Abstract
We generated an evolutionary computer program that generates complementary peptide (C-pep) sequences, with the potential to interact with a target peptide, by comparing several physico-chemical parameters of each pair of the complementary peptides being analyzed. We generated C-peps to target several molecules. About 30% of synthesized C-peps interfered with the function of their targets. C5a stimulates generation of TNFα and other inflammatory cytokines. Inhibition of C5a should be effective against sepsis, which impairs the status of cancer-bearing patients. One of the inhibitory C-peps of C5a, termed AcPepA, was effective in Cynomolgus monkeys intravenously infused with a lethal dose of bacterial LPS (4 mg/kg) destined to die. The monkeys were rescued by intravenous administration of 2 mg/kg/h of AcPepA. The excellent therapeutic effect of AcPepA is likely to be due to restriction of high mobility group box 1 (HMGB1) surge induced by the effect of C5a on C5L2, which is the second C5a receptor, since the released HMGB1 has the capacity to stimulate TLR4 as an endogeneous ligand resulting in further activation of inflammatory cells to release inflammatory cytokines forming a positive feedback circuit of inflammation.
After proposal of the possible role of antisense peptides for molecular interaction among proteins by Blalock et al. (1) in 1984, the theory was reviewed later (2, 3). Many examples of sense–peptide and antisense–peptide relationships have been found between receptors and their protein ligands (4-8). We speculated that such interactions between sense and antisense peptides should play a role in formation of the tertiary structure of proteins. We developed a novel computer program named ANTIS to find antisense peptide sequences between proteins to be compared (9). By analyzing intramolecular antisense peptides within a single protein molecule, we found that there are an appreciable number of sense and antisense peptide pairs within a protein molecule and we designated these as antisense homology boxes (AHB) (9). Using ANTIS, we analyzed sense and antisense peptide relationships in the endothelin receptor (ETR) molecule and between endothelin and ETR. One of the AHB peptides of ETR, named ETR-P1/fl, had the capacity to interfere with the function of ETR (10). We expected that it would be possible to generate candidates of complementary peptide reactive to a target amino acid sequence based upon the sense-antisense amino acid relationship. We generated an evolutionary computer program that runs on any PC, and generates complementary peptide (C-pep) sequences, with the potential to interact with a target peptide, by comparing several physico-chemical parameters of each pair of the complementary peptides being analyzed (11). With the program named MIMETIC, we generated complementary peptides (C-peps) to HIV-reverse transcriptase (11, 12), procarboxypeptidase R, thrombomodulin (13), and C5a anaphylatoxin (14) as listed in Table I. About 30% of the synthesized peptides interfered with the function of their target molecules. Out of 19 complementary peptides (C-peps) targeted to C5a anaphylatoxin, 7 exhibited an inhibitory effect.
C5a is a 74 amino acid peptide released from the fifth component of complement (C5) by C5 convertase generated during complement activation (15). C5a anaphylatoxin is considered to be an effective target for treatment of hyperinflammation since C5a stimulates generation of tumor necrosis factor alpha (TNFα) and other inflammatory cytokines (16-18). Although C5a generated in vivo is regulated by carboxypeptidase N and more efficiently by carboxypeptidase R (CPR) (19, 20), excessive generation of C5a appears to exceed the capacity of CPR, since administration of lipopolysaccharide (LPS) at a lethal dose to rats exhausted CPR capacity before death (21). However, attempts to restrict the effect of C5a with C5a receptor (C5aR) antagonists would not be successful because C5aR is not only expressed on inflammatory leukocytes, but also on many other cell types (17). Furthermore, C5aR numbers increase in an acute inflammatory state (22).
On the other hand, antibodies to C5a have been demonstrated to be effective in treating experimental primate models of sepsis (16, 23), indicating that C5a inhibitors should be useful for treatment of patients suffering from hyperinflammation such as in sepsis and multiple organ failure (24). If an inhibitor of C5a has a therapeutic effect on sepsis, which impairs the status of cancer-bearing patients, the inhibitor could be beneficial for the cancer patients in order to improve their performance status.
AHB in C5aR, and between C5aR and C5a were analyzed by ANTIS program, and we found that amino acids 37 to 53 of C5a (RAARISLGPRCIKAFTE) comprise an antisense peptide to AHB peptides (9) of the C5aR, and this has been designated PL37 (25). This region of C5a is presumed to be a potential site for C5aR stimulation (26). Using the computer program MIMETIC (11), we generated 19 C-peps to PL37. One of the 7 inhibitory C-peps to PL37 which interfered with C5a function was termed PepA (ASGAPAPGPAGPLRPMF) (14). To improve stability, we modified PepA by acetylation of its N-terminal alanine generating acetylated PepA (AcPepA) which was more stable in animal experiments (27). In preliminary experiments with human lung tissues, AcPepA successfully suppressed the allergic response in vitro (28). Therefore, we performed experiments in Cynomolgus monkeys in lieu of using humans.
Materials and Methods
Peptides. PepA (ASGAPAPGPAGPLRPMF) whose N-terminal alanine is acetylated (AcPepA) was synthesized and purified (over 95% purity) by Biologica Co. Ltd. (Nagoya, Japan). The peptide was dissolved in saline at a concentration of 2 mg/ml and passed through a 0.22 μm Millipore filter prior to administration intravenously with an automated injection pump.
Monkeys. Cynomolgus monkeys were supplied from a breeding colony maintained at the Corporation for Production and Research of Laboratory Primates (CPRLP), Tsukuba, Japan. The Institutional Animal Ethical Committee of the Choju Medical Institute, Fukushimura Hospital, and the Institutional Animal Care Use Committees of the Tsukuba Primate Research Center, National Institute of Biomedical Innovation approved the study protocol. Animals weighed 4 to 5.5 kg, had hematocrits exceeding 36% and were free of infection, including tuberculosis. Animals were held for one month prior to LPS-lethal shock studies at CPRLP.
Titration of cytokine levels in plasma. TNFα in monkey plasma was determined using an ELISA kit purchased from Quantikine Immunoassay (Minneapolis, MN, USA). Macrophage inhibitory factor (MIF) was determined by use of an ELISA kit (29) prepared by Sapporo Immuno Diagnostic Laboratory (Sapporo, Japan). For high mobility group box 1 (HMGB1) determination, an ELISA kit from Shino-Test Co. (Sagamihara-shi, Kanagawa, Japan) was used.
Treatment of Cynomolgus monkeys. Fourteen monkeys were used for the experiment, and 13 were administered a lethal dose of LPS (4 mg/kg) sufficient to kill a monkey within 2 days, and 1 monkey was not administered LPS as an untreated control. Following sedation using ketamine hydrochloride (14 mg/kg, subcutaneously), monkeys were anesthetized with sodium pentobarbital administered through the capalic vein via a percutaneous catheter to maintain light surgical anesthesia. Oral intubation allowed animals to breathe spontaneously. Under anesthesia with sodium pentobarbital, 13 monkeys were intravenously administered 4 mg/kg LPS within 30 min. Thirty minutes after the LPS injection, 8 of the 14 animals were administered 2 mg/kg of AcPepA in 2 min followed by 2 mg/kg/h of AcPepA for 3 h. The other 6 LPS-injected monkeys were injected with saline instead of AcPepA treatment as untreated controls. Six hours after the LPS administration, anesthesia was terminated and monkeys were returned to their cages to observe their status without any additional interference. However, half (3 animals) of the untreated controls and one of the AcPepA-treated monkeys were euthanized for autopsy for histopathological analysis of inflammation at 6 h after the LPS administration.
Results
Although all three monkeys administered saline alone as an untreated control died within two days (two in one day and one in two days), administration of 2 mg/kg of AcPepA in 2 min followed by 2 mg/kg/h of AcPepA for 3 h starting 30 min after the LPS injection rescued all of 7 monkeys who returned to a healthy condition in two days (Table II). Following LPS administration, significant leukopenia and thrombopenia were observed in peripheral blood obtained 6 h after the LPS injection both from monkeys treated with AcPepA and from control monkeys treated with saline instead of AcPepA (Figure 1). The increased TNFα level in plasma obtained during experiments in AcPepA-treated monkeys was lowered by only about 30% compared with that of untreated monkeys (Figure 2). The increase in the level of MIF (Figure 3) and HMGB1 (Figure 4) after LPS injection tended to be suppressed in the AcPepA-treated monkeys. Some of the monkeys were sacrificed under anesthesia 6 h after LPS administration in order to perform autopsies. Pathological analysis of organ tissues showed serious inflammatory changes, including leukocyte infiltration, to the same extent in the lungs of both treated and untreated monkeys (Figure 5).
Discussion
The monkeys treated with AcPepA might have escaped induction of a feedback inflammatory circuit which was progressing gradually in LPS-treated monkeys at a late stage of the endotoxin shock syndrome as a vicious circle (Figure 6). This may be because HMGB1 has the capacity to stimulate toll-like receptor 4 (TLR4) and TLR2 as an endogenous stimulator (30, 31).
Consequently, inhibition of HMGB1 release presumably rescued animals suffering from septic vicious circle (32, 33). Therefore, suppression of HMGB1 in monkeys treated with AcPepA (Figure 4) could explain the extreme therapeutic effect of AcPepA on endotoxin shock in these animals (34). In other words, continuous generation of C5a by LPS or bacteria in monkeys receiving as well as possibly in patients with sepsis likely induce a cytokine ‘storm’ amplified by the release of HMGB1, resulting in a lethal effect on the host as a vicious circle of inflammation. The suppression of HMGB1 induction by inactivation of C5a could directly correlate with the survival observed following AcPepA treatment of monkeys injected with a lethal dose of LPS. Furthermore, AcPepA was shown to suppress pathophysiological events and prolonged survival time of piglets with sepsis induced by cecal ligation and perforation (CLP) (35, 36). Survival times were longer in the AcPepA-treated group than in the group treated with CLP alone (19.3 h±2.7 h vs. 9.9 h±0.7 h, p<0.005). In this case, AcPepA also delayed the HMGB-1 surge (36).
Therefore, suppression of C5 anaphylatoxin by AcPepA interferes with the induction of a cytokine ‘storm’. Since C5a has the capacity to cause release of HMGB1 following stimulation of the second C5a receptor, termed C5L2, generated on activated monocytes (37-39), inhibition of C5a successfully interferes with the above release (Figure 7) (34).
Therefore, AcPepA would be beneficial for treatment of patients with sepsis and could be administered in large amounts at an acute stage, with little likelihood of an overdose, since the half-life of AcPepA in rats is 2.5 min. Administration of AcPepA to cancer patients at their terminal stage of their disease could improve their performance status.
Acknowledgements
This work was supported in part by a Research Grant from the Japanese Ministry of Health, Welfare and Labor (08062893).
Footnotes
-
This article is freely accessible online.
- Received April 7, 2011.
- Revision received June 8, 2011.
- Accepted June 9, 2011.
- Copyright© 2011 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved