferretti, E.; Palleschi, L.; Marchiafava, C.; Poletti, R.; Milandri, A.; Ceredi, A.; Pompei, M. Toxicon 1999, 37, 1187. 40. Matsuzaki, K.; Shioyama, T.; Okamura, E.; Umemura, J.; Takenaka, T.; Takaishi, Y.; Fujita, T.; Miyajima. K. Biochim. Biophys. Acta 1991, 1070, 419. 41. Takayama, M.; Itoh, S.; Nagasaki, T.; Tanimizu, I. Clin. Chim. Acta. 1977, 79, 93.
Supplementary Data Supplementary data associated with this article can be found, in the online version, at… Figure Legend Figure 1. A hypothesized recognition model of a ladder-shaped polycyclic ether (LSP) and a membrane protein. Figure 2. Gel filtration chromatograms of the fluorescently labelled peptides reconstituted in liposomes: (a) first gel filtration; (b) second gel filtration of combined fractions 4–6 shown in the first chromatogram. Figure 3. Interaction analysis strategy of transmembrane model peptides and YTX using Förster resonance energy transfer (FRET). The relative orientation (parallel or antiparallel) of the chromophores is not known.
22
Figure 4. FRET efficiency of the peptides. Figure 5. Helical wheels of the α-helical model peptides: (a) artificial sequences; (b) membrane protein-derived sequences. Figure 6. Schematic depiction of the mechanism mediating the interaction of YTX and the α-helical peptides, as proposed by FRET and ATR-IR experiments. orientation (parallel or antiparallel) of the chromophores is not known.
23
The relative
Figures
Figure 1. A hypothesized recognition model of a ladder-shaped polycyclic ether (LSP) and a membrane protein.
24
Figure 2. Gel filtration chromatograms of the fluorescently labelled peptides reconstituted in liposomes: (a) first gel filtration; (b) second gel filtration of combined fractions 4–6 shown in the first chromatogram.
25
Figure 3. Interaction analysis strategy of transmembrane model peptides and YTX using Förster resonance energy transfer (FRET). The relative orientation (parallel or antiparallel) of the chromophores is not known.
26
Figure 4. FRET efficiency of the peptides. 27
a
b
Figure 5. Helical wheels of the α-helical model peptides: (a) artificial sequences; (b) membrane protein-derived sequences.
28
Figure 6. Schematic depiction of the mechanism mediating the interaction of YTX and the α-helical peptides, as proposed by FRET and ATR-IR experiments. orientation (parallel or antiparallel) of the chromophores is not known. 29
The relative
Table 1. Amino acid sequences of α-helical model peptides. Peptide
N-CAP
1 2 3 4 5 6*
CGGKKK CKKK CGEP CGEP EP
1
5
10
15
20
C-CAP
25
KALKALAKLAKLWAKALAKLAKLAGGCGG KALKALAKLAKLWAKALAKLAKLA AALSALASLASLWASALASLASL AALAAALAAALWAALAAALAAA EITLIIFGVMAGVIGTILLISYGI EITLIIFSVMASVISTILLISYGI EITLIIFIVMAIVIGTILLISYGI
KKK RRR RRL RRL RRL
* 6 was not used in this study.
Table 2. The α-helix angle of the peptide and acyl chain of the lipid. R means the absorbance ratio of 0 degree to 90 degree polarization. Peptide
R of Amide I
α-helix angle
R of CH2
Lipid angle
1 2
1.95 1.95
Magic angle Magic angle
1.5 1.5
42° 42°
3 4
3.1 4.1
36° 24°
1.6 1.6
44° 44°
5
3.1
36°
1.6
44°
30
[Graphical Abstract]
31