Hydrazone-modulated peptides for efficient gene transfection

Please do not adjust margins a. Singular Research Centre in Chemical Biology and Molecular Materials (CIQUS), Organic Chemistry Department, University of Santiago de Compostela (USC), Santiago de Compostela, Spain. b. Centro de de Investigação em Química da Universidade do Porto (CIQUP), Department of Chemistry and Biochemistry, Faculty of Sciences, University of Porto, R. Campo Alegre, s/n, P-4169-007 Porto, Portugal. * Correspondence should be addressed to: javier.montenegro@usc.es Electronic Supplementary Information (ESI) available: [details of any supplementary information available should be included here]. See DOI: 10.1039/x0xx00000x Received 00th January 20xx, Accepted 00th January 20xx


Introduction
2][3][4][5] Although substantial advances have been made, we are still far from realizing the great potential that genetic manipulation will have for the future of humankind.Continuous breakthroughs and the development of different technologies have permanently been fuelling the field of genetic engineering and gene therapy. 1,4,5 Currently, the tool set for genetic manipulation can be divided into two main groups: transient active molecules and genome edition.The first group employs transient molecules such as poly or oligonucleotides to modulate gene expression. 1 The second group of editing techniques comprises several programmable nucleases such as the zinc fingers, the transcription activator-like effectors (TALENs) and the CRISPR-associated nuclease Cas9.
The application of transient active nucleotides for genetic modifications started with the introduction (transfection) of exogenous recombinant plasmid DNAs inside cells. 4In these protocols, the foreign DNA is delivered inside the nucleus to access the transcription machinery of the cell for the subsequent mRNA synthesis and protein expression.The following antisense techniques were based in the delivery of a single stranded oligonucleotide that binds the messenger RNA (mRNA), which is therefore inhibited for protein translation.
The discovery of the RNA interference mechanism (RNAi) has triggered new opportunities for genetic manipulation by interfering the gene expression at the cytoplasm post-transcriptional level. 1,3,6 Among these techniques we can find small interfering RNA (siRNA) and microRNAs.In general, the siRNA is a short exogenous double stranded oligonucleotide of about 20-25 base pairs.After delivery in the cytoplasm, the siRNA is processed by an enzyme called DICER and then integrated into the RNA-induced silencing complex (RISC).
The RISC complex chops the target mRNA in small pieces, a process that finally results in the inhibition of the synthesis of the corresponding encoded protein. 6On the other hand, microRNAs are single stranded oligonucleotides that are endogenously produced and processed by the cell in order to bind, sometimes promiscuously, to one or more mRNAs and inhibit the corresponding protein synthesis. 7Recently, the delivery of mRNAs has also emerged as a promising technique for the transient expression of the therapeutic protein of interest. 8This methodology avoids permanent recombination and allows the control over the protein expression until the mRNA is degraded.
Finally, the rising genome-editing techniques require the expression (or delivery) of nucleases, which are enzymes that catalyse DNA cleavage. 2Most commonly, these nucleases are designed to specifically recognise a particular loci in the genome and generate double strand breaks that can be edited by different cellular repair mechanisms.The simplicity and specificity of the CRISPR approach turn this technique as the most promising strategy towards gene This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins edition. 2 Paradoxically, the most efficient CRISPR genome edition is achieved by the transfection of the plasmids that encode for the corresponding nucleases. 2It is therefore of critical importance to innovate and to expand the chemical tool set for plasmid and polynucleotide intracellular delivery.
However, despite all the advances that have already been achieved, the previously described techniques suffer from important limitations.These could be related to the nucleotide cargo or to the delivery vehicle. 1,4,5,9The nucleotide cargo suffers from either potential permanent recombination of plasmids, from the sensitivity towards nuclease degradation of short oligonucleotides such as siRNA and from the problematic immunological response that can be triggered for any kind of nucleotide. 5The delivery vehicle can be either a viral or a non-viral vector. 1,3,5,101][32][33] However, non-viral vectors still present important barriers to overcome such as the poor packing and protection of the nucleotide cargo, the low stability of the resulting complex, the immune response, the escape from the endocytic route and most importantly, the low efficiency and the high cytotoxicity. 1,3 5][36] The field of nucleotide intracellular delivery by penetrating peptides has been developed by important contributions of authors such as Szoka, Divita, Deshayes, Andaloussi, Futaki, Langel, Gait, Dowdy and many others.9,16,18,24,36 The peptide vector can be covalently attached or non-covalently conjugated to the nucleotide cargo.In the later case, the nanoparticles formed, the polyplexes, exploit the ion pairing that is established between the anionic cargo and the cationic peptide. 16 As with the DNA/lipid formulations (lipoplexes) the peptides that form efficient polyplexes generally require the presence of hydrophobic residues that lead to the formation of amphiphilic nanoparticles. 16 8][39][40] This strategy has already been employed for the delivery of siRNA by peptide dendrons 28 and polymers. 27However, the hydrazone linkage has not been so far applied to the modulation of linear peptide sequences.Furthermore, although this methodology afforded efficient transfecting reagents for small double stranded nucleotides (i.e.siRNA), the same molecules that delivered short RNAs failed for the transfection of high molecular weight DNA plasmids in living cells.

27,28
In this paper, we exploit the concept of hydrazone activation for the transfection, for the first time, of a plasmid DNA by a cationic linear peptide template (Fig. 1).This parent peptide, bearing hydrazyde connectors, was combined with a broad range of aldehyde tails for the straightforward preparation and screening of hydrazonemodulated peptides for the delivery of plasmid DNA in living cells.We demonstrate the suitability of this methodology for the identification of excellent peptide vehicles that work with similar or even higher efficacy and lower toxicity than the routine commercial reagents (e.g.Lipofectamine 2000).

Results
Design and peptide synthesis.The peptide scaffold was prepared by solid phase peptide synthesis using a standard Fmoc solid phase strategy (Fig. 2 and ESI). 41Briefly, we started from a Rink amide solid support, which was subjected to consecutive deprotection and coupling cycles for the growth of the linear peptide sequence that was terminated by acetylation of the N-terminus (Fig. 2).The Please do not adjust margins Please do not adjust margins selective cleavage, under slightly acidic conditions, of the methyltrityl (Mtt) group (Fig. 2 in red) allowed the "on resin" amide coupling of the hydrazide connector 2 that was synthesized in a single step from glutaric anhydride and tert-butylcarbazate (Fig. 2 and ESI).The resin was finally treated with strong acid (TFA) for the deprotection of the side chain protecting groups and cleaved from the solid support (Fig. 2).The peptide was precipitated in Et 2 O, centrifuged and thoroughly washed with Et 2 O.The solid residue was re-dissolved in water, purified by reverse phase HPLC and characterized (see ESI and Fig. S1).The peptide sequence was designed to give rise to an amphiphilic alpha helix where the two reactive hydrazides would be aligned and at the interphase between the cationic and the hydrophobic domains (Fig. 1).The peptide (P1) showed a low helical content in PBS buffer but helicity increased after conjugation with certain aldehydes (vide infra).Please do not adjust margins Please do not adjust margins normalized against a control experiment with Lipofectamine 2000 (Fig. 4).To perform a broad screening of tails we initially performed transfection experiments at four fixed charge (+/-) ratios for a selection of 28 different aldehydes tails (Fig. 4).
We could confirm that short aliphatic and aromatic aldehydes did not perform as good activators for the transport of plasmid DNAs, although a slight level of transfection was observed for cyclohexanecarboxaldehyde (T 1 ), benzaldehyde (T 3 ) and hexanal (T 7 ) (Fig. 4).However, as previously observed for the delivery of siRNA with peptide dendrons, 28 the best aldehyde activators were long unsaturated hydrophobic molecules (Fig. 4).From this first screening we could identify 10-undecenal (T 19 ), dodecanal (T 20 ), oleic aldehyde (T 25 ) and linoleic aldehyde (T 27 ) as the best hydrophobic tails to activate plasmid delivery (Fig. 4).Dodecanal (T 20 ) was the only fully aliphatic molecule that triggered plasmid delivery with an efficiency comparable to the commercial reagent Lipofectamine 2000 (Fig. 4).Remarkably, the unsaturated oleic aldehyde was able to efficiently deliver the plasmid DNA at three different charge (+/-) ratios (1, 5 and 10).We therefore decided to fix the best charge ratio of the four best hits (P1T 19 , P1T 20 , P1T 25 and P1T 27 ) and perform dose-response transfection experiments (Fig. 5C).These experiments confirmed the unsaturated linoleic and oleic aldehydes as the best tails for plasmid transfection, leading to even higher transfection efficiencies than Lipofectamine 2000 (Fig. 5B).Interestingly, although dodecanal (T 20 ) did not achieve the best efficiency, this aldehyde tail maintained a good transfection activity even at the lowest concentration of 1 µM (Fig. 5C).
We next evaluated the toxicity of the hydrazone-modulated peptides (Fig. 5D).We employed the colorimetric MTT assay that quantifies cell viability by measuring the mitochondrial activity upon reduction of Thiazolyl Blue Tetrazolium Blue (MTT) to the purple formazan (see ESI for details).We could confirm that, at the working concentrations of the transfection experiments for three of the hydrazone-modulated peptides (i.e.P1T 20 , P1T 25 and P1T 27 at 10 and 5 µM), the cell viability was higher than that observed for the Lipofectamine 2000 transfection experiments (Fig. 5D).The 10undecenal (T 19 ), with an odd number of carbon atoms, showed some toxicity at the higher concentrations (Fig. 5D).
Flow cytometry analysis was performed in order to validate and further quantify the transfection efficiency of hydrazone-modulated peptides (Fig. S3).P1T 25 (5 µM) was combined with the plasmid pEGFP-C1 at a charge ratio (+/-) of 5 (see ESI for details).The resulting polyplex exhibited a 70% of GFP positive cells from the total cell counts (Fig. S3).This 70% was comparable with the control experiment carried out with Lipofectamine 2000 (Fig. S3D).However, although the percentage of transfected cells was similar for P1T 25 and Lipofectamine 2000, the total number of cells was more than ten times higher for the peptide vector (Fig. S3D).Please do not adjust margins

Please do not adjust margins
To study and characterize the polyplexes upon packing of the plasmid DNA, we carried out a gel retardation assay and we measured dynamic light scattering (DLS) and zeta potential of the peptide conjugated to the oleic aldehyde (P1T 25 ) and the dodecanal (P1T 20 ).Not surprisingly, gel electrophoresis confirmed DNA condensation at the corresponding charge (+/-) ratio and peptide concentration that were employed in transfection experiments (Fig. 5E).Dynamic light scattering (DLS) measurements after incubation of hydrazone-modulated peptides with the plasmid DNA revealed polyplexes of 122 nm for the dodecanal (P1T 20 ) and about 65 nm for the oleic aldehyde (P1T 25 ) (Fig. 5F and Figs.S4, S5).Size and globular shape of peptide polyplexes were confirmed by atomic force microscopy (AFM) and transmission electron microscopy (TEM) (Figs.S6 and S7, see ESI for details).The zeta (ζ) potential confirmed a positive charged surface for both polyplexes of 45 mVs for P1T 20 and of 23 mVs for P1T 25 (Fig. 5F).To study the stability of the polyplexes we measured a time course DLS of the peptide/DNA conjugates (Fig. S5).These experiments showed a progressive time dependent increase in the size and in the polydispersity index (PDI) of the polyplexes.This observation suggested a progressive particle aggregation and decomposition of the conjugates between DNA and the hydrazone-modulated peptide with oleic aldehyde and dodecanal tails (Fig. S5).In clear contrast, when the parent peptide without any hydrophobic tail was combined with the plasmid DNA, the PDI was already high at time zero (Fig. S5).The strong polydispersity for P1 without aldehyde tails clearly confirmed the  Please do not adjust margins Please do not adjust margins importance of hydrophobic tails to obtain stable and functional nanoparticles.
Circular Dichroism.In order to gain further experimental details into the amphiphile structural properties and helical character, circular dichroism (CD) was measured for three hits (P1T 20 , P1T 25 and P1T 27 ) and the parent peptide (P1), in three different conditions: trifluoroethanol (TFE), aqueous buffer (pH = 7.4) and liposomes (Fig. 6 and Fig. S8, see ESI for details).We could observe that hydrazone conjugation with the hydrophobic tails increased the helical content of the parent peptide in aqueous conditions and in liposomes (Fig. 6A).This increase in the helicity of the peptide was not observed in the presence of trifluoroethanol, a less competitive solvent for intramolecular hydrogen bonding.The maximum percentage of helicity in aqueous conditions (around 35%) was achieved for P1 after hydrazone formation with the long unsaturated linoleic aldehyde tail (T 27 ) (Fig. 6).However, this strong helicity of P1T 27 was sensitive to temperature increase of the buffered solutions (Fig. 6B in buffer).In contrast, the highest helical content in liposomes was achieved for the P1T 25 , the hydrazone peptide containing the oleic aldehyde (Fig. 6B in liposomes).Intriguingly, the helicity was very stable even at higher temperatures when the peptides where incorporated into the lipid bilayer (Fig. 6B).The peptide with the dodecanal tail (P1T 20 ) showed an intermediate level of helical increase in buffer and in liposomes (Fig. 6).Taken together, these observations revealed the impact of the aldehyde tails in the secondary structure of the peptide, a condition that was probably due to the enhanced hydrophobic effect in water and to the potential interactions with the hydrophobic part of the lipid bilayer (Fig. 6A).Please do not adjust margins Please do not adjust margins

Computational Chemistry
In order to get further insights in the peptide structural behaviour and mode of action, we performed series of all-atoms Molecular Dynamics (MD) simulations of the parent peptide P1 precursor and P1T 25 (Fig. 7 and ESI).We started from a preformed alpha helix and we studied the stability of this helix in water and in membranes.When inserted into a 1-palmitoyl-2-oleoyl-sn-glycero-3phosphocholine (POPC) lipid bilayer, the peptide bearing the hydrazone hydrophobic tails showed a slightly higher stability compared with the parent peptide P1 (Fig. 7B).This observation matches with the enhanced helicity, experimentally observed by circular dichroism, for P1T 25 in liposomes (Fig. 6A).Interestingly, we could observe that, during the MD simulation in water, the oleic tails of P1T 25 embrace the peptide backbone to minimize water repulsion (Fig. 7A).If this "hugged" peptide is approached to a POPC membrane, the peptide anchors to the membrane by inserting its hydrophobic tails into the bilayer.Interestingly, very shortly after landing on the membrane, the hydrophobic tails quickly untied from the backbone of the peptide to extend themselves into the hydrophobic core of the bilayer (Fig. 7C, Movie S1).This tail folding/unfolding process suggested that the hydrophobic side chains allowed the dynamic peptide reorganization in response to different environments (Fig. 7C).MD simulations of a membrane pre-inserted amphiphilic peptide (P1T 25 ) showed an important disruption of the lipidic structure (Fig. 7D and 7E).The peptide was not comfortable into the membrane and water was introduced into the bilayer hydrophobic core (Fig. 7D).3][44] The density maps of the lipids in Fig. 7E suggested a strong disorganization of the bilayer around P1T 25 .This membrane reorganization is a consequence of the severe tilting, with respect to the membrane plane, of the lipids that surround the peptide (Fig. S9).

Discussion
The objective of this work was to demonstrate that hydrazone conjugation enables to modulate the activity of a parent linear penetrating peptide in the delivery of a plasmid DNA.The results reported here confirmed a straightforward methodology involving the preparation and rapid screening of amphiphilic pseudo-helical peptides for gene transfection (Figs.1-4).This approach confirmed several combinations (P1T 19 , P1T 20 , P1T 25 and P1T 27 ) that performed as non-toxic and efficient plasmid delivery vehicles (Fig. 5).Some of these combinations (i.e.P1T 25 ) worked with better efficiency and lower toxicity than the commercial reagents usually employed in plasmid transfection experiments of biochemistry and cell biology laboratories (i.e.Lipofectamine 2000) (Fig. 5).As it should be expected, two of the best hits obtained (P1T 20 and P1T 25 ) efficiently packed the plasmid DNA at the optimal charge (+/-) ratio (Fig. 5E).Dynamic light scattering of these two hits confirmed the formation of nanoparticles with a positive zeta potential (Fig. 5F).
The protocol was simple and allowed the straightforward identification of three peptides that performed as optimal formulations for plasmid delivery.The hydrazone modulation was carried out in physiological compatible conditions.Therefore, the final amphiphilic peptides were readily combined with the DNA and incubated with the cells, without the need of any isolation or purification steps and/or any other special treatment (i.e.Opti-MEM).Circular dichroism in water and in liposomes indicated an increase in the helical character of the peptides after the conjugation with the hydrophobic tails (Fig. 6).This increase was particularly important for the most efficient amphiphilic peptides bearing long unsaturated hydrophobic tails such as oleic (T 25 ) and linoleic (T 27 ) aldehydes.Computational chemistry revealed that the hydrophobic tails flipped over the peptide backbone in aqueous buffer.However, these tails were extended again to anchor the peptide to the membrane and to span the lipid bilayer.This observation suggests that the flexibility of the hydrophobic tails may play an important role in the anchoring process of the peptide to the lipid membrane.In the reported ensemble peptide folding and polyplex formation are controlled by thermodynamics and kinetics. 45As shown by DLS, the hydrophobic tails were crucial for polyplex formation and stability (Figs.S4 and S5).Enthalpy driven hydrogen bonds would be responsible for peptide helicity but the entropic contribution of the hydrophobic tails seems to be of great importance for the formation of functional polyplexes.However, we should always consider the importance of counterion exchange. 46This counterion complexes are thermodynamically very stable but kinetically very labile. 467][48] Although more experiments would be needed to confirm this hypothesis, this hydrophobic folding/unfolding behaviour could have important implications in both the DNA packing and the cargo delivery of hydrazone-modulated peptides.
As previously described for other non-viral gene delivery systems, the recombination of charges (+/-) during the delivery process could lead to the formation of pores, or membrane holes, from where the cargo can escape and perform its corresponding function.

26,42-44
Interesting implications of the phase transition from lamellar to hexagonal 43 and gyroid cubic phases 13 have been studied for lipoplexes. 13,43Recently, it has been proposed that the monomer release from the supramolecular nanoparticles is the critical step to achieve an efficient delivery and not the particular nanoparticles themselves (poly or lipoplexes). 26Nevertheless, the presence of hydrophobic flexible pendants seems to be a critical structural feature for penetrating peptides in order to trigger the efficient delivery of supramolecular conjugated payloads such as nucleotides and proteins.

23,24
The purpose of this paper was to validate the methodology of hydrazone modulation for nucleotide transport across the cell membrane.We have applied this methodology for the particular case of linear peptides and plasmid transfection, two key features that were not demonstrated before. 27,28,37,40,417][28] We believe that we should pay attention to the molecular level features of non-viral vectors in order to understand and improve the efficacy of the bigger nanoplexes. 13,26In this study we have strictly focused in the hydrazone modulation from the synthetic and screening point of view.However, we hope that at due time, not only the synthetic, but also the dynamic properties of the hydrazone bond will help us in understanding and developing novel and improved gene and drug delivery vehicles.

Conclusions
Independently of the nature of the different nucleotides required for gene therapy they all need to be efficiently delivered.Therefore, the development of new tools for nucleotide delivery stands as a major challenge for chemistry and materials science.In this paper we have developed the hydrazone modulation of a parent penetrating peptide to trigger the cellular transfection of a DNA plasmid.This approach allowed us to identify three hydrazone modulated amphiphilic peptides that delivered a plasmid DNA in HeLa cells with better efficiency and less toxicity than the standard commercial reagents employed in routine transfection experiments.

Experimental Methods
Detailed methods and protocols for peptide synthesis and characterization, cell transfection and viability experiments, microscopy images as well as computational methods can be found in the electronic supplementary information (ESI) online.

Figure 1 .
Figure 1.Heptad-based wheel diagram for the potential helical conformation of peptide P1.Representation of the potential amphiphilic helical structure bearing two reactive hydrazides (in red).Treatment of P1 with a hydrophobic aldehyde affords an amphiphilic cationic peptide that complexes, transports and delivers a plasmid DNA across the membrane of HeLa cells.Colour code: Arginine: blue; leucine: orange; hydrazide: red; hydrophobic aldehyde: purple; pEGFP-C1 plasmid DNA: green.
Treatment of peptide P1 with different aldehydes afforded the corresponding hydrazones after two-hour incubation in DMSO/H 2 O (1:1) at 60 o C (Fig. 3 and Fig. S2, see ESI for details).These physiological compatible conditions allowed the direct combination of the hydrazone-modulated peptides with a DNA plasmid and the screening of the corresponding polyplexes for gene delivery.Polyplex screening.We started transfection experiments in HeLa cells with a plasmid that encoded for the expression of the enhanced green fluorescent protein (EGFP).For transfection experiments, after hydrazone formation, the amphiphilic peptides were readily incubated with the plasmid DNA (pEGFP-C1) in standard Dulbecco's Modified Eagle's Medium (DMEM) for 30 min at room temperature (see ESI).The polyplexes were diluted in DMEM and incubated with HeLa cells for 4 hours.The medium was then replaced and the fluorescence was quantified using a plate reader at 72 hours post-transfection (Fig. 4, Fig. 5A and 5B, see ESI for details).The fluorescence intensity in arbitrary units was

Figure 3 .
Figure 3.In hydrazone-modulated peptides (e.g.P1T 20), a parent peptide (e.g.P1) and aldehyde tails (e.g.dodecanal T 20 ) are connected via a hydrazone linkage.The physiological compatible conditions allowed the direct combination, without any purification or isolation steps, with a plasmid DNA in standard cell culture medium (e.g.DMEM) and the subsequent screening for polyplex delivery in HeLa cells (see Figure4).

Figure 2 .
Figure 2. Synthesis of P1.A) Solid phase peptide synthesis (SPPS) by Fmoc strategy consists in cycles of deprotection and amide coupling reactions to yield the linear peptide sequence.B) The final amino group of the peptide is terminated by acylation.C) Mtt selective removal allows the "on resin" incorporation of the Boc protected hydrazides pendants.D) TFA treatment cleaves the peptide from the solid support and removes all lateral protective groups to afford P1.

Figure 4 .
Figure 4. Heat map for the preliminary screening of P1 with the different aldehyde tails at different charge (+/-) ratios (10, 5, 1, 0.5).All data were obtained from the transfection experiments in HeLa cells and normalized to the 100 % transfection that was assigned to the corresponding control experiments with the commercial reagent Lipofectamine 2000.All data were obtained from triplicate transfection experiments.

Figure 5 .
Figure 5. HeLa cells transfection and polyplex characterization.A) and B) Superimposed differential interference contrast (DIC) and epifluorescence image in the green channel.A) Control of HeLa cells with plasmid alone.B) After transfection with P1T 25 + pEGFP-C1 C) Normalized transfection efficiency at 4 different concentrations (10, 5, 2.5 and 1 µM from left to right) of peptide P1 conjugated with the tails: T 19 (10-undecenal), T 20 (dodecanal), T 25 (oleic aldehyde) and T 27 (linoleic aldehyde) at a fixed charge ratio of 5 for P1T 19 and P1T 25 and a fixed charge ratio of 10 for P1T 20 and P1T 27 .In blue the results with Lipofectamine 2000 and in green the control with the plasmid alone.D) Cell viability obtained by the MTT assay of the different peptides (P1T n ) at 4 different concentrations (10, 5, 2.5 and 1 µM from left to right) and Lipofectamine 2000.E) Gel retardation assay.Lane 1: plasmid alone; Lanes 2-7: P1T 25 at 6 different charge ratios (10-0.01);Lane 8: Molecular weight marker; Lanes 9-14: P1T 20 at 6 different charge ratios (10-0.01);Lane 15: empty control experiment.[eGFP plasmid DNA] = 0.5 ng/µL in all cases.F) Dynamic light scattering and zeta potential of the plasmid alone and the corresponding DNA/peptide complexes at a 10 charge (+/-) ratio.Values in C) D) F) represent the average of three replicates with the corresponding standard deviation.

Figure 7 .
Figure 7. Computational Chemistry.A) Structures of P1T 25 at t = 0 and t = 30 ns from its simulation in H 2 O. B) Secondary structure evolution (number of residues in helix + turn) along the MD trajectory for P1 precursor (black) and P1T 25 (red).C) Different snapshots at t = 0, 5, 10, 20, 30, 40 and 50 ns from the simulation of P1T 25 approaching the surface of a POPC bilayer, applying an electric potential of 0.01 V/nm.A vdW representation is used for the peptide, where the Arg residues are in blue and the peptide backbone and the hydrophobic tails (T 25 ) in grey.D) Snapshot at t = 50 ns from the simulation of P1T 25 inserted in a POPC membrane.Only the peptide (in grey) and the lipids phosphorous atoms (in ochre) are represented for clarity.Water molecules in the analogous picture show the wetting of the hydrophobic core of the membrane.E) Density maps of lipids along the MD simulations of P1T 25 inserted into a POPC bilayer, averaged for 50 ns.