Phenylalanine Increases Membrane Permeability - Journal of the

Oct 1, 2017 - Video: Taking a Closer Look at Crystal Engineering of Self-Assembled Porous Protein Materials in Living Cells. Crystalline porous materi...
11 downloads 15 Views 2MB Size
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Communication Cite This: J. Am. Chem. Soc. 2017, 139, 14388-14391

pubs.acs.org/JACS

Phenylalanine Increases Membrane Permeability Russell Perkins and Veronica Vaida* University of Colorado Boulder, 215 UCB, Boulder, Colorado 80309, United States S Supporting Information *

room temperature, transition to a ripple phase at ∼38 °C, and a liquid crystal phase at slightly warmer temperatures (∼41 °C).7 DPPC vesicles in the gel phase are very impermeable, while the ripple phase is highly permeable, and the liquid crystal phase is of an intermediate permeability.8,9 Biological cell membranes have lipids in phases resembling both the gel (liquid-ordered, found in lipid rafts) and liquid crystalline (liquid-disordered) phases.10 Due to the low permeability of the gel phase, the liquid crystal phase makes the best biological mimic for our purposes. The permeability properties of DPPC vesicles were measured indirectly using a fluorescent probe. The basis of the measurement is using a membrane-impermeant, self-quenching dye such as the calcium binding fluorescein derivative, calcein. The dye is first encapsulated in the vesicles of interest, and residual dye is removed from the solution around the vesicles. When the vesicles are exposed to a change in osmotic pressure, as depicted in Figure 1, water leaves the vesicle, effectively

ABSTRACT: Biological membranes are a crucial aspect of living systems, controlling the organization and distribution of different chemical components. Control of membrane permeability is especially important for processes such as electron transport in metabolism and signal propagation in nerve cells. In this work, we show that the amino acid phenylalanine produces increased membrane permeability, which is likely responsible for some of the deleterious symptoms associated with high biological phenylalanine concentrations that occur with the genetic disorder phenylketonuria.

M

embrane interactions are critical for all living systems, and the details of these interactions can be investigated using bilayered vesicles as a representative model. In the genetic disorder phenylketonuria (PKU), phenylalanine (Phe) cannot be processed correctly and human serum levels of Phe can exceed 1.2 mM if untreated.1 This elevated Phe concentration often results in brain damage, and, in many parts of the world, infants are tested for PKU at birth.1 Adler-Abramovich et al.2 observed the formation of amyloid-like fibrils of Phe under biologically relevant conditions, as well as cytotoxicity induced from high Phe concentration. In our previous studies,3,4 interactions between Phe and interfacial or model membrane systems were investigated. Our experiments indicated that Phe does not spontaneously aggregate in solution, as was previously claimed,2 although we find evidence that aggregation occurs at the interface.3,4 Large changes to the morphology of a monolayer model membrane were observed both theoretically and experimentally and such changes could result in changes to the permeability of a bilayered system.3 In this work, we find that Phe does, in fact, increase the permeability of a bilayered vesicle membrane. This has profound implications for the study of PKU as well as the fundamental understanding of interactions between aromatic species and membrane systems. Our previous experimental studies3,4 utilized very simple model cell membranes composed of a monolayer of the phospholipid 1,2-dipalmitoyl-sn-glycero-3-phosphatidylcholine (DPPC), a major component of human cell membranes,5 and human lung surfactant.6 This system was advantageous due to its simplicity, which allowed for ease of interpretation but was lacking in its direct biological relevance. The bilayered membrane studies reported here allow for the investigation of permeability using a system that is a better proxy for biology. In this work, we choose to use vesicles composed of DPPC in order to allow for comparison with previous monolayer studies.3,4 DPPC is exceptionally well studied due to its prevalence in biology. DPPC vesicles exist in a gel phase at © 2017 American Chemical Society

Figure 1. Schematic of changes to vesicle size and encapsulated concentrations after osmolyte is added. Yellow dots represent calcein dye, green dots represent added osmolyte.

concentrating the encapsulated calcein. This in turn changes the fluorescence properties of the dye, which can be monitored and calibrated to vesicle size. At longer time scales, as the added osmolyte diffuses inside the vesicle, water returns along with it. This change results in a recovery in fluorescence, which can be used to derive permeability coefficients of the vesicle to the particular osmolyte. This technique has been employed successfully for a number of different vesicle systems and osmolytes.11−13 Both experimental and fitting procedures are discussed at length in the Supporting Information. Vesicles were characterized via cryogenic transmission electron microscopy (cryoTEM). Example images of prepared vesicles are shown in Figure 2. The average vesicle size was found to be ∼80 nm with a significant distribution of sizes, which were approximately Gaussian (Figure S3). This size distribution is ultimately not problematic for our analysis, for reasons discussed at the end of the Supporting Information. Received: August 29, 2017 Published: October 1, 2017 14388

DOI: 10.1021/jacs.7b09219 J. Am. Chem. Soc. 2017, 139, 14388−14391

Communication

Journal of the American Chemical Society

Table 1. Ribose Permeability Coefficients of DPPC Vesicles

Fit then Averaged Averaged then Fit

Control (×10−9 m/s)

5 mM Phe (×10−9 m/s)

20 mM Phe (×10−9 m/s)

2.67 ± 0.05

3.70 ± 0.04

7.00 ± 0.05

2.42

3.37

7.86

The observed trend is robust, regardless of the analysis method used. The “Fit then Averaged” data, where all individual mixing runs were fit separately, is expected to be more reliable, due to low levels of random noise that are observed in individual runs. The measured permeability coefficient, Ps, for ribose is in the same range as previous measurements vesicles composed of 20:15:2 DPPC:Cholesterol:1,2-dipalmitoyl-glycero-3-phosphate, with a Ps value of 0.9 × 10−9 m/s.13 Tyrosine did not produce a significant change in permeability near its solubility limit of 2.5 mM15 (Supporting Information), which is consistent with previous experiments.4 This increase in permeability is consistent with several features that we observed in previous experiments, where Phe addition resulted in larger (>10 μm) continuous regions of lesscondensed film.3 In MD simulations, small clusters of Phe caused small pores to form in the DPPC film.3 We have also reported indirect thermodynamic evidence that formation of Phe clusters occurs at the water−DPPC or water−air interface.4 In the work of Adler-Abramovich et al.,2 it is suggested that large Phe aggregates form in neat solution as well as several membrane-containing solutions. We have published experimental evidence concluding that Phe does not aggregate in bulk solution; however, the aggregation of Phe in a membrane is supported by our work.3,4 It is likely that the change in permeability is due to either a large-scale change in the morphology of the membrane and/or due to a small scale change immediately around a Phe cluster or aggregate. There have been several studies of what molecular level structures Phe aggregates may adopt.16−19 However, none of these investigations were performed or simulated under similar conditions to ours, that is, at low bulk concentration in an aqueous−DPPC interfacial region. These studies may still give hints as to the structure of Phe aggregates in this system. It seems likely that Phe would adopt a high aspect ratio, needlelike structure, as those have been found under a variety of

Figure 2. Example cryo-TEM images of unilamellar vesicles prepared without (left) and with 20 mM phenylalanine (right). Samples were frozen from room temperature after 1 day of incubation time. Darker areas indicate regions of increased electron density, and correlate with the two lipid layers. No visible changes are observed between the two samples.

Cryo-TEM images of the prepared vesicles do not show significant differences in size or shape due to the presence of Phe. If large Phe aggregates were present in solution, they should be visible to this technique. No aggregates of this type are observed, which is consistent with our previous findings.3,4 Inclusion of Phe into the membrane, either alone or as smaller aggregates, would not be visible in these images, and could still alter membrane behavior. It is possible that the Phe−vesicle interactions differ at higher temperature due to the change in vesicle phase;14 however, the size distribution should be conserved over the short time scale that the permeability experiments are performed on. Fluorescence assays were carried out in a stopped-flow setup, in order to achieve fast, even mixing, and high time resolution. Data that have been averaged and converted into vesicle volume fraction are shown in Figure 3. Even by inspecting this data by eye, it is apparent that Phe causes an increase in vesicle volume recovery, which due to an increase in permeability. This change is visible even at the relatively low concentration of 5 mM. Fitting of this data was performed in two ways, either by fitting individu-al traces and averaging permeability coefficients, or by averaging data and fitting a single permeability coefficient. Details of these procedures are available in the Supporting Information, and permeability coefficients obtained are listed in Table 1.

Figure 3. Vesicle volume fraction data for vesicle solutions following the addition of ribose. This data was first averaged as fluorescence, and then converted to volume fraction using the calibration that is discussed in the Supporting Information. 14389

DOI: 10.1021/jacs.7b09219 J. Am. Chem. Soc. 2017, 139, 14388−14391

Communication

Journal of the American Chemical Society conditions and contrast the crystal structure of solid Phe.17 Phe−Phe dimers,17,20 and larger peptides,21 appear to adopt similar needle-like structures, with larger hollow cores. This may suggest that the mechanism altering permeability is more local around a Phe “needle” embedded in the membrane. This mechanism could be similar to the mechanism of modern membrane proteins, where a protein channel penetrates the bilayer, extending into the interior and exterior aqueous environments. Regardless of the exact details of aggregation and assembly, the ability of Phe to alter membrane permeability is a likely cause of the damage associated with untreated PKU. This is consistent with previous studies exposing cytotoxic effects of low concentration Phe,2 as well as the detection of aggregates both immunologically and microscopically in cells cultured with Phe.2 It also seems consistent with the symptoms of untreated PKU, such as intellectual impairment and seizures,1 due to the stringent permeability requirements of nerve cells for proper function.22 In particular, permeability changes observed here are consistent with the observation that brain abnormalities in adults23 are reversed by decreasing Phe serum levels.24,25 This finding that even relatively small concentrations of amino acid are capable of altering membrane permeability, though important to the study of PKU, also has applications in other areas. Research in the origins of life has many different approaches and has focused on many difficulties in transforming a nonliving system into a living one, as well as the transformation of a very primitive “living” system into something more reminiscent of modern life. One key transition is generation of proteins from amino acids. In modern biology, this is carried out by protein enzymes and mediated by RNA templates. Clearly, the synthesis of proteins from enzymes that are themselves made of protein is not an acceptable source for the first proteins. Some RNA sequences have been shown to possess catalytic activity,26 leading to a common theory that the first generation of protein was catalyzed by RNA.27 Though this view certainly has merits, it is not obvious why the generation of random sequences of polypeptide would be advantageous to any organism. Without some advantage to the system, there is an apparent lack of selective pressure that would lead to the synthesis of the first proteins. In this work, however, we have shown that Phe is able to alter membrane permeability, likely because it is able aggregate within the membrane. Phe−Phe dimers have been shown to aggregate very efficiently into tubular structures, although they adopt additional structures depending on the conditions.17,20,28,29 Larger phenylalanine peptides also appear to function as efficient membrane channels.21 Given this behavior, it is not unreasonable to assume that useful interactions with membranes could set a direct path from amino acid monomers to functional proteins without the need for genetic encoding. Additionally, the water-restricted environment of the interfacial region has been shown to promote a variety of chemical reactions,30−33 including peptide bond formation.34 Although Phe is not thought to be a particularly primitive amino acid, based on its genomic coding,35,36 these experiments provide a proof-of-concept that aggregates of amino acids can fill at least some of the same roles as proteins, in an environment that favors peptide bond formation without enzymatic assistance. This ultimately complements recent work discussing different types of interactions between small molecules, peptides, and prebiotic membranes.37−41

In this work, we have shown that phenylalanine has the ability to alter the permeability of bilayered vesicles, serving as model membranes. This ability may be due to its ability to assemble into aggregates within the interfacial region. We propose that this permeability change drives the deleterious symptoms associated with elevated Phe serum levels in PKU. It is consistent both with the symptoms that are observed, as well as their reversible nature. The alteration of membrane permeability by Phe further relates to the origin of life. Membrane proteins are an essential component of modern biology, and the ability of a simple amino acid to perform some functions of a membrane protein helps bridge the gap between living and nonliving systems. This system is particularly advantageous in that regard, due to the ability of the water limited interfacial region to promote nonenzymatic peptide bond formation. Intermediate systems between isolated amino acids and proteins even appear to be beneficial, due to the aggregation-prone nature of Phe dimers and small polymers.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b09219. Experimental methods and instrumentation, including vesicle preparation, vesicle characterization, dye characterization, and details of the fitting procedure to obtain permeability coefficients (PDF)



AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Russell Perkins: 0000-0002-2351-2362 Veronica Vaida: 0000-0001-5863-8056 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS R.J.P. and V.V. acknowledge support from NSF CHE 161107, and NASA Habitable Worlds grant NNX15AP20G. R.J.P. acknowledges support from the NIH/CU Molecular Biophysics Training Program. The authors acknowledge Dr. Sheref Mansy for sharing his knowledge about vesicle preparation and permeability assays, as well as Dr. Annette Erbse for assistance with fluorescence and fluorescence stopped-flow measurements. Electron microscopy was done at the University of Colorado, Boulder EM Services Core Facility in MCDB, with the technical assistance of facility staff.



REFERENCES

(1) Hanley, W. B. Am. J. Med. 2004, 117 (8), 590. (2) Adler-Abramovich, L.; Vaks, L.; Carny, O.; Trudler, D.; Magno, A.; Caflisch, A.; Frenkel, D.; Gazit, E. Nat. Chem. Biol. 2012, 8 (8), 701. (3) Griffith, E. C.; Perkins, R. J.; Telesford, D.-M.; Adams, E. M.; Cwiklik, L.; Allen, H. C.; Roeselová, M.; Vaida, V. J. Phys. Chem. B 2015, 119 (29), 9038. (4) Perkins, R. J.; Kukharchuk, A.; Delcroix, P.; Shoemaker, R. K.; Roeselová, M.; Cwiklik, L.; Vaida, V. J. Phys. Chem. B 2016, 120 (30), 7408. 14390

DOI: 10.1021/jacs.7b09219 J. Am. Chem. Soc. 2017, 139, 14388−14391

Communication

Journal of the American Chemical Society

(41) Gordon, M.; Black, R. A.; Blosser, M. C.; Keller, S. L. Biophys. J. 2015, 108 (2), 542a.

(5) Sampaio, J. L.; Gerl, M. J.; Klose, C.; Ejsing, C. S.; Beug, H.; Simons, K.; Shevchenko, A. Proc. Natl. Acad. Sci. U. S. A. 2011, 108 (5), 1903. (6) Goerke, J. Biochim. Biophys. Acta, Mol. Basis Dis. 1998, 1408 (2), 79. (7) Nagarajan, S.; Schuler, E. E.; Ma, K.; Kindt, J. T.; Dyer, R. B. J. Phys. Chem. B 2012, 116 (46), 13749. (8) Xiang, T.-X.; Anderson, B. D. Biochim. Biophys. Acta, Biomembr. 1998, 1370 (1), 64. (9) Xiang, T.-X.; Xu, Y.-H.; Anderson, B. D. J. Membr. Biol. 1998, 165 (1), 77. (10) Sezgin, E.; Levental, I.; Mayor, S.; Eggeling, C. Nat. Rev. Mol. Cell Biol. 2017, 18 (6), 361. (11) Chen, P. Y.; Pearce, D.; Verkman, A. S. Biochemistry 1988, 27 (15), 5713. (12) Saito, H.; Shinoda, W. J. Phys. Chem. B 2011, 115 (51), 15241. (13) Sacerdote, M. G.; Szostak, J. W. Proc. Natl. Acad. Sci. U. S. A. 2005, 102 (17), 6004. (14) Gobrogge, C. A.; Blanchard, H. S.; Walker, R. A. J. Phys. Chem. B 2017, 121 (16), 4061. (15) CRC Handbook of Chemistry and Physics, 96th ed.; Haynes, W. M., Ed.; CRC Press: Boca Raton, FL, 2015. (16) Do, T. D.; de Almeida, N. E. C.; LaPointe, N. E.; Chamas, A.; Feinstein, S. C.; Bowers, M. T. Anal. Chem. 2016, 88 (1), 868. (17) German, H. W.; Uyaver, S.; Hansmann, U. H. E. J. Phys. Chem. A 2015, 119 (9), 1609. (18) Singh, V.; Rai, R. K.; Arora, A.; Sinha, N.; Thakur, A. K. Sci. Rep. 2015, 4, 3875. (19) Singh, P.; Brar, S. K.; Bajaj, M.; Narang, N.; Mithu, V. S.; Katare, O. P.; Wangoo, N.; Sharma, R. K. Mater. Sci. Eng., C 2017, 72, 590. (20) Rissanou, A. N.; Georgilis, E.; Kasotakis, E.; Mitraki, A.; Harmandaris, V. J. Phys. Chem. B 2013, 117 (15), 3962. (21) Shen, Y.; Si, W.; Erbakan, M.; Decker, K.; De Zorzi, R.; Saboe, P. O.; Kang, Y. J.; Majd, S.; Butler, P. J.; Walz, T.; Aksimentiev, A.; Hou, J.; Kumar, M. Proc. Natl. Acad. Sci. U. S. A. 2015, 112 (32), 9810. (22) Voet, D.; Voet, J. G. Biochemistry, 4th ed.; Wiley: Hoboken, NJ, 2010. (23) Thompson, A. J.; Tillotson, S.; Smith, I.; Kendall, B.; Moore, S. G.; Brenton, D. P. Brain 1993, 116 (4), 811. (24) Cleary, M. A.; Walter, J. H.; Wraith, J. E.; White, F.; Tyler, K.; Jenkins, J. P. J. Pediatr. 1995, 127 (2), 251. (25) Walter, J. H.; White, F.; Wraith, J. E.; Jenkins, J. P.; Wilson, B. P. M. J. Inherited Metab. Dis. 1997, 20 (3), 367. (26) Zaug, A. J.; Cech, T. R. Science 1986, 231 (4737), 470. (27) Kawamura, K.; Maurel, M.-C. Origins Life Evol. Biospheres 2017, 47, 281. (28) Görbitz, C. H. Chem. - Eur. J. 2001, 7 (23), 5153. (29) Guo, C.; Luo, Y.; Zhou, R.; Wei, G. ACS Nano 2012, 6 (5), 3907. (30) Vaida, V. Science 2016, 353, 650. (31) Tinel, L.; Rossignol, S.; Ciuraru, R.; Dumas, S.; George, C. Phys. Chem. Chem. Phys. 2016, 18 (25), 17105. (32) Martins-Costa, M. T. C.; Anglada, J. M.; Francisco, J. S.; RuizLopez, M. F. J. Am. Chem. Soc. 2012, 134, 11821. (33) Griffith, E. C.; Tuck, A. F.; Vaida, V. Acc. Chem. Res. 2012, 45, 2106. (34) Griffith, E. C.; Vaida, V. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (39), 15697. (35) Koonin, E. V.; Novozhilov, A. S. IUBMB Life 2009, 61 (2), 99. (36) Lu, Y.; Freeland, S. Genome Biol. 2006, 7, 102. (37) Black, R. A.; Blosser, M. C.; Stottrup, B. L.; Tavakley, R.; Deamer, D. W.; Keller, S. L. Proc. Natl. Acad. Sci. U. S. A. 2013, 110 (33), 13272. (38) Wilson, M. A.; Wei, C.; Pohorille, A. Origins Life Evol. Biospheres 2014, 44 (4), 357. (39) Mansy, S. S. Cold Spring Harb. Perspect. Biol. 2010, 2 (8), No. a002188. (40) Black, R. A.; Blosser, M. C. Life 2016, 6 (3), 33. 14391

DOI: 10.1021/jacs.7b09219 J. Am. Chem. Soc. 2017, 139, 14388−14391