Structural and Functional Consequences Induced by Post-Translational Modifications in α-Defensins
HNP-1 is an antimicrobial peptide that undergoes proteolytic cleavage to become a mature peptide. This process represents the mechanism commonly used by the cells to obtain a fully active antimicrobial peptide. In addition, it has been recently described that HNP-1 is recognized as substrate by the arginine-specific ADP-ribosyltransferase-1. Arginine-specific mono-ADP-ribosylation is an enzyme-catalyzed post-translational modification in which NAD+ serves as donor of the ADP-ribose moiety, which is transferred to the guanidino group of arginines in target proteins. While the arginine carries one positive charge, the ADP-ribose is negatively charged at the phosphate moieties at physiological pH. Therefore, the attachment of one or more ADP-ribose units results in a marked change of cationicity. ADP-ribosylation of HNP-1 drastically reduces its cytotoxic and antibacterial activities. While the chemotactic activity of HNP-1 remains unaltered, its ability to induce interleukin-8 production is enhanced. The arginine 14 of HNP-1 modified by the ADP-ribose is in some cases processed into ornithine, perhaps representing a different modality in the regulation of HNP-1 activities.
Antimicrobial host defense peptides (AMPs) are widely distributed throughout all kingdoms of life and represent one of the most ancient host defense factors. All of them display microbicidal activity against a wide spectrum of Gram-negative and Gram-positive bacteria, fungi, and viruses . The two main classes of AMPs produced by humans are represented by defensins and by the single cathelicidin LL-37, a peptide of 37 amino acids beginning with two leucin residues (L-L). Moreover, humans produce basic histatins, which are released in the oral cavity , dermicidins from the skin, and granulysin, which are released by proteolytic processing in sweat, and in leucocytes, respectively [3, 4]. Defensins are cationic multifunctional arginine-rich peptides with molecular masses ranging from 3.5 to 6 kDa, characterized by three intramolecular disulfide bonds. Based on the connectivity of the six conserved cysteine residues and sequence homology, human defensins are grouped into α- and β-defensins. Human neutrophil peptides, HNP-1 to 4, are the major component of the azurophilic granules of neutrophils while the β-defensins are mainly expressed in epithelial cells of various tissues and organs [5–7]. HNP-1 is synthesized as prepropeptide, which is processed via different sequential proteolytical cleavage steps to the final functional peptide of 30 amino-acid. Cleavage is by far the most studied post-translational modification for AMPs and is also the topic of several excellent reviews [1, 8, 9]. In this review we summarize the current knowledge regarding the influence of ADP-ribosylation on both structure and function of peptides belonging to the innate immune system and we describe this post-translational modification at the molecular level. Moreover, we aim to describe the strong impact that this modification exerts on the ability of HNP-1 to kill bacteria but also on the cytotoxicity versus eukaryotic cells. ADP-ribosylated-HNP-1 are present in the airways of patients with inflammatory-based diseases , therefore this novel discovery has the potential to open up new avenues of investigation with regard to pulmonary inflammatory diseases, which can be translated into tools for therapeutic applications.
2. Biochemical Characterization of Mono-ADP-Ribosylation
Mono ADP-ribosylation is a reversible enzymatic post-translational modification, which alters the chemical and functional properties of the modified proteins . In this reaction mono-ADP-ribosyltransferases (ARTs) catalyze the covalent attachment of an ADP-ribose unit from NAD+ to specific amino-acid residues with the simultaneous release of nicotinamide (Figure 1). The ART activity has been initially described in prokaryotes as one of the main mechanism of bacterial pathogenesis [12–14]. More recently five enzymes (ART1, ART2, ART3, ART4, and ART5) related to mono-ADP-ribosylating toxins have been identified in eukaryotes. Among them only ART1, ART2, and ART5 contain the R-S-EXE triad signature typical of a subfamily of arginine-specific ARTs. These enzymes selectively recognize and modify Arg as amino-acid acceptor. This signature is absent in the other two mammalian transfer enzymes ART3 and ART4, whose activity has not been characterized yet. ARTs are glycosylphosphatidylinositol (GPI) linked to the membrane with the catalytic site at the outer surface of cells with the exception of ART5, which represents a secretory form of ART [15, 16]. ADP-ribosylation of arginine residues can influence the function of target proteins by several mechanisms. The ADP-ribosylation reaction introduces a large and bulky group on the target protein, which is fivefold bigger (540 Da) than the average size of the amino-acid. Therefore, the presence of a single or multiple ADP-ribose modifications can considerably influence the structure and the size of antimicrobial peptides with an MW ranging between 3 and 6 kDa. The addition of ADP-ribose can induce a conformational change in the target protein, which often results in an activating effect or in the creation of a new docking site for ADP-ribose binding domains in other proteins . At the same time the presence of ADP-ribose can also impede the sterical interaction with molecular partners. In addition, at neutral pH while the arginine carries one positive charge, the ADP-ribose is negatively charged at the phosphate moieties. Therefore, the transfer of one or more negatively charged ADP-ribose moieties has the potential to considerably reduce the positive charge of the modified Arg and overall the cationicity of antimicrobials.
3. Physiological Implications of Post-Translational Modifications of HNP-1
The localization of ART1 on the surface of ciliated and intermediate epithelial cells lining the airways  led to the proposal that potential substrates for ART1-specific ADP-ribosylation could be present in the extracellular milieu. As shown in Table 1, α- and β-defensins are cationic peptides with a high content of Arg residues which are the amino-acid target for the ART1 catalyzed ADP-ribosylation. Azurophilic granules in inflammatory cells mainly store HNP-1 molecules. Since their main function is to defend the lung against pathogenic microorganisms, they are secreted at high concentration at the site of inflammation in the airways by neutrophils ; therefore, they could represent an ART1 substrate. In aqueous solution, HNP-1 is arranged in a dimeric form, where each monomer presents a positive net charge equal to +3 (Figure 2), conferred by four arginine residues (Arg5, Arg14, Arg15, and Arg24) and a negative glutamic acid (Glu13). The last three arginine residues of HNP-1 are directly involved in membrane perturbation while Arg5 is involved in the formation of a salt bridge with Glu13. The positively charged arginine residues provide an effective means of attracting dimeric HNP-1 peptide to target membranes facilitating its interaction with negatively charged surfaces, such as phosphatidyl glycerol or cardiolipin head groups (Figure 3). Previous studies demonstrate that the replacement of Arg residues by neutral or oppositely charged amino acids, and modification of the amino-acid sequence leads to dramatic reduction or loss of activity in bacterial killing not only in HNP-1 peptide but also in other cationic antimicrobial peptides [20–22]. It has been shown (Figure 4) that HNP-1 is ADP-ribosylated when assayed in vitro in the presence of ART1 and NAD+ . The primary site of modification is Arg14, but also Arg24 is shown to be ADP-ribosylated after a longer incubation time. Interestingly, Arg14 and Arg24, as described previously, are directly involved in membrane interaction. Arg5, which is involved in a salt bridge, is not targeted by ART1. Both forms were isolated from bronchoalveolar lavage (BAL) of volunteer with history of smoking . The ADP-ribosylation of selected Arg might well correlate with recent discoveries, which show that antibacterial activity is strictly dependent by cationicity  and that only selective Arg support this activity . While ART2 slightly modifies HNP-1, ART5 is completely inactive on HNP-1  supporting the evidence of a specific modification since the mouse ART1 ortholog and ART1 GPI-linked to the skeletal muscle membrane modifies HNP-1 with the same intensity . The ADP-ribosylated-HNP-1 exhibited a strong reduction of antimicrobial activity compared to the unmodified peptide. Moreover, while the chemotactic activity of the modified and unmodified HNP-1 resulted unaltered, the ability of HNP-1 to release interleukin-8 (IL-8) from A549 cells was enhanced by the presence of ADP-ribose linked to Arg14 . This is in line with the idea that the ADP-ribosylated-HNP-1 is a completely different molecule, which possesses unique biological characteristics that influence the ability to recruit neutrophils in the site of inflammation through the release of IL-8 from epithelial cells. This latter is an important point since mono- and di-ADP-ribosylated forms of HNP-1 were found in BAL of patients with pulmonary diseases such as idiopathic pulmonary fibrosis and asthma. Though the neutrophils in these diseases were recruited in high number, and HNP-1 and other isoforms were found in their granules, modified HNP-1 were not present. This suggests that HNP-1 have to be released from neutrophils to be ADP-ribosylated by ART1, which is expressed at the outer surface of epithelial cells. High concentrations of HNP-1 exert cytotoxic effects towards epithelial cells. In airways of patients with pulmonary diseases the concentration of HNP-1 produced by epithelial cells and infiltrating neutrophils, is high. Therefore, these cytotoxic effects may well be physiologically relevant at sites of inflammation and could result in a high local concentration of HNP-1, thus leading to cytotoxicity. Since electrostatic interactions play a key role in the interaction with biological membranes, it is conceivable that modifications like ADP-ribosylation, which link a negatively charged ADP-ribose moiety to the Arg14 and Arg24, may reduce the interaction with biological membranes and consequently the cytotoxicity that HNP-1 exerts when its concentration is high. Interestingly, the cytotoxicity of HNP-1 towards epithelial cells is drastically reduced when the peptide is ADP-ribosylated .
4. Ornithinylation and Phosphorylationof HNP-1
Post-translational modifications of HNP-1 are not restricted to the ADP-ribosylation. As shown in Table 1 defensins are also rich in Tyr, which are the sites, phosphorylated by tyrosine kinases. Using a novel phosphoproteomic approach to identify and validate phosphotyrosine sites it has been shown that the Tyr21 (YGTCIY*QGR) in HNP-1 and HNP-3 is phosphorylated in cancer cell lines . Because these studies represent an early step in the characterization of the tyrosine kinase signaling in lung cancer, the role that this modification plays in the defensin activities has to be established. The phosphoryl group is only 80 Da with respect to ADP-ribose, but it is interesting to note that in model peptide substrate phosphorylation affects ADP-ribosylation in the nearby arginyl residues . This possible link between phosphorylation and ADP-ribosylation could be relevant in the context of HNP-1 in which the Arg14 and the Arg24 are close to the Tyr21. Another modification that seems to occur in vivo is the generation of the ornithyl-HNP-1 from ADP-ribosylated-HNP-1 . The guanidino carbon of the ADP-ribosylated-arginine can undergo nonenzymatic hydrolysis resulting in the replacement of the ADP-ribosylarginine by the noncoded amino-acid ornithine (Figure 5). This modification is specific, since it targets only the ADP-ribosylated-arginine at position 14. Moreover, it prevents the modified Arg to serve as substrate for subsequent ADP-ribosylation since ornithine is not recognized as substrate by ART1. BAL of patients with idiopathic pulmonary fibrosis and asthma other than mono- and di-ADP-ribosylated-HNP-1 contained HNP-1 ADP-ribosylated at Arg24 with ornithine replacing Arg in position 14 suggesting a relevance of this conversion for the innate immunity in the airway . Interestingly, when 3 of the 4 Arg (not Arg5) were replaced by ornithine the bactericidal activity was decreased suggesting that the conversion of arginine with ornithine can be important in altering HNP-1 activities . Moreover, the chemical degradation of the ADP-ribosylated-arginine in ornithine creates a secondary post-translational modification that could potentially amplify the ability of the cell to regulate the biological activities of the HNP-1 and consequently the immune response in the airways.
5. Processing and Reversibility of the ADP-Ribose Moiety
Like many post-translational modifications with regulatory functions, mono-ADP-ribosylation of Arg is a potentially reversible process that could lead to a complete or partial removal of the ADP-ribose unit. A full removal of the ADP-ribose unit from the modified amino-acid is catalyzed by a specific mammal ADP-ribosylarginine hydrolase-1 (ARH1) , which regenerates free Arg for a new ADP-ribosylation cycle (Figure 5). Since ARH1 is localized in the intracellular compartment and modification of HNP-1 is catalyzed by the extracellular ART1, the restoration of unmodified HNP-1 due to the action of ARH1 is unlikely. In fact at least one of the two proteins should translocate across the membrane. Partial reversal of the ADP-ribose moiety may result from the action of extracellular phosphodiesterases that cleave adenosine-5′-monophopshate from ADP-ribose yielding a ribosyl moiety attached to the target Arg . This modification precludes a new ADP-ribosylation of the target Arg since it leaves attached a phosphoribose molecule with a negative charge (Figure 5). Phosphodiesterases are extracellularly expressed and could potentially cleave the adenosine-5′-monophopshate from the ADP-ribose of the ADP-ribosylated-HNP-1. These two latter examples of processing have not been described for HNP-1 so far.
6. Future Perspectives of ADP-Ribosylation of Antimicrobials
The molecular and functional comprehension of the biological activities of AMPs and how these activities are regulated in vivo could represent a crucial step in understanding deeply how innate immunity is modulated, since AMPs are an important defence mechanism against invading microorganisms. The biochemical characterization of ADP-ribosylated antimicrobial peptides, including the interaction with both their molecular targets and biological membranes, is important to reach this goal. The study of the functional role of arginine-ADP-ribosylation of antimicrobial peptides could represent the starting point for finding new therapeutic opportunities. Both NMR and crystallography are by far the most potent tools in investigating the protein structure properties. Furthermore, improved structure-function prediction algorithms are another opportunity to develop synthetic peptide that mimics modified portions or inhibitors of this reaction for therapeutic use. Monitoring the ADP ribosylated peptide should be an interesting strategy to identify biological targets even though, some AMPs are not directly bactericidal, but exert their effect by immunomodulation . Recent improved tools that include antibodies specific for ADP-ribosylarginine  or the possibility to generate site-specific ADP-ribosylated peptides [33, 34] will be useful to study and eventually treat pulmonary inflammation, other lung disorders as well as toxin-mediated diseases. The replacement with ornithine of the Arg linked with ADP-ribose could represent an alternative pathway for the regulation of the biological activities of HNP-1 through post-translational modification. It could be of interest to evaluate the effects of these modifications on HNP-1 functions. Since AMPs are rich in Arg but also contain amino acids potential substrate for phosphorylation (Table 1), it could be also of interest to evaluate whether the regulation of biological activities of antimicrobials by ADP-ribosylation or phosphorylation is restricted to HNP-1 or is a more generic mechanism.
In this review we have described how the ADP-ribosylation of specific Arg residues of HNP-1 affects its biological properties. The biological activities of HNP-1 are strictly dependent on the elevated cationicity due to the presence of Arg residues. Mono ADP-ribosylation is one example how post-translational modifications can affect the functions of peptides such as the α-defensin. In the era of microarray technology it has to be recognized that the analysis of the expression and regulation of single genes alone is not sufficient to understand the functional roles of proteins and peptides within normal or pathological situations. The interplay between ADP-ribosylation and phosphorylation, beyond the analysis of the functional effects that each single reaction can exerts on peptides belonging to the innate immunity system, should be also taken into consideration.
The authors are grateful to Giorgio Corsi for his precious artwork, to Elisabetta Soldaini for helpful discussion, valuable advice, and assistance. Novartis Vaccines and Diagnostics financially supported the paper.
R. I. Lehrer, A. K. Lichtenstein, and T. Ganz, “Defensins: antimicrobial and cytotoxic peptides of mammalian cells,” Annual Review of Immunology, vol. 11, pp. 105–128, 1993.View at: Google Scholar
F. G. Oppenheim, T. Xu, F. M. McMillian et al., “Histatins, a novel family of histidine-rich proteins in human parotid secretion. Isolation, characterization, primary structure, and fungistatic effects on Candida albicans,” Journal of Biological Chemistry, vol. 263, no. 16, pp. 7472–7477, 1988.View at: Google Scholar
B. Schittek, R. Hipfel, B. Sauer et al., “Dermcidin: a novel human antibiotic peptide secreted by sweat glands,” Nature Immunology, vol. 2, no. 12, pp. 1133–1137, 2001.View at: Publisher Site | Google Scholar
S. V. Peña, D. A. Hanson, B. A. Carr, T. J. Goralski, and A. M. Krensky, “Processing, subcellular localization, and function of 519 (granulysin), a human late T cell activation molecule with homology to small, lytic, granule proteins,” Journal of Immunology, vol. 158, no. 6, pp. 2680–2688, 1997.View at: Google Scholar
K. W. Bensch, M. Raida, H. J. Magert, P. Schulz-Knappe, and W. G. Forssmann, “hBD-1: a novel β-defensin from human plasma,” FEBS Letters, vol. 368, no. 2, pp. 331–335, 1995.View at: Publisher Site | Google Scholar
J. Harder, J. Bartels, E. Christophers, and J. M. Schröder, “Isolation and characterization of human β-defensin-3, a novel human inducible peptide antibiotic,” Journal of Biological Chemistry, vol. 276, no. 8, pp. 5707–5713, 2001.View at: Publisher Site | Google Scholar
J. Harder, J. Bartels, E. Christophers, and J. M. Schroder, “A peptide antibiotic from human skin,” Nature, vol. 387, no. 6636, p. 861, 1997.View at: Publisher Site | Google Scholar
L. Eckmann, “Defence molecules in intestinal innate immunity against bacterial infections,” Current Opinion in Gastroenterology, vol. 21, no. 2, pp. 147–151, 2005.View at: Publisher Site | Google Scholar
R. I. Lehrer and T. Ganz, “Cathelicidins: a family of endogenous antimicrobial peptides,” Current Opinion in Hematology, vol. 9, no. 1, pp. 18–22, 2002.View at: Publisher Site | Google Scholar
G. Paone, L. A. Stevens, R. L. Levine et al., “ADP-ribosyltransferase-specific modification of human neutrophil peptide-1,” Journal of Biological Chemistry, vol. 281, no. 25, pp. 17054–17060, 2006.View at: Publisher Site | Google Scholar
K. Ueda and O. Hayaishi, “ADP-ribosylation,” Annual Review of Biochemistry, vol. 54, pp. 73–100, 1985.View at: Google Scholar
K. A. Wreggett, “Bacterial toxins and the role of ADP-ribosylation,” Journal of Receptor Research, vol. 6, no. 2, pp. 95–126, 1986.View at: Google Scholar
S. S. Lahiri, “Bacterial toxins—an overview,” Journal of Natural Toxins, vol. 9, no. 4, pp. 381–408, 2000.View at: Google Scholar
M. J. Pallen, A. C. Lam, N. J. Loman, and A. McBride, “An abundance of bacterial ADP-ribosyltransferases—Implications for the origin of exotoxins and their human homologues,” Trends in Microbiology, vol. 9, no. 7, pp. 302–308, 2001.View at: Publisher Site | Google Scholar
I. J. Okazaki, H. J. Kim, and J. Moss, “Cloning and characterization of a novel membrane-associated lymphocyte NAD:arginine ADP-ribosyltransferase,” Journal of Biological Chemistry, vol. 271, no. 36, pp. 22052–22057, 1996.View at: Publisher Site | Google Scholar
I. J. Okazaki and J. Moss, “Characterization of glycosylphosphatidylinositiol-anchored, secreted, and intracellular vertebrate mono-ADP-ribosyltransferases,” Annual Review of Nutrition, vol. 19, pp. 485–509, 1999.View at: Publisher Site | Google Scholar
A. G. Ladurner, “Inactivating chromosomes: a macro domain that minimizes transcription,” Molecular Cell, vol. 12, no. 1, pp. 1–2, 2003.View at: Publisher Site | Google Scholar
E. Balducci, K. Horiba, J. Usuki, M. Park, V. J. Ferrans, and J. Moss, “Selective expression of RT6 superfamily in human bronchial epithelial cells,” American Journal of Respiratory Cell and Molecular Biology, vol. 21, no. 3, pp. 337–346, 1999.View at: Google Scholar
J. Ashitani, H. Mukae, M. Nakazato et al., “Elevated concentrations of defensins in bronchoalveolar lavage fluid in diffuse panbronchiolitis,” European Respiratory Journal, vol. 11, no. 1, pp. 104–111, 1998.View at: Publisher Site | Google Scholar
H. Steiner, D. Hultmark, and A. Engstrom, “Sequence and specificity of two antibacterial proteins involved in insect immunity,” Nature, vol. 292, no. 5820, pp. 246–248, 1981.View at: Google Scholar
W. Meyer-Glauner, E. Bernard, D. Armstrong, and B. Merrifield, “The antifungal activity of carrier peptides, L-arginyl-X-L-phenylalanine, containing amino acid antagonists or atypical non-biogenic D-amino acids in the central position,” Zentralblatt fur Bakteriologie Mikrobiologie und Hygiene A, vol. 252, no. 2, pp. 274–278, 1982.View at: Google Scholar
Z. Oren, J. Hong, and Y. Shai, “A repertoire of novel antibacterial diastereomeric peptides with selective cytolytic activity,” Journal of Biological Chemistry, vol. 272, no. 23, pp. 14643–14649, 1997.View at: Publisher Site | Google Scholar
G. Paone, A. Wada, L. A. Stevens et al., “ADP ribosylation of human neutrophil peptide-1 regulates its biological properties,” Proceedings of the National Academy of Sciences of the United States of America, vol. 99, no. 12, pp. 8231–8235, 2002.View at: Publisher Site | Google Scholar
G. Zou, E. De Leeuw, C. Li et al., “Toward understanding the cationicity of defensins: arg and Lys versus their noncoded analogs,” Journal of Biological Chemistry, vol. 282, no. 27, pp. 19653–19665, 2007.View at: Publisher Site | Google Scholar
E. de Leeuw, M. Rajabi, G. Zou, M. Pazgier, and W. Lu, “Selective arginines are important for the antibacterial activity and host cell interaction of human α-defensin 5,” FEBS Letters, vol. 583, no. 15, pp. 2507–2512, 2009.View at: Publisher Site | Google Scholar
K. Rikova, A. Guo, Q. Zeng et al., “Global survey of phosphotyrosine signaling identifies oncogenic kinases in lung cancer,” Cell, vol. 131, no. 6, pp. 1190–1203, 2007.View at: Publisher Site | Google Scholar
S. V. Kharadia and D. J. Graves, “Relationship of phosphorylation and ADP-ribosylation using a synthetic peptide as a model substrate,” Journal of Biological Chemistry, vol. 262, no. 36, pp. 17379–17383, 1987.View at: Google Scholar
L. A. Stevens, R. L. Levine, B. R. Gochuico, and J. Moss, “ADP-ribosylation of human defensin HNP-1 results in the replacement of the modified arginine with the noncoded amino acid ornithine,” Proceedings of the National Academy of Sciences of the United States of America, vol. 106, no. 47, pp. 19796–19800, 2009.View at: Publisher Site | Google Scholar
J. Moss, S. C. Tsai, R. Adamik, H. C. Chen, and S. J. Stanley, “Purification and characterization of ADP-ribosylarginine hydrolase from Turkey erythrocytes,” Biochemistry, vol. 27, no. 15, pp. 5819–5823, 1988.View at: Google Scholar
A. Zolkiewska and J. Moss, “Processing of ADP-ribosylated integrin α7 in skeletal muscle myotubes,” Journal of Biological Chemistry, vol. 270, no. 16, pp. 9227–9233, 1995.View at: Publisher Site | Google Scholar
D. M. E. Bowdish, D. J. Davidson, and R. E. W. Hancock, “A re-evaluation of the role of host defence peptides in mammalian immunity,” Current Protein and Peptide Science, vol. 6, no. 1, pp. 35–51, 2005.View at: Publisher Site | Google Scholar
H. Osago, M. Terashima, N. Hara, K. Yamada, and M. Tsuchiya, “A new detection method for arginine-specific ADP-ribosylation of protein—a combinational use of anti-ADP-ribosylarginine antibody and ADP-ribosylarginine hydrolase,” Journal of Biochemical and Biophysical Methods, vol. 70, no. 6, pp. 1014–1019, 2008.View at: Publisher Site | Google Scholar
P. M. Moyle and T. W. Muir, “Method for the synthesis of mono-ADP-ribose conjugated peptides,” Journal of the American Chemical Society, vol. 132, no. 45, pp. 15878–15880, 2010.View at: Publisher Site | Google Scholar
G. J. Van Der Heden Van Noort, M. G. Van Der Horst, H. S. Overkleeft, G. A. Van Der Marel, and D. V. Filippov, “Synthesis of mono-ADP-ribosylated oligopeptides using ribosylated amino acid building blocks,” Journal of the American Chemical Society, vol. 132, no. 14, pp. 5236–5240, 2010.View at: Publisher Site | Google Scholar