Interactions of antimicrobial peptides with the targets of their nephrotoxic action. Molecular dynamics simulations

doi:10.21203/rs.3.rs-1998579/v1

Download PDF

Research Article

Interactions of antimicrobial peptides with the targets of their nephrotoxic action. Molecular dynamics simulations

https://doi.org/10.21203/rs.3.rs-1998579/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objectives

The study aimed to investigate the structural aspects of polymyxins nephrotoxicity at the atomic level to promote the more purposeful development of the polymyxin’s derivatives with the lower nephrotoxic action.

Materials and methods

The molecular dynamics simulations of the complexes of polymyxin B and its derivative NAB7061 (that carries only three positive charges located within the macrocycle) with megalin were done in program package YASARA Structure with explicit water (TIP3P) and ions (0.9% NaCl) in NPT ensemble using the AMRER03 force field. After 10 ns equilibration, each system was simulated at 298 K and pH 7.4 for a 25 ns production phase. Simulations were run twice for each molecular system.

Results

By molecular dynamics simulations, the possibility was shown for polymyxin to form a stable complex with two neighbor structural domains of megalin in accord with the universal mechanism of binding the cationic ligands by ligand-binding CR repeats of the LDLR-family receptors. It was shown that interactions of megalin with polymyxin are stronger than with its derivative having no positively charged groups outside the macrocycle. The structural prerequisites of these differences were revealed that explain the less nephrotoxicity of such derivatives compared to polymyxin.

Conclusions

Comparative molecular dynamics simulations of megalin interactions with polymyxin B and its derivative NAB7061 that carries no positive charges located outside the macrocycle revealed the possible structural prerequisites for the lower nephrotoxic action of such polymyxin derivatives. The weakening of polymyxins binding with megalin may become an effective preventive measure against polymyxin-induced nephrotoxicity.

antimicrobial peptides

polymyxin B

nephrotoxic effect

megalin

molecular dynamics simulation

The rapid spread of multidrug-resistant gram-negative bacterial strains has necessitated the search for more efficient antimicrobial agents and prompted a renewed interest in polymyxins which have been invaluable for the therapy of serious nosocomial pathogens but which have been withdrawn due to their nephrotoxicity [1–5]. The group of polymyxin peptides contains several chemically different compounds (polymyxins A-E, M, S, and others) [6]. In clinical practice, only polymyxin B and polymyxin E (commonly known as colistin) are used. Polymyxin B (PMB, polymyxin, PmB) is a cyclic lipodecapeptide consisting of ten amino acids (2,4-diaminobutyric acid residues (Dab)б and Thr, Phe, and Leu) б and a fatty acid tail. Seven amino acids of polymyxin form a macrocycle (cycle [Dab4-Dab5-D-Phe6-Leu7-Dab8-Dab9-Thr10]), and three amino acids (Dab1-Thr2-Dab3) form a linear region connecting the macrocycle with the final methyl residue. octanoyl acid (MOA). The macrocycle is formed by an additional peptide bond between Thr10 and the γ-amino group of the Dab4 residue (Fig. 1).

Polymyxin B and colistin contain five free amino groups (as part of Dab residues) and, accordingly, five positive charges under physiological conditions. The only structural difference of PmE from PmB is the amino acid D-Leu at position 6 instead of D-Phe in polymyxin B.

When the use of β-lactams, aminoglycosides, or quinolones against extremely multidrug-resistant strains of gram-negative bacteria, including Pseudomonas aeruginosa, Acinetobacter baumannii, and Klebsiella pneumoniae, becomes ineffective, polymyxins remain the last treatment for these infections [3–5], [7]. The revival of polymyxins into clinical practice has stimulated further thorough investigations of their toxicity. During the last decades, the toxicity of PmB and colistin has been thoroughly studied (taking into account the chemical purity, homogeneity, dosing regimens, and other factors) and appeared to be not as high as reported earlier [8, 9]. Nevertheless, it still may substantially complicate therapy and even result in its stoppage. Thus, the development of less toxic polymyxin derivatives is a topical problem.

It is known that the nephrotoxic effect of polymyxins is determined by their accumulation in the epithelial cells of the proximal tubules of the kidney, where they persist for a long time, causing kidney damage. Their accumulation in increasing amounts in lysosomes leads to swelling and disruption of the latter and the release of polymyxins into the cytosol, where their nonspecific binding causes acute tubular necrosis [10–15]. The main factor in the accumulation of these antibiotics in the kidney is their interaction with megalin (formerly called the glycoprotein gp330), a giant receptor on the cell surface, which is the most abundant in the apical membrane of the proximal tubules of the kidney [11], [14, 16–21]. Megalin functions as an endocytic receptor that binds a wide spectrum of substances (including polymyxins) [19–21]. Thus, it may serve as a unique target to develop new polymyxin derivatives with minimized nephrotoxicity.

It was experimentally shown that the level of nephrotoxicity of polymyxin and its derivatives correlates with the peculiarities of their molecular structures [22–26] and, hence, with the peculiarities of their intermolecular interactions with megalin. In particular, derivatives of polymyxin, e.g., NAB7061 [23] (Fig. 2), which had only 3 positively charged residues located within the macrocycle of the molecule (residues Dab5, Dab8, and Dab9) had significantly lower nephrotoxicity and affinity for the kidney epithelium membrane than unmodified polymyxin. However, a comparative analysis of the peculiarities of their intermolecular interactions with megalin has remained behind the scope of the studies.

The revealing of a molecular picture of nephrotoxic action can make the search for non-toxic derivatives of polymyxins more targeted.

Megalin is a member of the low-density lipoprotein receptor (LDLR) gene family [27–29]. The LDLR family is a class of structurally homologous membrane receptors consisting of modular structures (domains) and represented in mammals by seven major glycoproteins: low-density lipoprotein receptor (LDLR); very low-density lipoprotein receptor (VLDLR); apolipoprotein E receptor 2 (ApoER2 or LRP8); multiple epidermal growth factor (MEGF7); LDLR-linked protein 1 (LRP1); LDLR-linked protein 1b (LRP1b) and LDLR-linked protein 2 (LRP1) or megalin [28–29]. All members of this receptor family have a modular structure (Fig. 3). In particular, they contain clusters of two or more cysteine-rich complement-type repeats, CR modules (indicated by red elongated rectangles in Fig. 3), which are the binding sites of the

most ligands of LDL-receptors. Megalin is the largest member of the family and contains 36 ligand-binding CR domains distributed in 4 clusters (Fig. 3). The smallest in this family is the LDLR, which contains only one cluster of 7 ligand-binding domains [28] (Fig. 4).

The binding site of the cationic ligands on the LDL-receptors contains a common structure motif, the so-called DXDXD motif (D – aspartic acid, X - any amino acid), which consists of three negatively charged aspartic acid residues coordinated by Са²⁺ ion and a hydrophobic residue (Fig. 5). As a rule, the cationic part of a ligand is represented by Lys residue.

Because of the high degree of homology of the amino acid sequences of CR domains, these binding sites have a remarkably similar folding of the polypeptide chain and spatial structure.

The Polymyxin binding site on the megalin is not yet determined experimentally; there are absent structural models for polymyxin interaction with megalin at the atomic level, as well. It is known, however, that polymyxin B is an effective competitive inhibitor of the RAP (Receptor-Associated Protein) protein binding to rat megalin [16] (megalin of humans and rats have a high degree of homology: their amino acid sequences are identical by 77%). But the competitive inhibition involves both the structural similarity between inhibitor and substrate and the same binding site at the receptor. The latter is known from the X-ray structure of the inhibitor RAP in complex with the human LDLR receptor [30]. Thus, there are reasons to suppose that the polymyxin binding site is the structural DXDXD motif of the ligand-binding CR domains of megalin as well, and the molecular fragments of polymyxins that interact with the receptor are their cationic Dab groups (analogs of lysine residues). It is known that the binding site of some ligands (including RAP) may involve structural DXDXD motifs from the adjacent repeats [30–35], i.e., there may be more than one center for binding of ligands. According to our estimates, in the expanded conformations of the polymyxin molecule, the distance between the nitrogen residues of Dab1 and Dab8 is approximately equal to the distance between Ca²⁺ ions in conserved binding sites of cationic ligands (2.12–2.15 nm] [30, 36]. Given this, we hypothesized that the higher affinity of polymyxin for the molecular target of megalin may be explained by the fact that, unlike NAB-derivatives, it interacts not with one, but simultaneously with two binding centers located on the adjacent CR repeats.

The study aimed to investigate the structural features of the polymyxins nephrotoxic action at an atomic level to promote the more purposeful development of the polymyxin’s derivatives with less nephrotoxicity.

The coordinates of the polymyxin B molecule were taken from the NMR experiment [37]. The structure of the NAB7061 derivative of polymyxin B1 was built based upon the structure of polymyxin B1. As a molecular model for megalin, we used the X-ray structure of two consecutive CR3-CR4 domains of the human LDLR receptor (PDB 2FCW [[30]]). The polymyxin molecule (in expanded conformation) was manually placed near the megalin molecule so that amino groups of Dab1 and Dab8 residues of polymyxin were located near the two DXDXD motifs of megalin at a distance not less than 8 Å from the corresponding Ca²⁺ atoms. The molecular systems were placed in a cubic periodic cell filled with TIP3P water molecules. The simulation cell was 1 nm larger than the molecular system. Before the simulations, the molecular systems were energy-minimized. After a short steepest descent minimization, the procedure continued by simulated annealing minimization. MD simulations were run in the NPT ensemble at 298 K and pH 7.4. Na⁺ and Cl ̄ counterions were added to neutralize the systems and to reach ion mass fraction 0.9% NaCl. The AMBER03 force field was used with a 1.05 nm force cutoff for dispersion interactions. Long-range electrostatic interactions were calculated by the Particle Mesh Ewald algorithm. Steps 1.25 and 2.5 fs were used for intramolecular and intermolecular forces, respectively. Simulations were run twice for each molecular system. For comparison, we separately performed MD simulations for polymyxin and NAB 7061 alone, as well. Trajectories were computed for 35 ns; the data were saved each 25 ps. Data were collected after 10 ns equilibration. For MD simulations and trajectories analysis, the molecular modeling program YASARA Structure was used [38].

During the MD simulations of the polymyxin-megalin system, polymyxin formed a stable supramolecular complex with megalin. In the most favorable configuration of the complex, polymyxin was bound to the two DXDXD binding motifs of the neighboring CR3 and CR4 domains of megalin simultaneously by its two cationic residues Dab8 and Dab1, respectively (Fig. 6). In literature, such a mode of binding was termed “bivalent” interaction [39].

The polymyxin-megalin complex was stable: the values of the binding energy showed only slight fluctuations near the average value that equals − 62.0 kcal/mol (Fig. 7). As can be seen from the dynamics of changes in the RMSD of C_α of the PmB backbone (Fig. 8), which are sensitive to conformational rearrangements of the molecule, PmB only slightly changes the conformation of its backbone in the interaction with megalin (the average value of RMSD C_α for it equals 0.97 Å), whereas the binding of PmB the decreases the mobility of its residues (especially, Dab1 and Dab8) (Fig. 9).

The bivalent binding occurs by the universal mechanism of binding of cationic ligands by CR repeats of LDLR family receptors: by the formation of a complex between two “critical” lysines of ligand (in our case, Dab residues) and a doublet of adjacent CR domains, where each lysine docks to a single CR domain [30, 39]. In the complex, the positively charged nitrogen atom of the residue Dab8 of PmB interacts with negatively charged residues Asp 108, Asp 110, and Asp 112 (Fig. 10 (a)), and the positively charged nitrogen atom of the Dab1 residue interacts with the negatively charged residues of Asp 147, Asp 149, and Asp 151 (Fig. 10 (b)). The binding occurs mainly due to the electrostatic interactions. Additional factors stabilizing the complex are hydrophobic interactions of aliphatic fragments of Dab8 and Dab1 residues with the conserved aromatic residues of Phe 105 and Trp 144 (Fig. 10 (a) and (b), respectively), which served as space fillers to create hydrophobic pockets between the molecules allowing other residues in the interface to make favorable interactions.

Similar to polymyxin B, the NAB7061 derivative binds to the DXDXD motif of the CR3 binding domain by its cationic residue Dab8 (Fig. 11). The binding occurs by the universal mechanism of binding of cationic ligands to LDLR receptors: a positively charged residue

Dab 8 of NAB7061 derivative interacts with negatively charged residues Asp108, Asp110, and Asp112 (Fig. 12 (a)).

Although like polymyxin, the NAB7061 derivative also binds by its cationic residue Dab5 to the second CR4 domain of megalin; its interactions occur with only two Asp residues (Fig. 12 (b)). Unlike polymyxin, the NAB7061 derivative has no stabilizing hydrophobic interactions with the structural domains CR3 and CR4 of megalin. Interactions of this derivative with megalin are not permanent, as manifested in larger values of root mean square fluctuations (RMSF) of NAB7061 residues (Fig. 13), and in large fluctuations of the energy of NAB7061 binding with megalin (Fig. 7). In total, the average value of binding energy, in this case, is lower (-45.4 kcal/mol), and the binding is weaker compared to polymyxin.

Our study shows that the structural prerequisite for a stronger interaction of PmB with megalin (compared to NAB7061) is the possibility to locate its charged amino groups (Dab8 and Dab1) at a certain distance apart. Such a possibility is absent for the polymyxin derivatives, which lack positive charges on the fatty acid tail; consequently, the stabilizing bivalent interactions with receptor are impossible in this case that, in turn, may be a reason for their lower affinity for the kidney epithelium membrane and, as a result, the lower nephrotoxicity compared to unmodified polymyxin. Thus, the weakening of polymyxins binding with megalin may become an effective preventive measure against polymyxin-induced nephrotoxicity.

New structural information was obtained on the interactions of polymyxins with megalin at the atomic level. By molecular dynamics simulations, the possibility was shown for polymyxin B to form a stable supramolecular bivalent complex with megalin, the target of its nephrotoxic action, in accord with the universal mechanism of binding the cationic ligands by ligand-binding CR repeats of LDLR-family receptors. The megalin interactions with polymyxin B were stronger than ones with its derivative having less nephrotoxicity. The possible structural prerequisites of these differences in nephrotoxic action of such derivatives compared to polymyxin were revealed. The weakening of polymyxins binding with megalin may become an effective preventive measure against polymyxin-induced nephrotoxicity.

Financing

This work was carried out with the financial support of the National Academy of Medical Sciences of Ukraine (grant No 0113U001516 “The study of the peculiarities of the interactions of the antimicrobial peptide of with targets”).

Conflict of interests

The authors declare that there is no conflict of interest.

Data Availability

1. In the article was used software Yasara https:// http://www.yasara.org/

2. Information about megalin’s structure available from PDB files https://www.rcsb.org/structure/1N7D

https://www.rcsb.org/structure/2FCW

3. Information about polymyxin’s structure available from

https://pubchem.ncbi.nlm.nih.gov/compound/49800004#section=2D-Structure

Paterson DL, Bonomo RA (2019) Multidrug-resistant gram-negative pathogens: The urgent need for ‘old’polymyxins. Polymyxin Antibiotics: From Laboratory Bench to Bedside 9–13. https://doi.org/10.1007/978-3-030-16373-0_2
Nang SC, Azad MA, Velkov T, et al (2021) Rescuing the last-line polymyxins: achievements and challenges. Pharmacological Reviews 73:679–728. https://doi.org/10.1124/pharmrev.120.000020
Jeannot K, Bolard A, Plesiat P (2017) Resistance to polymyxins in Gram-negative organisms. International journal of antimicrobial agents 49:526–535. https://doi.org/10.1016/j.ijantimicag.2016.11.029
Lisnyak Y (2015) Molecular targets of nephrotoxic action of polymyxins. Annals of Mechnikov’s Institute 8–24
Li J (2019) Reviving polymyxins: achievements, lessons and the road ahead. Polymyxin Antibiotics: From Laboratory Bench to Bedside 1–8. https://doi.org/10.1007/978-3-030-16373-0_1
Martin NI, Hu H, Moake MM, et al (2003) Isolation, Structural Characterization, and Properties of Mattacin (Polymyxin M), a Cyclic Peptide Antibiotic Produced byPaenibacillus kobensis M. Journal of Biological Chemistry 278:13124–13132. https://doi.org/10.1074/jbc.M212364200
Kwa AL, Tam VH, Falagas ME (2008) Polymyxins: a review of the current status including recent developments. Ann Acad Med Singapore 37:870–883
Falagas ME, Kasiakou SK (2006) Toxicity of polymyxins: a systematic review of the evidence from old and recent studies. Critical care 10:1–13. https://doi.org/10.1186/cc3995
Ouderkirk JP, Nord JA, Turett GS, Kislak JW (2003) Polymyxin B nephrotoxicity and efficacy against nosocomial infections caused by multiresistant gram-negative bacteria. Antimicrobial Agents and Chemotherapy 47:2659–2662. https://doi.org/10.1128/AAC.47.8.2659-2662.2003
Abdelraouf K, Chang KT, Yin T, et al (2014) Uptake of polymyxin B into renal cells. Antimicrobial Agents and Chemotherapy 58:4200–4202. https://doi.org/10.1128/AAC.02557-14
Suzuki T, Yamaguchi H, Ogura J, et al (2013) Megalin contributes to kidney accumulation and nephrotoxicity of colistin. Antimicrobial Agents and Chemotherapy 57:6319–6324. https://doi.org/10.1128/AAC.00254-13
Berg J, Spilker C, Lewis S (1996) Effects of polymyxin B on mammalian urinary bladder. The Journal of membrane biology 154:119–130. https://doi.org/10.1007/s002329900137
Lewis JR, Lewis SA (2004) Colistin interactions with the mammalian urothelium. American Journal of Physiology-Cell Physiology 286:C913–C922. https://doi.org/10.1152/ajpcell.00437.2003
Nielsen R, Christensen EI (2010) Proteinuria and events beyond the slit. Pediatric nephrology 25:813–822. https://doi.org/10.1007/s00467-009-1381-9
Christensen EI, Gburek J (2004) Protein reabsorption in renal proximal tubule—function and dysfunction in kidney pathophysiology. Pediatric nephrology 19:714–721. https://doi.org/10.1007/s00467-004-1494-0
Moestrup SK, Cui S, Vorum H, et al (1995) Evidence that epithelial glycoprotein 330/megalin mediates uptake of polybasic drugs. Journal of Clinical Investigation 96:1404–1413. https://doi.org/10.1172/JCI118176
Farquhar M (1996) Molecular analysis of the pathological autoimmune antigens of Heymann nephritis. The American journal of pathology 148:1331
Farquhar MG (1995) The unfolding story of megalin (gp330): now recognized as a drug receptor. The Journal of clinical investigation 96:1184–1184. https://doi.org/10.1172/JCI118149
Schmitz C, Hilpert J, Jacobsen C, et al (2002) Megalin deficiency offers protection from renal aminoglycoside accumulation. Journal of Biological Chemistry 277:618–622. https://doi.org/10.1074/jbc.M109959200
Nagai J, Takano M (2004) Molecular aspects of renal handling of aminoglycosides and strategies for preventing the nephrotoxicity. Drug metabolism and pharmacokinetics 19:159–170. https://doi.org/10.2133/dmpk.19.159
Christensen EI, Verroust PJ, Nielsen R (2009) Receptor-mediated endocytosis in renal proximal tubule. Pflügers Archiv-European Journal of Physiology 458:1039–1048. https://doi.org/10.1007/s00424-009-0685-8
MINGEOT-LECLERCQ M-P, Tulkens PM (1999) Aminoglycosides : Nephrotoxicity. Antimicrob Agents Chemother 43:1003–1012
Vaara M, Fox J, Loidl G, et al (2008) Novel polymyxin derivatives carrying only three positive charges are effective antibacterial agents. Antimicrobial agents and chemotherapy 52:3229–3236. https://doi.org/10.1128/AAC.00405-08
Vaara M, Siikanen O, Apajalahti J, et al (2010) A novel polymyxin derivative that lacks the fatty acid tail and carries only three positive charges has strong synergism with agents excluded by the intact outer membrane. Antimicrobial agents and chemotherapy 54:3341–3346. https://doi.org/10.1128/AAC.01439-09
Vaara M, Vaara T (2010) Structure–activity studies on novel polymyxin derivatives that carry only three positive charges. Peptides 31:2318–2321. https://doi.org/10.1016/j.peptides.2010.09.010
Ali FA, Cao G, Poudyal A, et al (2009) Pharmacokinetics of novel antimicrobial cationic peptides NAB 7061 and NAB 739 in rats following intravenous administration. Journal of antimicrobial chemotherapy 64:1067–1070. https://doi.org/10.1093/jac/dkp331
Saito A, Pietromonaco S, Loo AK-C, Farquhar MG (1994) Complete cloning and sequencing of rat gp330/" megalin," a distinctive member of the low density lipoprotein receptor gene family. Proceedings of the National Academy of Sciences 91:9725–9729. https://doi.org/10.1073/pnas.91.21.9725
Lillis AP, Van Duyn LB, Murphy-Ullrich JE, Strickland DK (2008) LDL receptor-related protein 1: unique tissue-specific functions revealed by selective gene knockout studies. Physiological reviews 88:887–918. https://doi.org/10.1152/physrev.00033.2007
Jeon H, Blacklow SC (2005) Structure and physiologic function of the low-density lipoprotein receptor. Annual review of biochemistry 74:535. https://doi.org/10.1146/annurev.biochem.74.082803.133354
Fisher C, Beglova N, Blacklow SC (2006) Structure of an LDLR-RAP complex reveals a general mode for ligand recognition by lipoprotein receptors. Molecular cell 22:277–283. https://doi.org/10.1016/j.molcel.2006.02.021
ANDERSEN OM, PETERSEN HH, JACOBSEN C, et al (2001) Analysis of a two-domain binding site for the urokinase-type plasminogen activator–plasminogen activator inhibitor-1 complex in low-density-lipoprotein-receptor-related protein. Biochemical Journal 357:289–296. https://doi.org/10.1042/bj3570289
Andersen OM, Vorum H, Honoré B, Thøgersen HC (2003) Ca2+ binding to complement-type repeat domains 5 and 6 from the low-density lipoprotein receptor-related protein. BMC biochemistry 4:1–7
Lazic A, Dolmer K, Strickland DK, Gettins PG (2006) Dissection of RAP-LRP interactions: binding of RAP and RAP fragments to complement-like repeats 7 and 8 from ligand binding cluster II of LRP. Archives of biochemistry and biophysics 450:167–175. https://doi.org/10.1016/j.abb.2006.04.007
Russell DW, Brown MS, Goldstein JL (1989) Different combinations of cysteine-rich repeats mediate binding of low density lipoprotein receptor to two different proteins. Journal of Biological Chemistry 264:21682–21688. https://doi.org/10.1016/S0021-9258(20)88240-4
Dolmer K, Gettins PG (2006) Three complement-like repeats compose the complete α2-macroglobulin binding site in the second ligand binding cluster of the low density lipoprotein receptor-related protein. Journal of Biological Chemistry 281:34189–34196. https://doi.org/10.1074/jbc.M604389200
Jensen GA, Andersen OM, Bonvin AM, et al (2006) Binding site structure of one LRP–RAP complex: implications for a common ligand–receptor binding motif. Journal of molecular biology 362:700–716. https://doi.org/10.1016/j.jmb.2006.07.013
Pristovšek P, Kidric J (1999) Solution structure of polymyxins B and E and effect of binding to lipopolysaccharide: an NMR and molecular modeling study. Journal of medicinal chemistry 42:4604–4613. https://doi.org/10.1021/jm991031b
Krieger E, Vriend G (2015) New ways to boost molecular dynamics simulations. Journal of computational chemistry 36:996–1007. https://doi.org/10.1002/jcc.23899
Marakasova E, Olivares P, Karnaukhova E, et al (2021) Molecular chaperone RAP interacts with LRP1 in a dynamic bivalent mode and enhances folding of ligand-binding regions of other LDLR family receptors. Journal of Biological Chemistry 297:. https://doi.org/10.1016/j.jbc.2021.100842

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Interactions of antimicrobial peptides with the targets of their nephrotoxic action. Molecular dynamics simulations

Status:

Version 1

Abstract

Objectives

Materials and methods

Results

Conclusions

Figures

1. Introduction

2. Materials And Methods

3. Results And Discussion

4. Conclusions

Declarations

References

Additional Declarations

Status:

Version 1