Enzymatically Forming Cell Compatible Supramolecular Assemblies of Tryptophan-Rich Short Peptides

Abstract

Here we report a new type of tryptophan-rich short peptides, which act as hydrogelators, form supramolecular assemblies via enzymatic dephosphorylation, and exhibit cell compatibility. The facile synthesis of the peptides starts with the production of phosphotyrosine, then uses solid phase peptide synthesis (SPPS) to build the phosphopeptides that contain multiple tryptophan residues. Besides exhibiting excellent solubility, these phosphopeptides, unlike the previously reported cytotoxic phenylalanine-rich phosphopeptides, are largely compatible toward mammalian cells. Our preliminary mechanistic study suggests that the tryptophan-rich peptides, instead of forming pericellular assemblies, largely accumulate in lysosomes. Such lysosomal localization may account for their cell compatibility. Moreover, these tryptophan-rich peptides are able to transiently reduce the cytotoxicity of phenylalanine-rich peptide assemblies. This rather unexpected result implies that tryptophan may act as a useful aromatic building block for developing cell compatible supramolecular assemblies for soft materials and find applications for protecting cells from cytotoxic peptide assemblies.

1. INTRODUCTION

This article reports the enzymatic self-assembly of tryptophan-rich short peptides as cell-compatible supramolecular hydrogelators for generating peptide assemblies. In the last two decades, the self-assembly of peptides has become an active reaseach area.^[1-7] Among them, aromatic group containing short peptides have received considerable attentions in the research of peptide assemblies because aromatic-aromatic interactions^[8] greatly enhance the ability of peptides to self-assemble in water.^[9-12] For example, Gazit et al. reported that a diphenylalanine self-assembles to form nanotubes.^[13] We found that the attachment fluorenylmethyloxycarbonyl (Fmoc) to simple dipeptides, such as diglycine or dialanine, enables the self-assembly of the dipeptide to result in supramolecular hydrogels.^[10] This rather serendipitous observation has led to works that use Fmoc to enable the self-assembly of peptides in water, especially to generate hydrogels for a variety of applications.^[14-30] The exploration of Fmoc naturally has led us and others to explore other aryl groups, such as naphthyl group, for the self-assembly of short peptides.^[31-36] Indeed, naphthyl-capped diphenylalanine (Nap-FF) is particularly effective to enable a large group of molecules to self-assemble in water.^[37]

Besides promoting the self-assembly of peptides, Nap-FF also acts as a key motif for developing phosphopeptides for enzyme-instucted self-assembly (EISA).^[38-46] For example, Nap-D-Phe-D-Phe (Nap-ff) conjugates with a D-phosphotyrosine to form a phosphotripeptide (Nap-ff_py), which undergoes EISA selectively around the cancer cells that overexpress alkaline phosphatases. Such an EISA process results in pericellular hydrogel/nanonets to block cellular mass exchange to induce apoptosis of the cells (e.g., HeLa cells).^[47] The use of Nap-ff in other phosphopeptides also generates EISA precursors that are cytotoxic and selectively inhibit cancer cells.^[48-51] Recently, Kuang et al. also show that the conjugate of Nap-fff and D-phosphotyrosine to produce a phosphotetrapeptide (Nap-fff_py) for selectively eliminate iPS cells via EISA.^[52] Although considerabe number of studies have incorported phenylalanines in the phosphopeptides for EISA, another aromatic amino acid, tryptophan, has received little attention in the exploration of phosphopeptides for EISA, except the conjugation of Fmoc with tryptophan to form peptide assemblies to induce apoptosis.^[53] Thus, we decided to examine the phosphopeptides that contain multiple tryptophan for understanding the roles of heteroaromatic group in the context of EISA.

To compare with Nap-ff_py or Nap-FF_pY, we synthesized Nap-ww_py (1p), Nap-WW_pY (2p). Our study shows that 1p and 2p, with the critical micelle concentrations (CMC) of about 54 and 60 μM, respectively, self-assemble to form nanoparticles, which turn into the nanofibers of Nap-wwy (1) or Nap-WWY (2) upon enzymatic dephophorylation catalyzed by alkaline phosphatase (ALP). The mixture of 1p and 2p, after being dephosphorylated by ALP, affords a slightly stronger hydrogel that those formed by dephosphorylation of 1p or 2p separately. In contrast to Nap-ff_py or Nap-FF_pY,^[47] both 1p and 2p are cell compatible. Moreover, the examination of the relevant analog of 1p and 2p (or 1 and 2) indicates that Nap-WW motif or homochirality favors hydrogelation. The fluorescent analogs of 1p and 2p form moderate amount of fluorescent punta inside cells. The addition of the inhibitor of alkaline phosphatase confirms intracellular dephosphorylation of 1p or 2p to form 1 or 2, which localizes in lysosomes (Scheme 1). Moreover, these tryptophan-rich peptides are able to transiently reduce the cytotoxicity of phenylalanine-rich peptide assemblies. These results implies that tryptophan may serve as a useful building block for developing cell compatible supramolecular peptide assemblies or soft materials.

Scheme 1.

The difference of EISA processes of Nap-ff_py and Nap-ww_py in cellular environment.

2. MATERIALS AND METHODS

2.1. Materials and instruments.

All amino acids were purchased from GL Biochem, and NBD-Cl was purchased from TCI America. All the solvents and chemical reagents were used directly as received from the commercial sources without further purification. All the products were purified with HPLC system (Agilent 1100 Series). Confocal microscopy images were obtained on by ZEISS LSM 880 confocal laser scanning microscope. Electron microscopy imaging was performed on an Morgagni 268 transmission electron microscrope. LC-MS was operated on a Waters Acquity Ultra Performance LC with Waters MICRO-MASS detector.

2.2. Critical micelle concentration (CMC) measurements.

All the solutions of the tryptophan peptides, from the concentration of 500 μM to 1.25 μM, were prepared from 1 mM of stock solutions in pH 7.4 PBS buffer. After incubating with pyrene,^[54] the absorbance from 330 to 400 nm were measured on a Shimadzu RF-5301-PC fluorescence spectrophotometer. Emission spectra of pyrene were acquired by exciting the samples at 332 nm. The fluorescence intensities of the first (373 nm, I₁) and the third (384 nm, I₃) peak were used for calculation of CMC. The value of I₁/I₃ was monitored as a function of the peptides concentration, where CMC can be obtained at the onset of the curve.

2.3. Transmission electron microsopy.

Following the established protocol for EM study,^[55] 5 μL samples were placed on 400 mesh copper grids coated with continuous thick carbon film (~ 35 nm). After the sample was loaded on the grid, stained with the uranyl acetate, then are allowed to dry in air. The TEM images were obtained with Morgagni 268 transmission electron microscrope.

2.4. Cell cultrue.

All cell lines were purchased from the American Type Culture Collection (ATCC). All the cell lines used in this work were authenticate by authenticated by CellCheck 9 - human (9 Marker STR Profile and Inter-species Contamination Test, IDEXX), confirming 100% match of the cell lines. Mycoplaama detection kit (from ATCC) was used to confirm the cell culture free of mycoplasma. The HeLa cells were cultured in Minimum Essential Media (MEM) supplemented with 10% fetal bovine serum (FBS) and antibiotics. The HepG-2 cells were propagated in Dulbecco's Modified Eagle's Medium (DMEM), supplemented with 10 % FBS and antibiotics. The Saos-2 cells were cultured in McCoy's 5A supplemented with 15% FBS and antibiotics. All the cells were incubated in a fully humidified incubator containing 5% CO₂ at 37°C.

2.5. Cell viability assay.

The cytotoxicity was determined by the viability of cells with MTT assay. Cells were seeded in 96-well plates at 1 x 10⁴ cells/well. The cells were allowed to attach to the wells for 24 h at 37 °C, 5% CO₂. The culture medium was removed, and 100 μL culture medium containing the compounds were added to each well. After incubation of 24, 48, and 72 h, each well was added by 10 μL of 5 mg/mL MTT ((3-(4, 5-DimethylthiazoL-2-yl)-2, 5-diphenyltetrazolium bromide), followed by addition of 100 μL of 0.1% sodium dodecyl sulfate (SDS) 4 h later. After incubation of the cells at 37 °C for overnight, the solution was measured in a microplate reader.

2.6. Confocal microscopy.

Cells in exponential growth phase were seeded in glass bottomed culture chamber at 1 x 10⁵ cell/well. The cells were allowed for attachment for 12 h at 37 °C, 5% CO₂. The culture medium was removed, and new culture medium containing desired concentrations of the compound was added. After incubation for certain time, cells were stained with 1.0 μg/ml Hochst 33342 or 1.0 μg/ml Lyso-Tracker for 10 min at 37 °C in the incubator. After rinsing the cells with three times by PBS buffer, the images were taken after addition of the peptides in live cell imaging solution for imaging.

3. RESULTS AND DISCUSSION

3.1. Molecular design.

Scheme 2 shows the structures of the tryptophan-rich peptides and phosphopeptides. The molecules consist of four or five segments: naphthyl motif, tryptophan backbone, amino acid linker, ALP trigger, or a fluorophore. The naphthyl group enhances intermolecular aromatic-aromatic interactions for promoting peptides to self-assemble in water.^[56-58] The tryptophan backbone, which can be L-Trp or D-Trp residues, is the subject of this study. That is, since the indole ring in tryptophan has a permanent dipole moment from nitrogen in the five-membered ring to the phenyl ring, the tryptophan backbone provides dipole-dipole interactions. Furthermore, the larger aromatic ring of tryptophan is available for π-π and/or cation-π interactions respect to phenylalanine or tyrosine. The amino acid linker, which is a glycine (Gly), can be optional or be inserted in any position of the designed peptides. The ALP trigger, a D- or L-phosphotyrosine, serves as the substrate of ALP for EISA. The fluorophore, 4-nitro-2,1,3-benzoxadiazole (NBD), is useful for visualizing the cellular uptake and distribution of the peptide assemblies.^[59]

Scheme 2.

The design of the tryptophan-rich peptides and phosphopeptides.

According to the above design, the key molecules for EISA to generate supramolecular hydrogels/assemblies would be Nap-D-Trp-D-Trp-D-_pTyr (1p) and Nap-L-Trp-L-Trp-L-_pTyr (2p), which would be a pair of enantiomers. As shown in Scheme 2, inserting glycine between naphthyl group and tryptophan in 1p and 2p gives 3p and 4p, respectively. Moving the glycine between tryptophans in 3p and 4p produces 5p and 6p, respectively. These molecules should reveal the effect of the flexibility of backbones in the peptides for self-assembly. Changing the stereochemistry of the second tryptophan in 1p and 2p generates heterochiral 7p and 8p, respectively. Removing the naphthyl group in 1p and 2p leads to 9p and 10p, respectively, which should help determine the contribution of the naphthyl group for the self-assembly of the peptides. To investigate how the number of tryptophan residues affects the formation of assemblies, tetrapeptide (11p) and dipeptide (12p) were synthesized. Attaching NBD to the C-terminal of 1p, 2p, 9p, and 10p creates 1p-NBD, 2p-NBD, NBD-9p, and NBD-10p, respectively, for visualizing cellular uptake and distribution of the assemblies of these tryptophan-rich peptides.

3.2. Synthesis.

We used 2-chlorotrityl chloride resin for the typical Fmoc solid-phase peptide synthesis (SPPS)^[60] to make the peptides shown in scheme 2. We first synthesized D- or L-phosphotyrosine in 90% yield,^[61] followed by the N-Fmoc protection. This step produces Fmoc-Tyr(PO₃H₂)-OH.^[62] According to the procedure of SPPS, using HBTU as the coupling agents in the presence of DIEA in dimethylformamide (DMF) attaches Fmoc-protected tyrosine phosphate to the 2-chlorotrityl chloride resin. Piperidine (20%) removes Fmoc protecting group to expose free amine for the next coupling reactions that attach the Fmoc-protected D- or L-tryptophan amino acids. Except the cases of 9p and 10p, 2-(naphthalen-2-yl)acetic acid directly caps the N-terminal of the peptides on the resin. In the final step, using a trifluoroacetic acid (TFA)/Triisopropylsilane (TIPS)/water (95: 2.5: 2.5 in volume) at room temperature for 2 h cleaves the crude peptides from the resin and deprotects the side chains. Adding diethyl ether precipitates the crude peptides for centrifugation and three times wash. In the cases of 1p-NBD and 2p-NBD, we kept the tert-butyloxycarbonyl (Boc) protecting groups of tryptophan for the coupling reaction with NBD-amine. That is, we used 2,2,2-trifluoroethanol (TFE) in dichloromethane for 2 h to cleave the crude peptides from the resin. After using diethyl ether to precipitate the crude, protected peptides, we used HBTU in the presence of DIEA in dimethylformamide (DMF) to react the peptides with NBD-amine for overnight. Then, we removed the Boc protection. We used reverse phase HPLC and acetonitrile (0.1% TFA) and double-distilled water (0.1% TFA) as the eluents to purify all compounds. The purity of the compounds was examined by LCMS and listed in Table S1.

3.3. Self-assembly and hydrogelation.

To evaluate the self-assembly of the peptides for enzymatic hydrogelation, we used transmission electon microscopy (TEM) to examine the nanoscale assemblies formed before and after the enzymatic dephosphorylation of the phosphopeptides. We treated the solution of the the phosphopeptides with alkaline phosphatase (ALP) and checked whether the corresponding, dephosphorymated peptides form hydrogels or form nanofibers. As shown in Figure 1, without the addition of ALP, the solution of 1p, 2p, or the mixture of 1p and 2p produce nanoparticles on the TEM graids. The addition of ALP (1 U/mL) to the solutions of the 1p (1.0 wt % in PBS buffer) results in a hydrogel that consists the nanofibers of 1. The diameters of the nanofibers, with a width of 7 ± 2 nm, appear to be monodispersed. The order and monodispersity likely are resulted from EISA.^[63] That is, the enzymatic reaction controls the noncovalent interactions and minimizes interfibrillar association for resulting in monodispersity of the diameters of the nanofibers. The addition of ALP (1 U/mL) to the 1.0 wt% of 2p, however, produces a viscous liquid, which contains nanoribbons of 2 with the diameter of 32 nm ± 4 nm. Although 1 and 2 are enantiomer pairs, enzymatic dephosphorylation generate different peptide assemblies. This result indicates that the interaction between the substrate or the product with the enzyme affects the final supramolecular assemblies of the peptides. This result agrees with our previous observation that EISA generates different assemblies of the peptides that are enantiomeric pairs,^[64] and indicates that EISA is an integral process rather than a simple addition of dephosphorylation reaction and self-assembly (as unrelated processes). In addition, this result also agree with the observation that the interactions between proteins and low molecular weight hydrogelators control nanostructures of peptide assemblies.^[65] The addition of ALP to the solution of 1p and 2p (1:1 ratio, total concentration = 1.0 wt%) results in a hydrogel, which contains twisted ribbon and nanofibers. The diameters of the nanofibers still are largely monodispersed, with the diameter of 7 ± 2 nm. The twisted nanoribbons, however, exhibit varied pitches and have diameters of 27 ± 6 nm. Unlike 1p and 2p, 3p-10p are not able to form hydrogels even at 5.0 wt % (Figure S1, table 1, Figure S11) after the addition of ALP. These results suggest several trends for the formation of assemblies from these tryptophan-containing peptides: (i) The naphthyl group, ditryptophan, and tyrosine must be at adjacent position; (ii) tryptophans should be in the same stereochemistry; (iii) glycine, as a flexible amino acid, hardly enhances the self-assembly ability of the ditryptophan-containing peptides; and (iv) naphthyl group enhances the self-assembly of the peptides WWY or wwy.

Figure 1.

TEM and optical images of (A) 1p and (B) 2p or (C) their mixture (1:1) treated with ALP (1 U/mL) after 24 hours at 37 °C in pH 7.4 PBS buffer (scale bar = 100 nm, total concentrations of the molecules = 1.0 wt%). Black arrows indicated the aggregates of the micelles.

Table 1.

Gelations of the peptides after being dephorylated by ALP

Entry no	Peptides	1.0 wt%	1.5 wt%	2.0 wt%	5.0 wt%
1	1p	G	-	-	-
2	2p	G	-	-	-
3	1p + 2p	G	-	-	-
4	3p	P	S	S	S
5	4p	S	S	S	S
6	3p + 4p	P	P	S	S
7	5p	S	S	S	S
8	6p	S	S	S	S
9	5p + 6p	S	S	S	S
10	7p	P	P	-	-
11	8p	P	P	-	-
12	9p	S	S	-	-
13	10p	S	S	-	-

Conditions : The phosphopeptides were incubated with ALP (1 U/mL) for 24 hours at 37 °C in pH 7.4 PBS buffer. G: gel, P: precipitate, S: solution, - : Not tested.

To further quantify the self-assembly ability of the precursors and the hydrogelators, we measured the critical micelle concentration (cmc) of the precursors (1p, 2p) and their corresponding self-assembling molecules (1, 2). As shown in Figure 2, 1p and 2p exhibit comparable cmc values. After the dephosphorylation, the cmc values of 1 and 2 are about a quarter of those of 1p and 2p. The cmc values of 1p-NBD and 2p-NBD are about the half of those of 1p and 2p, indicating that the attachment of NBD at C-terminal of the peptides increases the self-assembling ability of the resulting phosphopeptides. Since it is possible that the purity of sample affects the values of CMC, the CMC of 1p-NBD likely deviates more from its actual value than 2p-NBD does.

Figure 2.

CMC of 1p ( 54 μM), 2p (60 μM), 1 (10 μM), 2 (17 μM), 1p-NBD (27 μM), and 2p-NBD (31 μM) in PBS buffer (pH=7.4).

3.4. Cytotoxicity.

To examine cell compatibilities of the tryptophan-rich peptide hydrogelators towards mammalian cells, we added 1p or 2p into the culture of HeLa cells, HepG2, and Saos2 cells and measured the cell viability using the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay. As shown in Figure 3, after being incubated with 400 μM of 1p and 2p for three days, the viabilities of the cells remains above 80%, with the exception that viability of HepG2 cells is about 60% at Day 3. This result drastically differs from the cytotoxicity of Nap-ff_py,^[47] which potently inhibits HeLa, HepG2, and Saos2 at the concentration of 400 μM. All other peptides shown in Figure 1 are also cell compatible (Figures S3-S10). These results indicate that these tryptophan-rich peptides and their assemblies are largely cell compatible.

Figure 3.

Cell viabilities of HeLa, HepG2, Saos-2 cells treated with 1p and 2p for 3 days. All cell viability test are done in triplicate (n = 3).

Since the tryptophan-rich hydrogelators exhibit excellent cell compatibility at day 1, we decided to test whether they can reduce the cytotoxicity of phenylalanine-rich peptide assemblies formed by in-situ dephosphorylation of Nap-ff_py. Similar to our previous report,^[66] being incubated with Nap-ff_py at 400 μM for 24h, HeLa exhibits viability of about 10%. In this study, the addition of 1p or 2p to co-incubated with 400 μM of Nap-ff_py and HeLa cells, significantly increases the cell viability, to above 60% or 80%, respectively, at 24 h (Figure 4). Since the cell rescue is largely independent to the concentrations of 1p and 2p, it is unlikely that 1p or 2p competes with Nap-ff_py for alkaline phosphatases. However, 1p or 2p is unable to rescue the cells at 48h and 72h incubation of Nap-ff_py, indicating that 1p or 2p only transiently reduces the cell stress caused by the EISA of Nap-ff_py.

Figure 4.

Cell viabilities of HeLa cells treated with Nap-ff_py and treated by the combination of Nap-ff_py (400 μM) and 1p (or 2p) for 24 h. All cell viability test are done in triplicate (n = 3).

3.5. Cellular internalization.

Because, at 400 μM, 2p is slightly more effective than 1p to rescue the cells from Nap-ff_py at 24 h incubation, we speculate that the intracellular proteolysis of 2p may play a role. So we decided to visualize the internalization of 1p-NBD and 2p-NBD, the fluorescent analogs of 1p and 2p, respectively. We use confocal laser scanning microscopy (CLSM) to examine the internalization of 1p-NBD and 2p-NBD at 200 μM in HeLa cells because 1p-NBD and 2p-NBD, at 200 μM, are innocuous to the HeLa cells (Figure S9). Figures 5 show the time-dependent accumulation of the fluorescent peptide assemblies on the HeLa cells incubated with 1p-NBD and 2p-NBD. At 2 hours, HeLa cells uptake 1p-NBD and 2p-NBD to form some green fluorescent puntas, which remain inside cells at 6 hours. The fluorescence of 1p-NBD becomes weaker than 2p-NBD after 12 h incubation, indicating that the assemblies of 1p-NBD likely diffuse out of the cells, though it is also possible that 2p-NBD is proteolytically cleaved to left NBD inside the membrane of lysosomes. To further evaluate the cellular distribution of 1p-NBD or 2p-NBD, we used Lyso-Tracker^[67] for co-staining. As shown in Figure 6A, the green fluorescence from NBD co-localizes well with the red fluorescence signal from the Lyso-Tracker in the cytosol after co-incubation for 2 h, suggesting that 1p-NBD or 2p-NBD, being uptaken by the cells, eventually localized in the lysosomes. Moreover, the green fluorescence on the cell surface indicates that the assemblies of 1p-NBD or 2p-NBD not only enter the cell to accumulate in lysosome, but also self-assemble on the cell surface. We use inhibitors of ALPL/TNAP, (2,5-dimethoxy-N-(quinolin-3-yl)benzenesulfonamide, DQB)^[68] in the cell culture of HeLa to confirm that ALP catalyzed the assemblies of the peptides inside cells because HeLa cells overexpress ALPL. The treatment of DQB reduces the fluorescence on cell surface and inside cell (Figure 6B). This observation indicated that ALPL/TNAP involves the uptake of 1p-NBD and 2p-NBD by the HeLa cells.

Figure 5.

Time dependent fluorescent images of HeLa cells incubated with 200 μM of 1p-NBD and 1p-NBD for 2 hours, 6 hours, and 12 hours. The nuclear dye Hoechst 33342 was used to stain nucleus after the peptide treatment.

Figure 6.

(A) Fluorescent images of HeLa cells incubated with 200 μM of 1p-NBD and 2p-NBD. LysoTracker was used to stain lysosomes after the peptide treatment for 2 hours. (B) Fluorescent images of HeLa cells co-incubated with 200 μM of 1p-NBD and 2p-NBD with or without the addition ALPL/TNAP inhibitor DQB (5 μM). The incubation time is 2 hours.

4. CONCLUSION

In summary, this study, for the first time, reveals that ditrytophan-based hydrogelators and peptide assemblies not only are cell compatible, but also transiently reduce the cytotoxicity of hydrophobic peptides. Considering the association of aberrant hydrophobic peptides with neurodegenerative diseases, a future direction of this study would be to understand the molecular and cellular mechanism of the cell rescue conferred by these tryptophan-containing peptides. Moreover, this work highlights that EISA is a context dependent molecular process. That is, the functions of EISA also depend on the molecular building blocks. This feature requires more extensive molecular design for EISA processes, which is actively explored by multiple labs.^[69-73]