Julia
Rivera-Jiménez
*,
Carmen
Berraquero-García
,
Raúl
Pérez-Gálvez
,
Pedro J.
García-Moreno
,
F. Javier
Espejo-Carpio
,
Antonio
Guadix
and
Emilia M.
Guadix
Department of Chemical Engineering, University of Granada, 18071, Granada, Spain. E-mail: julia_rivera@ugr.es
First published on 21st November 2022
Inflammation is the response of the immune system to harmful stimuli such as tissue injury, infection or toxic chemicals, which has the aim of eliminating irritants or pathogenic microorganisms and enhancing tissue repair. Uncontrolled long-lasting acute inflammation can gradually progress to chronic, causing a variety of chronic inflammatory diseases that are usually treated with anti-inflammatory drugs, but most of them are inadequate to control chronic responses and are also associated with adverse side effects. Thus, many efforts are being directed to develop alternative and more selective anti-inflammatory therapies from natural products. One main field of interest is the obtaining of bioactive peptides exhibiting anti-inflammatory activity from sustainable protein sources like edible insects or agroindustry and fishing by-products. This work highlighted the structure–activity relationship of anti-inflammatory peptides. Small peptides with molecular weight under 1 kDa and amino acid chain length between 2 to 20 residues are generally the most active because of the higher probability to be absorbed in the intestine and penetrate into cells when compared with the larger size peptides. The presence of hydrophobic (Val, Ile, Pro) and positively charged (His, Arg, Lys) amino acids is another common occurrence for anti-inflammatory peptides. Interestingly, a high percentage (77%) of these bioactive peptides can be found in alternative sustainable protein sources such as Tenebrio molitor or sunflower, apart from its original protein source. However, not all of these peptides with anti-inflammatory potential in vitro achieve good scores by the in silico bioactivity predictors studied. Therefore, it is essential to implement current bioinformatics tools, in order to complement in vitro experiments with prior prediction of potential bioactive peptides.
Chronic inflammation is associated with increased risk of chronic diseases and disorders such as asthma, inflammatory bowel disease (IBD), cancer, cardiovascular disease, obesity and type-2 diabetes.4–7 Indeed, chronic inflammatory diseases dominate present-day morbidity and mortality worldwide with more than 50% of all deaths.8 Although the etiologies of chronic inflammatory diseases differ, the pathways that lead to pathological abnormalities are common.2 As a result, these inflammatory pathways can be explored as prospective targets for developing therapeutic treatments.
Regardless of initial stimuli's nature and location, the inflammatory cascade always involves the same steps: (1) noxious stimuli are recognized by cell surface receptors known as pattern-recognition receptors (PRRs); (2) inflammatory signaling pathways are triggered; (3) inflammatory mediators and signaling molecules are produced and released; and (4) blood vessels are dilated allowing inflammatory cells to accumulate in the inflamed tissue.2 Pattern-recognition receptors (PRRs), such as toll-like receptors (TLRs), are found in both immune and non-immune cells. PRRs can recognize conserved motifs of molecules expressed by bacterial structures called pathogen-associated molecular patterns (PAMPs), such as lipopolysaccharide (LPS) which is the major component of Gram-negative bacteria cell walls, or endogenous signals activated during tissue and cell injury named alarmins or danger-associated molecular patterns (DAMPS).9–12 Inflammatory mediators comprise cytokines of the interleukin-family (ILs) such as IL-6, IL-1β and IL-10, interferons (IFNs) like interferon-γ (INF-γ), tumor necrosis factors (TNFs) like tumor necrosis factor-α (TNF-α) and chemokines such IL-8, and they mediate inflammation through interaction with diverse cellular components or receptors such as IL-1 receptor (IL-1R), IL-6 receptor (IL-6R), and the TNF receptor (TNFR) among others.2 Once recognition of stimuli occurs, receptor activation triggers common signaling pathways including the nuclear factor kappa-B (NF-κB) and mitogen-activated protein kinase (MAPK). Regarding the NF-κB pathway, inactive protein complex NF-κB bounds to IκB protein; when PRRs recognize noxious stimuli, IκB protein degradation is induced. This releases NF-κB subunits p50 and p65 which translocate to the nucleus regulating genes involved in the inflammatory response.13 The NF-κB pathway regulate the production of pro-inflammatory cytokines (IL-1β, IL-6, IL-8, and TNF-α), anti-inflammatory cytokines (IL-10), expression of iNOS, which produces nitric oxide (NO), and COX-2, which is a key enzyme in the biosynthesis of inflammatory mediators such as prostaglandins and leukotrienes.14,15 Besides, it regulates the expression of adhesion molecules including vascular cell adhesion molecule-1 (VCAM-1) and intercellular adhesion molecule-1 (ICAM-1).16 In the case of MAPK pathway, it consists of a family of serine/threonine protein kinases, divided in three subfamilies: extracellular signal-regulated kinases (ERKs); C-Jun N-terminal kinases (JNKs); and p38 MAPKs; which upregulate or downregulate inflammation-related genes through protein phosphorylation.17 Particularly, the MAPK pathway stimulates enzyme phospholipase A2 (PLA2) activity together with translocation of NF-κB subunits to nucleus.18
While a great number of commercially anti-inflammatory drugs exist, these pharmacological therapies are often related with adverse side effects due to prolonged consumption.19,20 Non-steroidal anti-inflammatory drugs (NSAIDs), like ibuprofen, naproxen or aspirin, are some of the most frequently conventional drugs prescribed to combat chronic Inflammation.21 NSAIDs acts by inhibiting the activity of cyclooxygenase enzymes involved in the synthesis of pro-inflammatory mediators and they are associated with severe toxicity and hypertension and are contraindicated to older patients with cardiovascular, renal or hepatic complications.22 Therefore, natural compounds exhibiting anti-inflammatory activity, which can be included in functional foods or within nutraceutical formulations, are a potentially better alternative to synthetic drugs for the prevention and treatment of inflammatory diseases. For instance, the variety of immunomodulatory properties attributed to bioactive peptides has focused the attention on them for the treatment of inflammatory diseases. These peptides are small fragments of proteins that usually contain 2–20 amino acid residues per molecule and they achieve their biological activity once they are released from their parental protein. The release of bioactive peptides is accomplished by enzymatic or chemical hydrolysis, in vivo or in vitro simulated gastro-intestinal digestion or bacterial fermentation of the parental proteins.23 Although there are many studies that have observed the anti-inflammatory activity of single purified or synthesized peptides, others have evaluated the activity of whole protein hydrolysates composed of a mixture of diverse bioactive peptides.24–27 Once crossed the gastrointestinal barrier and survive enzyme degradation, active peptides are absorbed in the human body and perform a wide range of functions, including modulating physiological systems and inflammatory response due to the modulation of MAPK and NF-κB inflammatory signaling pathways through the downregulation of pro-inflammatory mediators and upregulation of anti-inflammatory mediators expression.28,29 The biological and functional properties of peptides are normally determined by their amino acid sequence, relative proportion of specific amino acids and the type of residues present at C- and N-terminals, as well as by their hydrophobicity, molecular weight/length and net charge.30 However, limited information about the relationship between structure and anti-inflammatory activity of bioactive peptides is available, which makes it difficult to understand the specific molecular mechanisms involved in their action.
Animal proteins from eggs, milk (casein and whey) and meat have been widely used for extracting bioactive peptides, but the production of these conventional animal-based protein sources is not feasible to meet the growing global protein demands in a sustainable manner.31,32 Plant protein has also been commonly used as a source of bioactive peptides, being legumes such as soybean among the most used crops. Nonetheless, the use of soybeans may be inconvenient because of the amount of antinutritional components that can cause negative effects. Moreover, soybean cultivation is typically associated with environmentally unfriendly and unsustainable practices.33–35
The difficulties with conventional protein sources from animals and plants have prompted a search for other alternative or unconventional protein sources, such as those derived from edible insects and by-products from the food or agriculture industry. Edible insects can be produced with less negative impact on the environment than existing livestock.36 From the wide array of edible insects, yellow mealworms known as Tenebrio Molitor have been reported to provide a source of high-quality protein (53% protein in a dry basis) containing high amounts of essential amino acids.37–40 A recent study show that insect-derived protein can be rapidly digested and effectively absorbed, with no different from a high-quality dairy protein source.36 Alternative plant-based proteins are gaining increasing interest and a wide range of flour or defatted flours from different sources (especially legumes, cereals and oilseeds) are already in the market.41 By-products of the extraction of sunflower (Helianthus annuus L.) oil are considered an interesting raw material for biopeptides production due to their high content of protein. For instance, defatted sunflower meal (principal by-product) contains about 28–42% protein (dry basis).42,43 Other interesting crops, due to their favorable composition and availability, are Lupine family such as Lupinus albus or Lupinus angustifolius which have a high protein content (about 35%–40% in a dry basis) containing a high proportion of essential amino acids.44 Nutritional characteristics along with low water requirements and high productivity make chickpeas (Cicer arietinum) a great alternative to conventional protein sources. Chickpeas are an inexpensive source rich in protein (approximately 20% dry mass) with excellent balance of essential amino acid composition, high bioavailability, and low level of antinutritional factors.26,45 It is important to mention that pea protein, that is usually used as feed with a low added value, also contains a high content of protein (about 18% protein in a dry basis) and an adequate content of essential amino acids.46,47 Olive (Olea europaea) is another interesting source of protein (17% protein in seeds), since olive oil consumption has experienced a continuous increase, and hence waste and by-products derived from the olive production (e.g. olive seed flour) have also increased. Recently, it has been reported a novel strategy for the revalorization of olive residues based on the extraction of bioactive peptides.48 In addition, the recovery of protein-rich fish by-products resulting from the processing of sardine (Sardina pilchardus) (15.5–19.1% protein in a wet basis) is also gaining importance as well as the valorization of low commercial value and traditionally discarded fish species as blue whiting (Micromesistius poutassou) and horse mackerel (Trachurus trachurus), which contain between 14%–19% of protein (wet basis).49–52
One of the preferred approaches to identify bioactive peptides is the empirical approach which begins with the hydrolysis of the selected protein source, testing different types of protease and operating conditions such as enzyme/substrate ratio, time, pH or temperature.53 Afterwards, a bioactivity screening of the protein hydrolysate is carried out. However, these hydrolysates are a complex mixture of peptides with different sizes and compositions, as well as other undesirable components such free amino acids, enzymes and nonreacted native proteins. Therefore, fractionation methods are required to separate active peptides from other non-active hydrolysate components, using common techniques such as ultrafiltration.54,55 Subsequently, the potential active peptides are identified from the most active fraction.56 Generally, a further purification process is required, consisting of sequential chromatographic treatments including gel-filtration chromatography and high-performance liquid chromatography (HPLC),57–59 which are combined with mass spectrometry (MS), particularly tandem MS (MS/MS) or by a combination of liquid chromatography and mass spectrometry (LC-MS).25,60 Finally, the bioactivity of purified peptides is validated in vitro and/or in vivo through the quantification of inflammatory mediators such as prostaglandin E2 (PGE2), TNF-α, IL-6, IL-1β, IL-10, IL-8, VCAM-1 or NO in cell lines or by measuring the inhibition of enzymes involved in inflammation such as COX-2, LOX and iNOS.61–64 Nonetheless, current techniques for separation, purification, identification, and quantification of biopeptides from their natural matrices are limited by high technicality and high cost, which restricts their therapeutic competitiveness and industrial application.53 Processes for peptide preparation and identification represent an analytical challenge because they implie high yield and purity for the quantification of its potential bioactivity and the validation of their structure–activity relationship.
Recently, computational methods for predicting bioactivities of peptides have emerged to reduce the experimental effort, and in silico tools have become an inevitable approach prior to in vitro and/or in vivo investigation.65 Bioinformatics-driven techniques have recently been suggested to solve the major drawbacks of the empirical methodology (reagent and time investments) to identify biopeptides, although there is still room for improvement. This approach is helped by the development of profiles of protein fragments with potential biological activity, based on a template sequence from the transcriptome of the organism. Once the bioactive sequence or sequences have been identified, its bioactivity could be predicted by machine learning web-based tools or prediction servers and for further validation, they are chemically synthesized and tested individually for their bioactivity either in vitro or in vivo.66,67 In comparison to classical approaches, in this case the trial-and-error approach of the initial stage can be avoided, and the obtaining of biologically active peptide becomes less laborious and time-consuming.
This work reviews the current literature on anti-inflammatory peptides and protein hydrolysates from natural sources, such as animals and plants, and synthetic sources, including their production and mechanism of action. Particularly, this work systematically studies the role of peptides sequence, structure and physicochemical properties on their anti-inflammatory or immunomodulatory activity. In addition, the potential release of the most bioactive sequences from sustainable protein sources was assayed. Moreover, this study evaluates the correlation between predicted anti-inflammatory activity by two available bioinformatics tools.
Among all the peptide sequences identified in the selected literature for exhibiting anti-inflammatory potential in vitro, a classification was made according to the inhibition and/or production values of inflammatory markers reported for each peptide. From this classification, a threshold for each inflammatory marker was chosen. The selection threshold is determined by the value at which the activity reported is significantly higher than the rest of the activities reported for that specific inflammatory marker. The threshold selected for inhibition values of NO was 70% (30% production), 52% for iNOS, 65% for COX-2, 50% for 5-LOX, 65% for TNF-α, 75% for IL-1β, 75% for IL-6, and 800 pg mL−1 for the production of the anti-inflammatory cytokine IL-10. Peptide sequences showing values below these thresholds were not selected for further study. The threshold selected for production values of TNF-α was 80 pg mL−1, 99 pg mL−1 for IL-1β, 131 pg mL−1 for IL-6, 50 pg mL−1 for IL-8 and PGE2 and 14 μM for nitrite. Peptide sequences whose values exceed this threshold were excluded from the selection. Finally, peptides that meet the selection threshold for more than one inflammatory marker (e.g., they reached the inhibition threshold for COX-2 and also for TNF-α, presenting high values in both cases) were selected for further analysis.
Protein type | Protein source | Peptide obtention | Peptide sequencea | Regulatory mechanism | Cell type | Ref. |
---|---|---|---|---|---|---|
a Amino acid sequence has been written from the first residue at N-terminus to the last residue at C-terminus (considering C-terminus the end of the peptide amino acid chain). Caco-2: human colon adenocarcinoma cell line. HUVECs: human umbilical vein endothelial cell line. RAW 264.7: murine macrophage cell line. J774A.1: mouse reticulum cell sarcoma cell line. BMDM: murine bone marrow-derived macrophages cell line. BMDC: murine done marrow-derived dendritic cell line. HD11: chicken macrophage cell line. THP-1: human leukemia monocytic cell line. U937: human monocytic cell line. C518: rat C518 knee joint degenerative cartilage cells. | ||||||
Egg and dairy | Egg | Enzymatic hydrolysis in silico (pepsin and trypsin) | FL | Reduce IL-8 secretion, TNF-α, IL-8, IL-6 and IL-1β expressions and increased IL-10 expression | Caco-2 | 82 |
MK | ||||||
LL | ||||||
CR | Reduce IL-8 secretion, TNF-α, IL-8, IL-6 and IL-1β expressions, phosphorylation of pJNK and p-p-38 and increased IL-10 expression | |||||
HC | ||||||
GID (pepsin and pancreatin) | MLGATSL (ML-7) | Inhibit expressions of TNF-α, IL-8, IL-6, IL-1β, IL-12, JNK, IκB, and p38 and increase IL-10 expression | Caco-2 | 84 | ||
DEDTQAMPFR (DR-10) | ||||||
DEDTQAMPF (DF-9) | ||||||
GW | Reduce the expression of TNF-α, ICAM-1 and VCAM1 | HUVECs | 83 | |||
Milk | Enzymatic hydrolysis (trypsin) | KVLPVPQK (K-8-K) | Reduce levels of NF-κB in the nucleus and TNF-α, IL-6, and IL-1β gene expression | Caco-2 | 86 | |
Enzymatic hydrolysis (alcalase) | DQWL | Inhibit gene expression of IL-1β, COX-2, and TNF-α | RAW 264.7 | 85 | ||
Marine | Baijiao sea bass (Lateolabrax maculatus) | Chemical extraction | DAPAPPSQLEHIRAA | Inhibit NO production | RAW 264.7 | 89 |
AADGPMKGILGY | ||||||
Sea anemone (Heteractis crispa) | Chemical extraction | RGICSEPKVVGPCKAGLRRFYYDSETGECKPFIYGGCKGNKNNFETLHACRGICRA (HCRG1) | Reduce TNF-α and IL-6 secretion and IL-1β precursor (proIL-1β) expression levels | J774A.1/RAW 264.7 coculture | 90 | |
RGICLEPKVVGPCKARIRRFYYDSETGKCTPFIYGGCGGNGNNFETLHACRGICRA (HCRG2) | ||||||
Clam worm (Marphysa sanguinea) | Chemical extraction | NCWPFQGVPLGFQAPP | Inhibit IL-1β, TNF-α, NO, iNOS, and COX-2 gene expression | RAW 264.7 | 73 | |
Sea snake venom gland (Hydrophis cyanocinctus) | Chemical synthesis | DEQHLETELHTHLTSVLTANGFQ (Hydrostatin-SN1) | Inhibit TNF-α, IL-6, and IL-1β | RAW 264.7 | 91 | |
Mollusk Abalone (Haliotis discus hannai) | GID (pepsin, trypsin, α-chymotrypsin) | PFNEGTFAS | Inhibit NO production, reduce gene transcription levels of IL-1β, IL-6, and TNF-α, and suppress phosphorylation of p-p38 and p-JNK MAPKs | RAW 264.7 | 94 | |
Oyster (Ostreidae) | Enzymatic hydrolysis (Protamex, Neutrase) + GID (pepsin, trypsin, α-chymotrypsin, carboxypeptidase A) | YA | Inhibit NO production | RAW 264.7 | 183 | |
Bivalve visceral mass (Meretrix meretrix) | Enzymatic hydrolysis (trypsin, pepsin, alcalase and papain) | HKGQCC | Reduce iNOS activity, inhibit production of TNF-α and IL-1β, and prevent activation of COX-2 | RAW 264.7 | 74 | |
Enzymatic hydrolysis (trypsin, pepsin, alcalase, papain) + GID (pepsin, trypsin, α-chymotrypsin) | GQCC (MMV2) | |||||
Sturgeon muscle (Acipenseridae) | Enzymatic hydrolysis (alcalase) | HLDDALRGQE | Reduce NO, IL-6, and IL-1β production and inhibit phosphorylation levels of MAPKs | RAW 264.7 | 63 | |
Herring milt (Clupea harengus) | Enzymatic hydrolysis (alcalase) | IVPAS | Inhibit iNOS pathway decreasing NO production | RAW 264.7 | 96 | |
FDKPVSPLL | ||||||
Peanut worm (Sipunculus nudus Linn.) | Enzymatic hydrolysis (alcalase) | TVNLAYY | Inhibit NO production and reduce gene expression of iNOS, IL-6, TNF-α, and COX-2 | RAW 264.7 | 97 | |
LSPLLAAH | ||||||
Sea cucumber (Actinopyga lecanora) | Enzymatic hydrolysis (bromelain) | LREMLSTMCTARGA | Inhibit NO production | RAW 264.7 | 99 | |
VAPAWGPWPKG | ||||||
AVGPAGPRG | ||||||
Salmon skin (Salmo salar) | Enzymatic hydrolysis (Flavourzyme) | QA | Reduce NO, IL-6, IL-1β, and TNF-α secretion | RAW 264.7 | 100 | |
APD | ||||||
KA | ||||||
WG | ||||||
Salmon bones (Salmo salar) | Enzymatic hydrolysis (papain) | SNKGGGRPN | Inhibit NO production and iNOS, IL-6, TNF-α and COX-2 mRNA levels | RAW 264.7 | 101 | |
TVTVYSLLR | ||||||
PGVATAPTH | ||||||
SLPEANSLRHR | ||||||
LLGLGLPPA | ||||||
TLGTGLCPV | ||||||
LKPKGGSVP | ||||||
FGLLVNPGA | ||||||
NGRACSYKLWD | ||||||
Salmon pectoral fin (Salmo salar) | Enzymatic hydrolysis (pepsin) | PAY | Inhibit NO, PGE2, IL-6, TNF-α and IL-1β production | RAW 264.7 | 60 | |
Sturgeon muscle (Acipenseridae) | Enzymatic hydrolysis (pepsin) | KIWHHTF | Reduce NO, IL-6, and IL-1β production and inhibit phosphorylation of MAPKs | RAW 264.7 | 63 | |
VHYAGTVDY | ||||||
Terrestrial | Spider venom gland (Pardosa astrigera) | Chemical synthesis | AMMAESRKDNCIPKHHECTSRPKDCCKQNLMQFKCSCMTIIDKNNKETERCKCDNSIFQKVAKTSVNIGKAVVTK (Lycotoxin-Pa4a) | Reduce NO production via downregulation of the iNOS gene and COX-2, TNF-α and IL-1β expression | RAW 264.7 | 66 |
Cecropia moth (Hyalophora cecropia) | Chemical synthesis | KWKLFKKIEKVGQNIRDGIIKAGPAVAVVGQATQIAK (Cecropin A) | Reduce NO production, mTNF-α and mIL-1β cytokine release, and COX-2 expression | RAW 264.7 | 107 | |
European red frog skin (Rana temporaria) | Chemical synthesis | FVQWFSKFLGRIL (Temporin-1TL) | Inhibit TNF-α and NO production, and TNF-α and iNOS mRNA expression | RAW 264.7 | 134 | |
Brazilian scorpion venom (Tityus obscurus) | Chemical synthesis | KIASVLGGILSPILSFF (ToAP4) | Reduced TNF-α and IL-1β gene expression and protein levels | J774, BMDMs and BMDCs | 102 | |
KIASILGGILGPIMGIF (ToAP3) | ||||||
Frog skin (Hylarana guentheri) | Chemical synthesis | GLFSKKGGKGGKSWIKGVFKGIKGIGKEVGGDVIRTGIEIAACKIKGEC (Esculentin-1GN) | Inhibit NO production, NO, IL-1β, IL6, and TNF-α gene expression and enhanced IL-10 expression | RAW 264.7 | 104 | |
Centipede (Scolopendra subspinipes mutilans) | Chemical synthesis | KKASKSVIKIFYKCM (Scolopendrasin IX) | Inhibit TNF-α, IL-6, and enhanced IL-10 production | Mouse neutrophil | 184 | |
Chicken (Gallus gallus domesticus) | Chemical synthesis | FRASGQITITVKPRFRRIKRLFRGFR (CATH-2) | Inhibit IL-1β transcription and NO production | HD11 | 185 | |
Human (Homo sapiens) | Chemical synthesis | SIFGKIFKRIIRVAWK (Hs02) | Inhibit TNF-α production | C57BL/6/THP-1 coculture | 71 | |
Hard tick (Amblyomma variegatum) | Chemical synthesis | HLHMHGNGATQVFKPRLVLKCPNAAQLIQPGKLQRQLLLQ (Amregulin) | Inhibit TNF-α, IL-8 and IFN-γ production | Rat splenocytes | 103 | |
Dynastid Beetle (Allomyrina dichotoma) | Chemical synthesis | AFWCLIRRTVAA (Allomyrinasin) | Reduce NO and PGE2 production and COX-2 expression | RAW 264.7 | 67 | |
Chinese scorpion (Mesobuthus martensii) | Chemical extraction | HYGH | Reduce NO, TNF-α, IL-6, and IL-1β production and inhibit IκBα degradation, p65 nuclear translocation, NF-κB activation and phosphorylation of ERK, JNK, and p38 MAPKs | RAW 264.7 | 64 | |
Horsefly salivary glands (Tabanus yao) | Chemical extraction | RGQANILAGKNIKIRSGAAAGVGKTPQKANVEVLALGIW (Cecropin-TY1) | Inhibit NO, IL-1β, IL6, and TNF-α production | RAW 264.7 | 106 | |
Black fly salivary glands (Simulium bannaense) | Chemical extraction | GKLTKDKLKRGAKKALNVASKV (SibaCec) | Inhibit NO, TNF-α, IL-1β, and IL-6 transcription and production along with inhibiting MAPKs and NF-κB pathways | RAW 264.7 | 105 | |
Red deer velvet antler (Cervus elaphus Linnaeus) | GID (pepsin and pancreatin) | LAN | Inhibit NO production | RAW 264.7 | 59 | |
VH | ||||||
IA | ||||||
AL | ||||||
Locust (Schistocerca gregaria) | GID (α-amylase, pepsin, pancreatin, and bilioA) | FDPFPK | Inhibit LOX and COX activity | Comercial assay | 62 | |
Gastropod visceral mass (Harpa ventricosa) | Enzymatic hydrolysis (trypsin, alcalase and pepsin) | AKGTWK | Inhibit NO, TNF-α and IL-1β production | THP-1 | 58 | |
Hen spent muscle (Gallus gallus domesticus) | Enzymatic hydrolysis (protease M and Protex 50FP) | SFMNVKHWPW | Inhibit IL-6 production | U937 | 108 | |
AFMNVKHWPW | ||||||
Chicken Feather Meal (Gallus gallus domesticus) | Enzymatic hydrolysis (Flavourzyme) | SNPSVAGVR | Reduce gene transcription levels of iNOS, TNF-α, COX-2 and IL-6 | RAW 264.7 | 110 | |
Chicken sternal cartilage (Gallus gallus domesticus) | Enzymatic hydrolysis (papain) | VAIQAVLSLYASGR | Inhibit IL-1β, TNF-α, and PGE2 production | C518 | 109 |
Protein type | Protein source | Peptide obtention | Peptide sequencea | Regulatory mechanism | Cell type | Ref. |
---|---|---|---|---|---|---|
a Amino acid sequence has been written from the first residue at N-terminus to the last residue at C-terminus (considering C-terminus the end of the peptide amino acid chain). EA.hy926: human umbilical vein cell line. 3T3-L1: murine preadipocyte cell line. IEC-6: rat intestinal epithelial cell line. ARPE-19: human retinal pigment epithelial cell line. HepG2: human hepatocellular carcinoma cell line. BV-2: murine microglial cell line. | ||||||
Gramineae | Rice (Oryza sativa L.) | Enzymatic hydrolysis (trypsin) | IGVAMDYSASSKR | Inhibit gene expression of iNOS, IL-1β, IL-6, and TNFα | RAW 264.7 | 111 |
DNIQGITKPAIR | ||||||
IAFKTNPNSMVSHIAGK | ||||||
QRDFLLAGNKRNPQAY | Inhibit nuclear translocation of p65, NO and TNF-α production and reduced TNF-α, iNOS, IL-6 and IL-1β transcription | RAW 264.7 | 112 | |||
Corn zein (Zea mays) | Enzymatic hydrolysis (alcalase, neutral protease, thermolysin) + GID (pancreatin) | PPYLSP | Reduce gene expression of TNF-α and VCAM-1 | EA.hy926 | 113 | |
IIGGAL | ||||||
FLPPVTSMG | ||||||
Leguminosae | Black soybean (Glycine max) | Extraction | RGD | Inhibit NO, TNF-α, IL-1β, and IL-6 production | RAW 264.7 | 120 |
Soybean (Glycine max) | Chemical synthesis | FLV | Inhibit TNF-α and IL-6 release | RAW 264.7/3T3-L1 coculture | 72 | |
Enzymatic hydrolysis (alcalase and Neutrase) | YGGGGE | Reduce NO production and gene expression, TNF-α, IL-1β and IL-6 production and promote IL-10 gene expression | IEC-6 | 61 | ||
SEGGFLE | ||||||
Enzymatic hydrolysis (trypsin) | SLVNNDDRDS (S-10-S) | Reduce levels of NF-κB in the nucleus and TNF-α, IL-6, and IL-1β gene expression | Caco-2 | 86 | ||
Fermented sorghum Baijiu vinasse (Sorghum bicolor) | Enzymatic hydrolysis (corolase PP) | KLPDHPKLPK (VPH-1) | Inhibit NO, TNF-α, IL-6 and IL-1β production | RAW 264.7 | 124 | |
VDVPVKVPYS (VPH-2) | ||||||
Lupin seeds (Lupinus angustifolius L.) | Enzymatic hydrolysis (alcalase) | GPETAFLR | Reduce IL-1β, IL-6, and TNF-α production and increase IL-10 production and gene expression | THP-1 | 122 | |
ARPE-19 | 121 | |||||
GID (pepsin and pancreatin) | IQDKEGIPPDQQR (IQD) | Inhibit TNF-α, IL-6, IL-1β production | RAW 264.7 | 186 | ||
Foxtail Millet (Setaria italica) | GID (pepsin and pancreatin) | QNWDFCEAWEPCF | Inhibit NO, IL-6, and TNF-α production | RAW 264.7 | 125 | |
EDDQMDPMAK | ||||||
Other plants | Hempseed (Cannabis sativa) | Enzymatic hydrolysis (pepsin) | IGFLIIWV | Reduce NO production levels and modulate IFN-γ, TNF-α, IL-6 and IL-10 levels | HepG2 | 126 |
WVSPLAGRT | ||||||
Lychee seeds (Litchi chinensis Sonn.) | Enzymatic hydrolysis (alcalase, Flavourzyme, Neutrase) | KVRTKLLPP | Inhibit NO production by down-regulation of iNOS and IL-6 | RAW 264.7 | 127 | |
RPLVTHK | ||||||
MKLCWQKSIHGS | ||||||
Walnut (Juglans regia) | Enzymatic hydrolysis (Viscozyme L and pancreatin) | GVYY | Inhibit gene expression of COX2 and iNOS, and production of NO and PGE2 | BV-2 | 128 | |
APTLW | ||||||
LPF |
Protein source | Parental protein | Peptide sequencea | Regulatory mechanism | Cell type | Ref. |
---|---|---|---|---|---|
a Amino acid sequence has been written from the first residue at N-terminus to the last residue at C-terminus (considering C-terminus the end of the peptide amino acid chain). HOb-OA: human osteoblasts–osteoarthritis cell line. | |||||
AMP | Tick defensin OsDef1 | KGIRGYKGGYKGAFKQTKY (Os–C) | Inhibit NO and TNF-α production | RAW 264.7 | 130 |
KGIRGYKGGYCKGAFKQTCKCY (Os) | |||||
Rana temporaria temporin-1TL (TL) | FVQWWSKWLGRIL (TL-1) | Inhibit TNF-α and NO production, and TNF-α and iNOS mRNA expression | RAW 264.7 | 134 | |
FVRWWSKWLRRIL (TL-2) | |||||
FVRWWSRWLRRIL (TL-3) | |||||
FVKWWSKWLKKIL (TL-4) | |||||
Pseudin-2 (Ps) | GLNALKKVFQGIHEAIKKINNHVQ (Ps-K18) | Inhibit NF-κB gene expression and TNF-α, IL-6 and IL-1α production | RAW 264.7 | 133 | |
Chensinin-1 | SAVWRHWRRFWLRKHRKH (MC1-1) | Inhibit TNF-α and IL-6 production | RAW 264.7 | 132 | |
Chemokine CXCL14 | YKRWKKNWAKYWKIFRK (CXCL14-C17-a2) | Inhibit NO, TNF-α and IL-6 production | RAW 264.7 | 131 | |
YKRWKKRWAKYWKKFRK (CXCL14-C17-a3) | |||||
Chicken cathelicidin-2 (CATH-2) | QITITVKPRFRRIKRLFR | Inhibit IL-1β transcription and NO production | HD11 | 185 | |
QITITVKPRFRRIKRLFRGFR | |||||
Hormone | Glucagon-like peptide-1 (GLP-1) | HAEGTFTSDVSSYLEGQAAKEFI (Liraglutide) | Reduce NO, PGE2, and IL-6 production, IL-6, COX-2, and TNF-α gene expression | RAW 264.7 | 136 |
Parathyroid hormone–related protein (PTHrP) | ETNKV | Reduce IL-6 and PGE2 production | HOb-OA | 135 | |
ETNKVETYKEQPLKTPGKKKKGKP GKRREQEKK | Reduce IL-6, PGE2, TNF-α production and NF-κB activation | ||||
Protease-activated receptor-1 (PAR-1) | NPNDKYEPFWEDEEKNESGL | Reduce IL-1β production | THP-1 | 187 | |
Protease-activated receptor-3 (PAR-3) | GAPPNSFEEFPFSALEGWTGATIT | ||||
Hybrid | Bovine lactoferrin (LfcinB)-human cathelicidin (LL-37) hybrid | RLWKKILKVIRKPRWQWRR (Lf-KR) | Inhibited NO and TNF-α expression and production | RAW 264.7 | 139 |
Cathelicidin-2 (CATH-2)-thymopentin (TP5) hybrid | RWGRFLRKIRRFRRKDVY (CTP) | Reduce TNF-α, IL-1β, and IL-6 production | RAW 264.7 | 138 | |
De novo designed | Non-existent | KKIRVRLSA (SET-M33D) | Reduce TNF-α, IL6, COX-2, iNOS and NF-κB gene expression | RAW 264.7 | 143 |
Inhibit TNF-α, IL-6, and IL-1β, IL-8 and COX-2, iNOS and NF-κB activation | RAW 264.7 | 142 | |||
GAKYAKIIYNYLKKIANALW (GW-A2) | Reduce NO, iNOS, COX-2, TNF-α and IL-6 expression levels, phosphorylation of MAPK and NF-κB activation | RAW 264.7 | 141 | ||
LKWLKKLLKKL (WALK11.3) | Inhibit NO, COX-2, IL-1β, IL-6, INF-β, and TNF-α expression | RAW 264.7 | 140 | ||
Spider venom glands (Agelena koreana) | FKGLAKLLKIGLKALAKVIQ (Ak-N′) | Reduce TNF-α, IL-6, and IL-1β gene expression | THP-1 | 188 | |
NKGLAKLLKIGLKALESVIQ (Ak-N′m) |
In addition to cell assays, there are other types of enzyme-type assays to study the anti-inflammatory potential. Inflammatory enzymes are also very attractive therapeutic targets since their expression and elevation is associated with most forms of inflammation. Several studies make use of commercial inhibition assays for COX-2, LOX and PLA2.27,75 For instance, the potential anti-inflammatory capacities of lupin protein hydrolysate were studied by determining its in vitro inhibition of enzymes involved in the inflammatory process including PLA2 and COX-2.76 By studying the inhibition of COX-1 and COX-2 together with LOX activities, anti-inflammatory peptides have been identified in the digest of millet grains as well as digested protein fractions from chia seed.77,78 Furthermore, the inhibitory capacity against COX-2 and LOX of biopeptides obtained from edible insects have been measured.62
Similarly to egg-derived peptides, anti-inflammatory peptides from milk were among the first-studied biopeptides. In this case, the reviewed studies carried out the hydrolysis with a single enzyme instead of combining pepsin with different enzymes like is done for egg proteins. For example, properties against inflammation of peptides derived from whey bovine protein treated with alcalase have been evaluated in RAW 264.7 cells and peptide DQWL showed strong inhibitory ability on IL-1β and COX-2 with a reduction of 49.5% and 62.1%, respectively. Additionally, DQWL treatment suppressed nuclear translocation of the p65 component of NF-κB and blocked IκB kinase phosphorylation, IκB degradation and p38 activation.85 Another study on milk-derived peptides identified a peptide with sequence KVLPVPEK (K-8-K) exerting anti-inflammatory activity through inhibition of the NF-κB pathway.86
Apart from chemical extraction of peptides naturally present in marine-animal proteins, anti-inflammatory peptides have also been obtained by enzymatic hydrolysis of marine proteins. Anti-inflamamtory peptides with sequences PFNEGTFAS, YA and GQCC were obtained by gastrointestinal digestion with a combination of trypsin, pepsin and α-chymotrypsin of proteins from mollusk abalone (Haliotis discus hannai) intestine, oyster (Ostreidae) and bivalve (Meretrix meretrix) visceral mass, respectively.74,94,95 It should be noted that tetrapeptide GQCC, produced from the digestion of its parental peptide HKGQCC obtained from hydrolysis with a combination of four enzymes (trypsin, pepsin, alcalase and papain), outperformed its parental peptide in inflammatory potential, suppressing NO production by 58% against 45% supressed with HKGQCC. These three peptides have the same regulatory action on iNOS activation by inhibiting the production of NO, showing a good IC50 value of 54.07 μg ml−1 in the case of PFNEGTFAS. Besides, they have other effects such as the regulation of the NF-κB pathway (except in the case of the dipeptide YA).
Furthermore, recent findings on fishing by-products treated with Alcalase led to the identification of two novel peptides IVPAS and FDKPVSPLL from herring milt hydrolysate and peptide HLDDALRGQE from sturgeon muscle protein hydrolysate, that showed anti-inflammatory activity by decreasing NO production in murine macrophage cells.63,96 Another study reported the hydrolysis with Alcalase of peanut worms protein, where two identified peptides TVNLAYY and LSPLLAAH exhibited high NO-inhibitory activity along with pro-inflammatory cytokines regulation, such as IL-6 and TNF-α, and COX-2 suppression.97 Moreover, anti-inflammatory peptides KIWHHTF and VHYAGTVDY from pepsin–sturgeon muscle hydrolysate also suppressed NO production by reducing nitrite levels and down-regulate the MAPK pathway.98
Another enzymatic treatment of marine proteins using bromelain has also reported good results with respect to the production of peptides with anti-inflammatory activity. For example, bioactive peptides with sequences LREMLSTMCTARGA, AVGPAGPRG and VAPAWGPWPKG from sea cucumber (Actinopyga lecanora) bromelain-hydrolysate, obtained using bromelain, inhibited NO production in murine macrophages showing an IC50 value of 572.096 mg mL−1 and 674.435 mg mL−1, respectively.99
Alternatively, enzymes such as Flavourzyme and papain have also been used to produce anti-inflammatory peptides from salmon by-products. Atlantic salmon provides large quantities of low-value protein rich co-products, such as salmon skin and bones that can be upgraded from by-products to high-value functional ingredients such as biopeptides. As an example of this, hydrolysis with Flavourzyme of salmon skin produced dipeptides QA, KA and WG and tripeptide APD, which significantly reduced production of IL-6, IL-1β and TNF-α in LPS-stimulated murine macrophages. Peptide QA had the highest overall anti-inflammatory impact of all of them having an IC50 value against NO production of 849.3 μM.100 Moreover, salmon bone hydrolysis with papain, in comparison with Alcalase, Flavourzyme and Neutrase hydrolysates, produced a range of potential anti-inflammatory peptides such as peptides SNKGGGRPN and TVTVYSLLR, which were found to have a marked NO-inhibitory activity with an IC50 value of 2.56 μg mL−1 along with IL-6, TNF-α and COX-2 regulatory activity.101 In addition, a tripeptide from salmon pectoral fin by-product, PAY, produced by a different enzyme treatment using pepsin instead of Flavourzyme or papain, also exhibited anti-inflammatory action via inhibiting production of NO by 63.80% plus reducing production of inflammatory markers such as PGE2 among others and activation of COX-2.60
Regarding peptides naturally present in these animals, RNA sequencing has proven to be an interesting tool to assist the identification of novel anti-inflammatory peptides. As an example, after sequencing the RNA of spider Pardosa astrigera venom gland and of Dynastid beetle (Allomyrina dichotoma) potential anti-inflammatory peptides named Lycotoxin-Pa4a (peptide toxin) and Allomyrinasin, from the two sources respectively, were selected based on their physicochemical properties and structural characteristics. Both peptides showed the ability to decrease NO production via downregulation of iNOS gene and suppressed the expression of COX-2 apart from regulating the production of other inflammatory mediators such as IL-1β in LPS-stimulated murine macrophages.66,67
Another powerful tool towards the identification of anti-inflammatory peptides with desired properties is cDNA display, which enables selection of peptide libraries that encompass thousands of sequences. Thanks to this technique, peptides ToAP3 and ToAP4 from Brazilian scorpion (Tityus obscurus) venom, peptide Esculentin-1GN from the skin of the frog Hylarana guentheri and peptide Amregulin from hard tick Amblyomma variegatum were identified as potentially anti-inflammatory peptides.102–104 These four peptides act by inhibiting the production of different proinflammatory cytokines such as TNF-α, and in the case of peptides ToAP3 and ToAP4 they are capable of increasing the production of anti-inflammatory cytokine IL-10, and exerted their properties in various cell types: J774, murine bone marrow-derived macrophages (BMDMs) and dendritic cells (BMDC). In addition, peptide Esculentin-1GN is the only one that shows the ability to regulate NO production and iNOS transcription.
In addition to the in silico tools for peptide identification discussed above, anti-inflammatory peptide discovery has also been made by chemical extraction, as in the case of tetrapeptide HYGH purified from the crude protein extract of the Chinese scorpion (Mesobuthus martensii). Peptide HYGH showed different mechanisms of action by regulation of NF-κB together with MAPK inflammatory signaling pathways along with suppressing NO production by 76.8%.64 Anti-inflamamtory peptide GKLTKDKLKRGAKKALNVASKV (Sibasec) has also been extracted from the salivary glands of black fly (Simulium bannaense) after determining its amino acid composition by cDNA sequencing.105 Since flies attract attention due to their good nutritional value (17.5–67 g per 100 g protein), more peptides with anti-inflammatory potential such as Cecropin-TY1 were discovered from salivary glands of horsefly.106 Cecropin TY1 and Sibasec share the same mechanism of action inhibiting the production of NO, TNF-α, IL-1β, and IL-6 and the activation of iNOS by 58.8% and 51.9%, respectively. Furthermore, synthesis of known antimicrobial peptides (AMPs) has also been investigated to test whether they also have effect against inflammation. A proof of this is peptide Cecropin A, an antibacterial 37-residue peptide isolated from the cecropia moth that proved to be a potential agent for prevention and/or treatment of inflammatory disorders, since its mechanism of action is very complete. Cecropin A was able to reduce NO production by total suppression of nitrite production along with reducing TNF-α and IL-1β release, COX-2 expression and phosphorylation of ERK, JNK, p38 inhibiting MAPK pathway.107
Apart from identifying already existing anti-inflammatory peptides in the protein sequence of terrestrial animal sources, several studies have identified peptides with this bioactivity after in vitro simulated gastrointestinal digestion with pepsin/pancreatin mixture, either in combination with other enzymes such as α-amylase or alone. Results of these studies reported that edible insect locust Schistocerca gregaria derived peptide FDPFPK has the ability to inhibit COX-2 and LOX activity giving a IC50 values of 7.40 and 2.85 mg mL−1, respectively, and that red deer Cervus elaphus Linnaeus velvet antler derived peptides LAN, VH, IA and AL had also a strong inhibitoty activity against NO, showing greater effect in the case of the three dipeptides compared to the longer peptide LAN.59,62
There are also terrestrial animal-protein hydrolysates produced by with a single enzyme where peptides with good effects against inflammation have been identified. For example, visceral mass protein of gastropod Harpa ventricose was hydrolysed by trypsin, Alcalase and pepsin individually. The most active hydrolysate was the tryptic one where an hexapeptide AKGTWK was found to suppresed NO production up to 61.6% along with inhibition of TNF-α and IL-1β production.58 Moreover, anti-inflammatory peptides were generated from chicken by-products using hydrolysis with individual commercial enzymes, i.e., peptides SFMNVKHWPW and AFMNVKHWPW identified from spent hen muscle protein and treated with Protex 50FP reported a good IC50 value of 100 μg mL−1 for IL-6 production.108 Another examples is chicken-collagen peptides such as VAIQAVLSLYASGR from papain hydrolysate that significantly inhibited the secretion of inflammatory cytokines IL-1β, TNF-α, and PGE2.109 Furthermore, peptide SNPSVAGVR obtained after hydrolysis with Flavourzyme of chicken feather meal that showed inhibitory activity against NO production with an IC50 value of 55.2 mM and reduced gene transcription levels of COX-2 and IL-6 by 83 and 96%, respectively.110
Apart from enzymatic hydrolysis with trypsin, peptides PPYLSP, IIGGAL and FLPPVTSMG with strong anti-inflammatory activity have also been identified by hydrolyzing zein, a by-product of corn starch, with an enzymatic mixture of Alcalase, neutral protease and thermolysin in combination with a subsequent simulated gastrointestinal digestion with gastric fluid and pancreatin. These three peptides exhibit their effect by decreasing TNF-α expression and inhibiting ICAM-1 by 36.5–28.6% along with VCAM-1 by 54–38.9%; and their action is believed to be due to their successful transport across Caco-2 cell monolayers without digestion by peptidases.113
Other novel peptides with anti-inflammatory activity have been obtained by enzymatic hydrolysis of soybean. Specifically, peptides YGGGGE and SEGGFLE obtained by hydrolysis with a mixture of Alcalase/Neutrase reported protective activity against intestinal inflammation, as well as peptide SLVNNDDRDS, named S-10-S, obtained by trypsin hydrolysis, that exerted its effect through inhibition of the NF-κB pathway and downregulation of the gene expression of IL-1β and TNF-α.61,86
The enzymes Corolase PP and Alcalase have also been used successfully to hydrolyze leguminosae plant proteins and identify anti-inflammatory peptides such as GPETAFLR an octapeptide derived from lupine protein hydrolyzed with alcalase that exerted protection against inflammatory damage in retinal pigment epithelium cells, as well as modulated the inflammatory response and plasticity in human primary monocytes.121,122 Moreover, GPETAFLR inhibited neuroinflammation by preventing inflammation in BV-2 microglial cells and potentiating neuroprotection in mouse brain.123 Alternatively, hydrolysis with Corolase PP produced anti-inflammatory peptides VDVPVKVPYS and KLPDHPKLPK with NO inhibitory activity (89% and 84%, respectively) from Baijiu vinasse, which is a by-product of sorghum fermentation.124
It also can be highlighted the potential of in vitro gastrointestinal digested foxtail millet protein with pepsin and pancreatin as a source of anti-inflammatory peptides like QNWDFCEAWEPCF and EDDQMDPMAK, which stood out for their effect inhibiting NO, IL-6, and TNF-α production between seven novel peptides identified.125
Mutations of several residues or amino acid sequence modifications of known antimicrobial peptides have been made in order to test the anti-inflammatory potential of these new synthetic peptides since AMPs are key components of the immune system as one of the most important defense mechanisms against bacterial infections in many types of organisms. A defensin-like AMP, named OsDef2, was found in the hemolymph of the tick Ornithodoros savignyi and used as template for the synthesis of peptide Os and its analogue Os-C. Both peptides derived from the C-terminal region of OsDef2 and succesfully inhibited LPS/IFN-γ-induced production of NO and TNF-α by 36% and 25%, respectively.129,130 It seems that C-terminal region of antimicrobial peptides is a good template for the creation of anti-inflammatory peptides since, in addition to peptides Os and Os–C, peptides YKRWKKRWAKYWKKFRK and YKRWKKNWAKYWKIFRK were engineered by introducing changes in C-terminal amino acids of antimicrobial chemokine CXCL14, like the addition of lysine, arginine or tryptophan, and reported almost 100% inhibition of NO production along with TNF-α and IL-6 production decrease.131 Substitutions of other amino acid residues have also been made resulting in several anti-inflammatory peptides such as peptide MC1-1, which was created by glycine and histidine mutations in AMP chensinin-1 derived from Chinese frog (Rana chensinensi) skin secretions or peptide Ps-K18 produced by substituting lysine for leucine at position 18 of AMP pseudin-2 (Ps) from paradoxical frog (Pseudis paradoxa) skin.132,133 Another AMP that has been modified to enhance anti-inflammatory activity is temporin-1TL (TL), in this case 4 analogous peptides TL-1, TL-2, TL-3 and TL-4 were synthesized by substituting tryptophan, arginine and lysine from the original template sequence. It should be noted that synthetic TL analogs improved anti-inflammatory activity compared with TL thanks to the mutations performed.134 Moreover, anti-inflammatory peptides derived from modifications of hormones involved in inflammatatory response, like parathyroid hormone (PTH) and Glucagon-like peptide-1 (GLP-1), have also been reviewed.135 Peptide ETNKV and ETNKVETYKEQPLKTPGKKKKGKPGKRREQEKK are derived from N-terminus (amino acids 107–111) and C-terminus (107–139) of parotid hormone-related protein or PTH-rP (composition extracted from UniProt database for human PTH-rP) and both favored osteoblastic function, although the C-terminal domain was more efficient than N-terminal domain which is in agreement with the previous data for peptides Os, Os–C and CXCL14-derivivatives. Another hormone-derived peptide is Liraglutide, a glucagon-like peptide 1 receptor agonist that also exerted anti-inflammatory and anti-degradative actions in osteoarthritis being suggested as a potencial therapeutic candidate for Osteoarthritis (OA) treatment along with ETNKVETYKEQPLKTPGKKKKGKPGKRREQEKK and its shorter fragment.135,136 Besides, Liraglutide is already commercially available for treatment of type II diabetes under the name Victoza® with a longer half-life compared to endogenous GLP-1.137
Apart from modifying natural peptides, hybridization is also suggested as an effective approach to enhance anti-inflammatory activity. An hybrid peptide named CTP and created by combining the active center of AMP cathelicidin-2 (CATH-2) with thymopentin (TP5), which is the active site of the naturally occurring hormone thymopoietin and showed enhanced anti-inflammatory activity when amine modification was made at C-terminus.138 These results that are in agreement with the previously suggested hypothesis that C-terminus is a region involved in the anti-inflammatory effect. Nevertheless, the hybridization of a peptide derived from the N-terminus of bovine lactoferrin, LfcinB6, with another peptide KR-12-a4 composed of amino acids from the central part of human cathelicin (LL-37) resulted in an anti-inflammatory peptide named Lf-KR, which exerts inhibitory effects against NO and TNF-α production.139 Therefore, it can be argued that the composition of other regions in natural peptides, apart from the C-terminal end, also have an influence on the activity against inflammation.
Finally, in addition to hybridizing or modifying known peptide sequences, designing de novo peptides that meet structural and physicochemical requirements is also a successful strategy for creating anti-inflammatory peptides. As an example, model peptide isomers with simple amino acid composition containing tryptophan and combinations of leucine/lysine residues were designed for the development of therapies against inflammation and one of these peptides, named WALK11.3, reported potent anti-inflammatory activity by inhibiting NO, COX-2 and cytokines TNF-α, IL-6 and IL-1β production.140 Through the design based on structural determinants such as charge or hydrophobicity peptide GW-A2 was created and reported both antimicrobial activity and anti-inflammatory activity, attracting attention for its complete mechanism of action regulating MAPK and NF-κB pathways along with pro-inflammatory cytokines, COX-2 and iNOS expression levels.141
Regarding the bioactivity of these modified or designed synthetic peptides compared with the previously mentioned natural peptides, it could be assumed that “non-natural” peptide sequences present a slight improvement in terms of in vitro activity against inflammation, particularly for the suppression of proinflammatory cytokines TNF-α, IL-6 and IL-1β production and iNOS enzyme inhibition. Supporting this theory it is worth mentioning that synthetic peptide SET-M33D (KKIRVRLSA) was reported to have a very strong potential for protection against in inflammation murine macrophage cells by inhibiting the enzymes iNOS (by 83.3%) and COX-2 (by 84%) as well as significantly reducing cytokines TNF-α (by 95.7%), IL-1β (by 77.6%) among other inflammatory mediators regulation.142,143 This bioactivity improvement may be due to these sequences being specifically designed or mutated to fulfil the purpose of fighting inflammation and they are “upgraded” versions of natural sequences. However, no reported measures have been found for inflammatory mediators such as LOX, IFN-γ, VCAM-1 or IL-10 and their effects in vivo have not been evaluated in many studies. Therefore, the spectrum of action against inflammation for natural peptides is largest since more mechanisms of action have been studied. Nevertheless, in view of the variability of methods and types of measurements, doses, etc., it is very difficult to establish comparisons between the studies.
Mechanism of inhibition | Selection criteria | Threshold | Number of peptides |
---|---|---|---|
NO production | Peptide sequences that report values below the threshold | 30% | 19 |
Nitrite production | Peptide sequences that report values below the threshold | 14 μM | 16 |
PGE2 production | Peptide sequences that report values below the threshold | 50 pg mL−1 | 2 |
TNF-α production | Peptide sequences that report values below the threshold | 80 pg mL−1 | 11 |
IL-1β production | Peptide sequences that report values below the threshold | 99 pg mL−1 | 15 |
IL-6 production | Peptide sequences that report values below the threshold | 131 pg mL−1 | 14 |
IL-8 production | Peptide sequences that report values below the threshold | 50 pg mL−1 | 2 |
IL-10 production | Peptide sequences that report values above the threshold | 800 pg mL−1 | 1 |
TNF-α inhibition | Peptide sequences that report values above the threshold | 65% | 2 |
IL-1β inhibition | Peptide sequences that report values above the threshold | 75% | 3 |
IL-6 inhibition | Peptide sequences that report values above the threshold | 75% | 3 |
iNOS inhibition | Peptide sequences that report values above the threshold | 52% | 5 |
COX-2 inhibition | Peptide sequences that report values above the threshold | 65% | 3 |
5-LOX inhibition | Peptide sequences that report values above the threshold | 50% | 1 |
Protein type | Protein source | Peptide obtention | Peptide sequencea | Regulatory mechanism | Cell type | Ref. |
---|---|---|---|---|---|---|
a Amino acid sequence has been written from the first residue at N-terminus to the last residue at C-terminus (considering C-terminus the end of the peptide amino acid chain). | ||||||
Plant | Corn silk (Zea mays L.) | Enzymatic hydrolysis (trypsin) | TMKLLLVTL (FK2) | Inhibit IL-1β, IKKβ activity, IκB phosphorylation and NF-κB activation | BALB/c mice | 144 |
Animal | Beetle (Allomyrina dichotoma) | Chemical synthesis | AFWCLIRRTVAA (Allomyrinasin) | Inhibit IL-6 and TNF-α production | BALB/c mice and mouse skin infection model | 67 |
Synthetic | Synthetic hybrid | Chemical synthesis | RWGRFLRKIRRFRRKDVY (CTP) | Reduce TNF-α, IL-1β, and IL-6 secretion levels | C57/BL6 mice | 138 |
Animal | Snake venom gland (Hydrophis cyanocinctus) | Chemical synthesis | DEQHLETELHTHLTSVLTANGFQ (Hydrostatin-SN1) | Inhibit TNF-α, IL-6, and IL-1β production | C57BL/6 mice | 91 |
Synthetic | AMP (Chensinin-1) | Chemical synthesis | SAVWRHWRRFWLRKHRKH (MC1-1) | Inhibit TNF-α and IL-6 production | Kunming mice | 132 |
Animal | Earthworm coelomic fluid (Eisenia foetida) | Chemical synthesis | AMADQ | Inhibit TNF-α and COX-2 production, degradation of IκB and MAPK signaling pathway | Mice | 57 |
Plant | Jiuzao (Baijiu vinasse) | Chemical synthesis | AYI | Inhibit TNF-α, IL-1β, IL-6, and NO expressions | Rats | 145 |
Animal | Snake venom (Heloderma suspectum) | Chemical synthesis | HGEGFTSDLSKQMEEEAVRLFIEWLKNGGPSSGAPPPS (Exendin-4) | Reduce IL-1β, IL-6, TNFα, and IFNγ production | Wistar rats | 146 |
One of the animal models in which the anti-inflammatory response to peptide treatment has been studied the most is BALB/c mice. Particularly, protection against inflammation of peptide TMKLLLVTL (FK2) obtained from hydrolysis with trypsin of corn silk and peptide was investigated in BALB/c mice model Besides, in vitro anti-inflammatory potential of peptide Allomyrinasin from Beetle Allomyrina dichotoma was verified in BALB/c mice together with a mouse skin infection in vivo study using Staphylococcus pseudintermedius bacteria.67,144
Another widely used strain in biomedical research to study the anti-inflammatory activity of peptides is mice strain C57BL/6, which has been used to show that apart from having in vitro activity, peptides CTP (cathelicidin-2 and thymopentin hybrid) and Hydrostatin-SN1 from sea snake venom gland also exerted anti-inflamatory activity in vivo.91,138 However, these are not the only animal models used to investigate anti-inflammatory effects in vivo. For example, another powerful anti-inflammatory peptide named MC1-1 with in vitro effects, was tested in a different animal model of the two named above, Kunming mice.132
Aside from mouse, rat models have also been used to test anti-inflammatory activity. Treatment with Exendin-4, a GLP-1 receptor agonist purified from snake Heloderma suspectum venom, significantly attenuated inflammation in a LPS-induced rat model of inflammation and the potential mechanism of action of tripeptide AYI, isolated from a byproduct of baijiu distillation, was also investigated using a rat model.145,146
Interestingly, peptides CTP, Hydrostatin-SN1, AYI and Exedin-4, obtained from completely different protein sources, seem to share a mechanism of action consisting of reducing proinflammatory cytokines TNF-α, IL-1β, and IL-6 levels, although it can not be assumed that this is their only mechanism of action since they may have effects on other inflammatory mediators, but no trials have been carried out in this regard. In the same way, peptides Allomyrinasin and MC1-1 share in vivo effect against the cytokines IL-6 and TNF-α, but in this case the protein source is similar, even though MC1-1 is a modification of Chensinin-1 peptide found in Chinese brown frog and Allomyrinasin peptide is found naturally in a beetle, both are derived directly or indirectly from animal sources. On the contrary, peptide FK2 derived from a plant protein source acts in a different way, inhibiting compounds and mediators involved in NF-κB pathway.
Despite the fact that several studies investigated the effects of anti-inflammatory peptides in vivo, more research on animal models is needed before applications to human health.
Protein source | Enzyme treatment | Regulatory mechanism | Study type | Ref. |
---|---|---|---|---|
Tuna cooking juice (Thunnini) | Alcalase | Suppress TNF-α, IFN-y and IL-2 expression | In vitro (RAW 264.7) | 25 |
Skipjack tuna dark muscle (Katsuwonus pelamis) | Alcalase | Reduce TNF-α, IL-6 and IL-1β secretion and inhibit NO production | In vitro (RAW 264.7) | 148 |
Milk | Alcalase | Suppress TNF-α and IL-1β gene expression | In vitro (RAW 264.7) | 151 |
Bovine bone-gelatin (Bos Taurus) | Alcalase | Reduce IL-6, NO and TNF-α release and in serum, suppress TNF-α, IL-6 and IL-1β production along with decreasing COX-2 activation | In vitro (RAW264.7) | 152 |
In vivo (C57BL/6 mice) | ||||
Egg | Alcalase | Inhibited TNF-α activation | In vitro (HDFs) | 153 |
Alcalase and pepsin | Inhibit NO, PGE2, TNF-α, IL-1β and IL-6 production and iNOS and COX-2 expression | In vitro (RAW 264.7) | 154 | |
Sandfish (Arctoscopus japonicus) | Alcalase and collupulin MG | Inhibit NO production | In vitro (RAW 264.7) | 147 |
Milk | Bacterial food-grade enzyme (unknown) | ReducedIL-1α, IL-1β, IL-8 and TGF-β expression and increase IL-17 expression | In vitro (Caco-2) | 24 |
Ex vivo (Porcine colonic tissues) | ||||
Sea cucumber (Actinopyga lecanora) | Bromelain | Inhibit NO production | In vitro (RAW 264.7) | 99 |
Chicken (Gallus gallus domesticus) | Corolase PP, Protamex | Reduce IL-1β, IFN-γ, TNF-α, IL-1α, IL-2, IL-6, IL-10 and MCP-1 levels in plasma | In vivo (C57BL/6 mice) | 159 |
Sea cucumber (Stichopus Japonicus) | Flavourzyme | Suppress IL-6, TNF-α and IL-1β mRNA expression, phosphorylation of JNK, ERK and p38 and inhibit degradation of IκBα and nuclear transposition of NF-κB p65 | In vitro (RAW 264.7) | 189 |
Milk | Flavourzyme | Reduce NO production and synthesis of TNF-α and IL-1β | In vitro (RAW 264.7) | 190 |
Silkworm pupae (Bombyx mori) | Flavourzyme and alcalase | Inhibit NO production | In vitro (RAW 264.7) | 150 |
Bee pollen (Apis mellifera) | Neutrase | Suppress COX-2, NO, iNOS, IL-6 and TNF-α production | In vitro (RAW 264.7) | 191 |
Milk | Neutrase and Protamex | Inhibit production of NO and reduced IL-1α, IL-6, and TNF-α production | In vitro (RAW 264.7) | 192 |
Salmon bones (Salmo salar) | Papain | Inhibit NO production and reduced iNOS, IL-6, TNF-α and COX-2 mRNA levels | In vitro (RAW 264.7) | 101 |
Crocodile hemoglobin (Crocodylus siamensis) | Pepsin | Reduce NO, IL-6, IL-1β, and PGE2 production | In vitro (RAW 264.7) | 156 |
Milkfish (Chanos chanos) | Pepsin | Reduce LOX activity and NO production | In vitro (assay) | 155 |
Mussel (Mytilus edulis) | Pepsin | Inhibit translocation of NF-κB through the prevention of IκB phosphorylation and degradation, NO and PGE2 production, iNOS and COX-2 protein and gene expressions and reduce IL-1β, IL-6 and TNF-α secretions and also inhibit the MAPK signaling pathway | In vitro (RAW 264.7) | 193 |
Sturgeon muscle (Acipenser schrenckii) | Pepsin | Reduce NO, IL-6, TNF-α and IL-1β production and suppress the expression of MAPK, IκB, and NF-κB p65 | In vitro (RAW 264.7) | 98 |
Skipjack tuna (Katsuwonus pelamis) | Pepsin and animal protease (unknown) | Reduce IL-6 and TNF-α expression | In vivo (BALB/c mice) | 149 |
Oyster soft tissue (Crassostrea talienwhanensis) | Pepsin, trypsin and Maxipro | Suppress TNF-α, I L-1β, IL-6 and i-NOS expression | In vitro (RAW 264.7) | 158 |
Chum salmon (Oncorhynchus keta) | Pepsin and trypsin | Reduce NO, IL-6 and TNF-α secretions, as well as TNF-α, IL-6, iNOS and COX-2 mRNA expression | In vitro (RAW 264.7) | 157 |
In vivo (C57BL/6 mice) | ||||
Cricket (Gryllodes sigillatus) | Alcalase, pepsin and pancreatin (GID) | Inhibit expression of NF-κB | In vitro (RAW 264.7) | 150 |
Sardine (Sardina pilchardus) | Brewer's spent yeast (BSY) proteases, pepsin and trypsin (GID) | Inhibit IL-8 and ICAM-1 secretion | In vitro (EA.hy926/Caco-2 co-culture) | 160 |
Hen spent muscle (Gallus gallus domesticus) | Protease M, Protex 50FP | Inhibit IL-6 production | In vitro (U937) | 108 |
Egg | Trypsin | Inhibit NO and iNOS production and reduce the phosphorylation levels of JNK and ERK | In vitro (RAW 264.7) | 194 |
Sturgeon cartilage (Acipenser schrenckii) | Trypsin and papain | Inhibit NO production and reduced IL-6 levels | In vitro (RAW 264.7) | 174 |
Milk | Virgibacillus halodenitrificans SK1-3-7 proteinase | Suppress IL-1β, IL-6, IL-8, TNF-α and COX-2 production | In vitro (THP-1) | 195 |
To produce hydrolysates with bioactivity against inflammation, Alcalase has proved to be among the most efficient proteases for these protein sources. This protease has produced hydrolysates with significant anti-inflammatory activity from egg and dairy products as well as from terrestrial and marine animals (Table 6). An experiment on sandfish (Arctoscopus japonicus) meat and roe reported that products from hydrolysis with Alcalase and Copulline MG separately showed higher NO inhibitory activity in vitro than the ones produced by Flavourzyme, Neutrase and Protamex.147 Moreover tuna cooking juice the hydrolyzate with alcalase exhibited a more powerful inhibitory effect on inflammatory mediators such as TNF-α and IFN-y in murine macrophage RAW 264.7 calls than the hydrolysates produced with Orientase and Flavourzyme.25 These results follow the line of those reported for hydrolysate of Skipjack tuna (Katsuwonus pelamis) dark muscle where Alcalase produced a hydrolyzate with better anti-inflammatory activity in terms of inhibiting TNF-α, IL-6, IL-1β and NO (IC50 > 45.44 mg mL−1) compared to Flavourzyme hydrolyzate in vitro.148 Nonetheless, a hydrolyzate of Skipjack tuna has also been produced with pepsin in combination with an animal protease of unspecified origin and showed good anti-inflammatory activity against TNF-α, IL-6, IL-1β and NO production in vivo.149 Therefore, two assumptions can be drawn from these data. First one is that the mechanism of action of a hydrolysate seems to be more influenced by the origin of the protein source than by the enzyme used, since for Alcalase hydrolysates from different fish species the anti-inflammatory effect reported is not the same, and in order to add extra support for this hypothesis, the same substrate, SKipjack tuna, treated with different enzymes acts by the same mechanism both in vitro and in vivo, reducing the expression of TNF-α, IL-6, IL-1β and NO. Second one is that particularly compared to Flavourzyme, Alcalase leads to better results in terms of production of hydrolysates with anti-inflammatory potential, but this refers to the use of these enzymes individually and whether the combination of both enzymes could give even better results has not been mentioned. This last question has been explored for edible insect Silkworm pupae and the results provided showed that treatment with Alcalase individually produced lower NO production values, presenting greater NO inhibition activity than treatment with Flavourzyme/Alcalase in combination.150 However, Flavourzyme hydrolysis of sea cucumber (Stichopus Japonicus) resulted in a pool of peptides reporting anti-inflammatory effect.99 Besides, thanks to Alcalase broad selectivity and specificity permit the use of a wide variety of protein substrates such as milk β-Lactoglobulin, egg ovomucin and livetins or bovine bone-gelatin apart from those already mentioned above.151–154
Pepsin, used individually and not in combination with pancreatin and other enzymes to mimic human gastrointestinal digestion as occurs in many anti-inflammatory peptides studies, also leads to protein hydrolysates with anti-inflammatory activity. Interestingly, the reviewed hydrolysates exhibiting in vitro anti-inflammatory capacity that are produced with pepsin are from diverse marine animal sources such as milkfish (Chanos chanos) or crocodile hemoglobin where two potential peptides SAFNPHEKQ (SQ9) and IIHNEKVQAHGKKVL (IL15) were also identified.155,156 Pepsin has also been used in combination with trypsin, reporting good results against inflammation for hydrolysate of Chum salmon (Oncorhynchus keta) myofibrillar protein (Mf) in vitro and in vivo.157 However, for Oyster (Crassostrea talienwhanensis) soft tissue, pepsin hydrolysate did not give as good results against inflammation as trypsin hydrolysate in vitro.158 Regarding the mechanism of action, since both pepsin hydrolysates act by different mechanisms, the first reducing LOX activity and NO production and the second reducing IL-6, IL-1β, NO and PGE2 production, we can continue with the previous assumption where we reasoned that the mechanism of action of a hydrolyzate can not be attributed directly to the enzyme used for its production.
Furthermore, different enzymes such as Corolase PP or Protamex have also produced hydrolysates with strong anti-inflammatory effect in vivo, even showing greater pro-inflammatory cytokine-lowering potential individually with respect to the mixture of papain/bromelain or single Alcalase in chicken protein hydrolysates, although these last two treatments also reported anti-inflammatory activities.159 Besides, simulated gastrointestinal digestion with combinations of pepsin/pancreatin or pepsin/trypsin has also reported good results in different types of protein sources after previous hydrolysis with other proteases, as is the case of Sardine (Sardina pilchardus) which decreases IL-8, ICAM-1 and NO levels in EA.hy926/Caco-2 cell co-culture or Cricket (Gryllodes sigillatus) where three novel anti-inflammatory peptides YKPRP, PHGAP and VGPPQ were identified from the bioactive hydrolysate that reported in vitro inhibition of NF-κB expression.150,160 Interestingly, apart from studies of bioactivity of hydrolysates in vitro and in vivo, the anti-inflammatory potential of hydrolyzed milk-casein with a bacterial food-grade enzyme (unknown) has been tested also in ex vivo models of the gastrointestinal tract, using porcine colonic tissue.24
Protein source | Enzyme treatment | Regulatory mechanism | Study type | Ref. |
---|---|---|---|---|
White sorghum grain (Sorghum bicolor) | Alcalase | Reduce IL-1β, IFN-γ and TNF-α levels | In vitro (Human skin cultures) | 164 |
Lupin (Lupinus angustifolius L.) | Alcalase | Reduce TNF-α, IL-1β, and IL-6 levels and promote expression of IL-10 | In vitro (Caco-2/THP-1 coculture) | 162 |
Pigeon pea (Cajanus cajan) | Alcalase and bromelain | Inhibit NO production | In vitro (RAW 264.7) | 161 |
Lentil (Lens culinaris) | ||||
Chickpea (Cicer arietinum) | ||||
Pinto bean (Phaseolus vulgaris L.) | Alcalase and savinase | Inhibit IL-6 secretion | In vitro (CD-18Co) | 163 |
Lychee seeds (Litchi chinensis Sonn.) | Neutrase | Inhibit NO production and reduce iNOS and IL-6 expression | In vitro (RAW 264.7) | 127 |
Corn silk (Zea mays L.) | Trypsin | Inhibit IL-1β, NF-κB activities, IKKβ activities, IκB phosphorylation and NF-κB activation | In vivo (BALB/c mice) | 144 |
Rice (Oryza sativa) | Trypsin | Inhibit NO and TNF-α release, TNF-α, iNOS, IL-6, IL-1β transcription and repress NF-κB pathway by impeding the nuclear translocation of p65 | In vitro (RAW 264.7) | 165 |
Microalgae (Synechococcus sp. VDW) | Trypsin | Reduce iNOS, TNF-α, COX-2, and IL-6 gene expression | In vitro (RAW 264.7) | 166 |
Pea (Pisum Sativum) | Virgibacillus halodenitrificans SK1-3-7 proteinase | Suppress IL-1β, IL-6, IL-8, TNF-α and COX-2 production | In vitro (THP-1) | 194 |
Quinoa (Chenopodium quinoa Willd.) | Papain, pepsin and pancreatin | Inhibit NO production | In vitro (RAW 264.7) | 169 |
Rice bran (Oryza sativa) | Alcalase, pepsin and pancreatin (GID) | Suppress iNOS, IL-6 and TNF-α mRNA levels and inhibited NO production | In vitro (RAW 264.7) | 112 |
Millet grain (Panicum miliaceum L.) | α-Amylase, pepsin and pancreatin (GID) | Inhibit COX-1, COX-2 and LOX activity | In vitro (assay) | 78 |
Chia seeds (Salvia hispanica L.) | Pepsin and pancreatin (GID) | Reduce NO, IL-1β, IL-6, and TNF-α levels and increase levels of IL-10 | In vivo (BALB/c mice) | 167 |
In vitro (peritoneal murine macrophages) | ||||
Common bean (Phaseolus vulgaris L.) | Pepsin and pancreatin (GID) | Reduce TNF-α, IL-1β and PGE-2 production | In vitro (THP-1) | 195 |
Soybean (Glycine max) | Pepsin and pancreatin (GID) | Inhibit NO and PGE2 production | In vitro (RAW 264.7, Caco-2, HT-29 and HCT-116) | 182 |
Chickpea (Cicer arietinum) | Pepsin and pancreatin (GID) | Inhibit NF-κB expression | In vitro (RAW 264.7) | 196 |
Chia seed (Salvia hispanica L.) | Pepsin and pancreatin (GID) | Inhibit 5-LOX, COX-1-2, and iNOS production | In vitro (assay) | 77 |
Amaranth flour (Amaranthus hypochondriacus) | Pepsin and pancreatin (GID) | Inhibit NO production | In vitro (RAW 264.7) | 168 |
As observed for animal protein hydrolysates, Alcalase remains one of the most used enzymes in terms of searching for anti-inflammatory activity in vitro for plant protein hydrolysates. Particularly, Alcalase showed potential in the production of hydrolysates with NO inhibitory properties from legume proteins such as pigeon pea (Cajanus cajan), lentil (Lens culinaris) or chickpea (Cicer arietinum).161 Nevertheless, it can not be assumed that this enzyme is responsible for this mechanism of action since for other legumes hydrolysates such as lupin (Lupinus angustifolius L.) or pinto bean (Phaseolus vulgaris L.) activity has been measured through other inflammatory markers and the effect showed is different, decreasing TNF-α, IL-1β, and IL-6 levels by 70, 40, and 45% (respectively) in the first case and inhibiting IL-6 secretion by 28% in the second case.162,163 Furthermore, anti-inflammatory peptides in lupin protein hydrolysate were able to cross Caco-2 monolayers successfully without any modifications in their bioactivity.162 Thus, it can be deduced that they are potentially resistant to the gastrointestinal tract and may reach the bloodstream to exert their beneficial effect. Despite this, further investigation is required to confirm these properties in vivo and apply this information to a tissue and organ level like as has been done in an innovative study which determined the protective anti-inflammatory effects of Alcalase-derived peptide extracts of white sorghum against the damage induced on human skin by the exposure to ultraviolet-B irradiation (UVB).164
In addition to the use of Alcalase, the use of trypsin has also resulted in hydrolysates of plant protein from gramineae family (known as grasses) displaying anti-inflammatory abilities by NF-κB signaling pathway regulation. Trypsin hydrolysates from rice and corn silk protein inhibited inflammatory response in vitro and in vivo, respectively.144,165 Moreover, a recent study reported the potential used of peptides derived from trypsin hydrolysis of microalgae (Synechococcus sp.) to develop natural anti-inflammation food-grade ingredients, drugs, and cosmetic products.166 The authors recovered some active fractions, reporting an IC50 value of 34.51 μg mL−1 for NO inhibitory activity but due to their interest for biotechnological applications more studies on anti-inflammatory peptides from microalgae should be done.
Apart from the use of enzymes individually, pepsin and pancreatin have been used in combination to simulate the physiological process of digestion providing hydrolysates from different plant sources with anti-inflammatory potential. To this regard, peptides fractions generated from in vitro gastrointestinal digestion of chia seeds showed high inhibitory activity against 5-LOX, COX-1-2 and iNOS pro-inflammatory enzymes, and its bioactivity was also validated in vivo murine models.77 Similarly, gastrointestinal digested peptide fractions generated from heat treated millet grains had high COX-1 and COX-2 inhibitory activity, IC50 value of 0.08 and 0.12 mg mL−1, along with LOX inhibitory activity, IC50 value of 0.14 mg mL−1in vitro.78 Additionally, in vitro digestion of chia seeds in another study resulted in peptide fractions exerting anti-inflammatory effect through the regulation of proinflammatory cytokines such as IL-1β, IL-6, TNF-α and IL-10.167 Therefore, bioactive peptides found in chia seeds hydrolysate may have different mechanisms of action against inflammation, but in order to study their individual effect these sequences, such as peptide TGPSPTAGPPAPGGGTH, should be isolated and purified. In contrast, the study of the anti-inflammatory activity of the Amaranthaceae family of plants, such as quinoa or amaranth, has resulted in hydrolysates with an inhibitory effect on NO production in murine macrophages in vitro, but their activity against other proinflammatory markers has not been investigated.168,169
Regarding the bioactivity of these protein hydrolysates, compared with single peptides from natural and “non-natural” origin, they may display better anti-inflammatory effect than single peptides giving understanding that possible synergies in the peptide pool could result in increased values of bioactivity. For example, hydrolysate fractions from locust Schistocerca gregaria protein preparation showed higher inhibitory potential for COX-2 and LOX, IC50 values between 0.13–0.26 μg mL−1, than the individual peptide FDPFPK identified in the hydrolysate fraction of the same insect, which reported IC50 values of 7.40 and 2.85 mg mL−1 for COX-2 and LOX inhibition, respectively.62 On the contrary, there are also cases in which the purified peptides exhibit greater bioactivity than the hydrolyzed proteins such as peptides SNKGGGRPN and TVTVYSLLR from salmon bone hydrolysis which were found to have a marked NO-inhibitory activity compared to Skipjack tuna dark muscle hydrolysate, reporting salmon peptides an IC50 value of 2.56 μg mL−1 for this inflamamtory marker, while tuna hydrolysate reported an IC50 value around 45.44 mg mL−1.101,148
Peptide sequence | M w (Da) | pI | Net charge (pH 7) | Water solubility | PCL | Hydrophobicity ratio (%) | Positively charged AA (%) | Polar AA (%) | Aromatic AA (%) | P (%) | G (%) | Q (%) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
M w: molecular weight. pI: isoelectric point. PCL: peptide chain length. P: presence of proline residue. G: presence of glycine residue. Q: presence of glutamine residue. | ||||||||||||
PAY | 349.38 | 3.85 | 0.00 | Poor | 3 | 100% | 0% | 0% | 33% | 33% | 0% | 0% |
VDVPVKVPYS | 1102.28 | 6.55 | 0 | Good | 10 | 70% | 10% | 30% | 10% | 20% | 0% | 0% |
VHYAGTVDY | 1024.08 | 4.87 | −0.9 | Poor | 9 | 56% | 11% | 33% | 22% | 0% | 11% | 0% |
GAKYAKIIYNYLKKIANALW | 2341.79 | 10.33 | 4.00 | Poor | 20 | 55% | 20% | 30% | 20% | 0% | 5% | 0% |
QA | 217.22 | 3.34 | 0 | Good | 2 | 50% | 0% | 50% | 0% | 0% | 0% | 50% |
KA | 217.27 | 9.91 | 1 | Good | 2 | 50% | 50% | 50% | 0% | 0% | 0% | 0% |
WG | 261.28 | 3.50 | 0 | Poor | 2 | 50% | 0% | 0% | 50% | 0% | 50% | 0% |
SNPSVAGVR | 885.97 | 10.57 | 1 | Good | 9 | 44% | 11% | 44% | 0% | 11% | 11% | 0% |
SIFGKIFKRIIRVAWK | 1962.43 | 12.18 | 5 | Good | 16 | 44% | 31% | 38% | 19% | 0% | 6% | 0% |
YKRWKKNWAKYWKIFRK | 2429.91 | 11.30 | 8 | Good | 17 | 41% | 47% | 53% | 35% | 0% | 0% | 0% |
RGQANILAGKNIKIRSGAAAGVGKTPQKANVEVLALGIW | 3971.61 | 11.57 | 5.00 | Good | 39 | 41% | 15% | 36% | 3% | 3% | 15% | 5% |
EDDQMDPMAK | 1179.28 | 3.32 | −3 | Good | 10 | 40% | 10% | 60% | 0% | 10% | 0% | 10% |
YKRWKKRWAKYWKKFRK | 2487.01 | 11.63 | 10 | Good | 17 | 35% | 59% | 59% | 35% | 0% | 0% | 0% |
FDPFPK | 749.85 | 6.39 | 0 | Good | 6 | 33% | 17% | 33% | 33% | 33% | 0% | 0% |
KKIRVRLSA | 1070.33 | 12.16 | 4 | Good | 9 | 33% | 44% | 56% | 0% | 0% | 0% | 0% |
QNWDFCEAWEPCF | 1674.81 | 0.65 | −3.1 | Poor | 13 | 31% | 0% | 54% | 31% | 8% | 0% | 8% |
KLPDHPKLPK | 1172.42 | 10.40 | 2.1 | Good | 10 | 30% | 40% | 50% | 0% | 30% | 0% | 0% |
KIWHHTF | 968.11 | 9.91 | 1.2 | Poor | 7 | 29% | 43% | 57% | 29% | 0% | 0% | 0% |
HAEGTFTSDVSSYLEGQAAKEFI | 2487.63 | 4.04 | −2.9 | Good | 23 | 26% | 9% | 52% | 13% | 0% | 9% | 4% |
HYGH | 512.52 | 7.70 | 0.2 | Poor | 4 | 25% | 50% | 50% | 25% | 0% | 25% | 0% |
GPETAFLR | 889.99 | 6.86 | 0 | Good | 8 | 25% | 13% | 38% | 13% | 13% | 13% | 0% |
QITITVKPRFRRIKRLFRGFR | 2689.26 | 12.81 | 8 | Good | 21 | 24% | 38% | 52% | 14% | 5% | 5% | 5% |
GKLTKDKLKRGAKKALNVASKV | 2353.85 | 11.38 | 7.00 | Good | 22 | 23% | 36% | 55% | 0% | 0% | 9% | 0% |
KGIRGYKGGYCKGAFKQTCKCY | 2459.92 | 9.99 | 5.80 | Good | 22 | 23% | 27% | 50% | 18% | 0% | 23% | 5% |
YGGGGE | 538.51 | 1.00 | −1 | Good | 6 | 17% | 0% | 17% | 17% | 0% | 67% | 0% |
HLDDALRGQE | 1153.20 | 4.16 | −1.9 | Good | 10 | 10% | 20% | 60% | 0% | 0% | 10% | 10% |
LKWLKKLLKKL | 1410.87 | 11.28 | 5.00 | Good | 11 | 9% | 45% | 45% | 9% | 0% | 0% | 0% |
DEQHLETELHTHLTSVLTANGFQ | 2620.78 | 4.49 | −3.7 | Good | 23 | 9% | 13% | 65% | 4% | 0% | 4% | 9% |
HC | 258.30 | 7.06 | 0 | Poor | 2 | 0% | 50% | 100% | 0% | 0% | 0% | 0% |
CR | 277.35 | 9.21 | 0.9 | Good | 2 | 0% | 50% | 100% | 0% | 0% | 0% | 0% |
SEGGFLE | 737.75 | 0.85 | −2 | Good | 7 | 0% | 0% | 43% | 14% | 0% | 29% | 0% |
It has been reported that the molecular weight and the length of the peptide chain (PCL) play an important role in their bioactivity. Out of the 31 peptides analyzed, 6 of them are di- or tri-peptides and 19 have a length between 4 and 20 amino acids, showing that bioactivity seems to be greater in smaller peptides with short amino acid sequences. This previous argument is also reflected in the molecular weight data where about 42% of the most bioactive peptides reviewed have a molecular weight below 1000 Da and the remaining does not exceed 3 kDa, except for one of them that almost reaches 4 kDa. Although bioactive peptides are generally short sequences of up to 20 residues, longer peptides have also been reported to exert anti-inflammatory activities like lunasin from defatted soybean flour or the peptides HCRG1 and HCRG2 from sea anemone (Heteractis crispa), both containing 56 amino acid residues.90,115 Results in literature support that both small (2–3 amino acids) and large (up to 50 amino acids) peptides can be absorbed intact through the intestine and produce biologic effects at the tissue level but the potency of the enterally administered peptides decreases as the chain length increases.170
Several investigations point out the role of positively charged residues on the anti-inflammatory response of bioactive peptides.59,156,171 The positively charged region of the peptide may act as a chemokine, so the peptides modulate the immune response through their union with respective chemokine receptors.172 In our case, the presence of specific amino acids attached to the N-terminal end was studied, being the most abundant arginine (R) and lysine (K), two basic amino acids with net positive charge at physiological pH. This feature has been reported for human lactoferricin or tuna juice hydrolysates, both presenting Arg residues at terminal position, and reported to block inflammation by binding to the popolysaccharides excreted by bacteria and macrophages.25,173 Almost half of the biopeptides analyzed in this work contain between 25–50% of positive amino acids in their sequence (i.e. lysine, arginine and histidine). Moreover, the majority of the peptides present a net positive charge at physiological pH. Another characteristic shared by anti-inflammatory peptides is the presence of polar amino acids at the C-terminus.30,135,138 According to our analysis on the peptides reviewed, 83% of the sequences have within their composition 25–75% of polar amino acids. Besides, 23% peptides reviewed have a polar residue at the C-terminus, such as lysine or histidine, indicating that this property may affect their anti-inflammatory response.
Furthermore, many studies have highlighted the importance of the presence of hydrophobic amino acids in biopeptides, especially at their N-terminus, in relation to their regulation of the inflammatory response.30,59,113,174 The presence of hydrophobic amino acids residues (i.e. Val, Leu, Ile, etc.) may improve the interaction between peptides and cell membrane, promoting their anti-inflammatory effect. This feature agrees with the peptide sequences shown in Table 8, where 19 out of the 31 peptides analyzed contain between 25 and 100% hydrophobic amino acids. The most abundant hydrophobic amino acid was alanine, followed by isoleucine. This was the case for the tripeptide PAY (100% hydrophobicity) from salmon and the peptide VDVPVKVPYS (70% hydrophobicity) from fermented sorghum, both displaying inhibitory activity against NO production.60,124 Otherwise, no obvious relationship has been found between the solubility of the peptides studied and their potential bioactivity. Despite the fact that short peptides (4–5 amino acid residues) are generally soluble in water, some of the anti-inflammatory dipeptides reviewed show poor solubility due to the presence of hydrophobic amino acids in their sequence.175–177
Other feature linked to anti-inflammatory response is the presence of aromatic amino acids in the peptide sequence.85,178 Our analysis indicates that 55% of the sequences are composed of between 10% and 50% aromatic amino acids such as tyrosine, tryptophan or phenilalanine.100 The anti-inflammatory response of peptides containing aromatic amino acids may be explained by their less susceptibility to intestinal peptidases and proteases. For instance, the ovotransferrin-derived peptide IRW presenting tryptophan at C-terminus, exhibited limited degradation by peptidases and proteases hydrolysis.179 Recent studies report that peptides rich in proline, glycine or glutamine have a great potential against inflammation response.30 For instance, milk-derived tripeptides IPP and VPP, both recognized as potent anti-inflammatory agents, are rich in proline residues.180 To this regard, the presence of proline has been related to less amenable gastro-intestinal digestion and more likely absorption upon oral consumption, a factor that can be beneficial for future therapeutic applications.181 This hypothesis has been confirmed by the high content of glycine in anti-inflammatory peptides from sturgeon cartilage, walnut or germinated soybean.128,174,182 Since the presence of proline, glycine, and glutamine could enhance peptide's anti-inflammatory potencial, the amount of these three residues was studied in the reviewed peptides. The sorghum peptide with the sequence KLPDHPKLPK stands out from the rest of the studied peptides due to the proline composition of its sequence, represented by 30%. The rest of biopeptides do not have a significant amount of proline in their sequences. Soybean peptide YGGGGE its composed by a 67% of glycine. In contrast, the rest of the peptides do not show high contents of this amino acid. Regarding glutamine, it has not been seen that its presence is relevant in the peptides analyzed in this study. From these data a clear relationship between the presence of proline, glycine or glutamine in the peptide sequence and anti-inflammatory activity cannot be determined.
Protein type | Protein source | Sequence | New sources (BLASTp) | Hits |
---|---|---|---|---|
Synthetic | Designed | KKIRVRLSA (SET-M33D) | Lupinus albus | 21 |
Cicer arietinum | 13 | |||
Olea europaea subsp. Europaea | 12 | |||
Helianthus annuus | 80 | |||
Tenebrio molitor | 7 | |||
Designed | LKWLKKLLKKL (WALK11.3) | Cicer arietinum | 8 | |
Lupinus albus | 26 | |||
Olea europaea subsp. Europaea | 24 | |||
Helianthus annuus | 35 | |||
Tenebrio molitor | 11 | |||
Designed | GAKYAKIIYNYLKKIANALW (GW-A2) | Helianthus annuus | 28 | |
Olea europaea subsp. Europaea | 10 | |||
Lupinus albus | 19 | |||
Cicer arietinum | 7 | |||
Tenebrio molitor | 29 | |||
Glucagon-like peptide-1 (GLP-1) | HAEGTFTSDVSSYLEGQAAKEFI (Liraglutide) | Cicer arietinum | 13 | |
Pisum sativum | 4 | |||
Lupinus albus | 25 | |||
Olea europaea subsp. Europaea | 24 | |||
Helianthus annuus | 29 | |||
Tenebrio molitor | 10 | |||
Chemokine CXCL14 | YKRWKKRWAKYWKKFRK (CXCL14-C17-a3) | Helianthus annuus | 65 | |
Olea europaea subsp. Europaea | 12 | |||
Cicer arietinum | 9 | |||
Lupinus albus | 26 | |||
Tenebrio molitor | 8 | |||
Chemokine CXCL14 | YKRWKKNWAKYWKIFRK (CXCL14-C17-a2) | Lupinus albus | 21 | |
Cicer arietinum | 16 | |||
Helianthus annuus | 52 | |||
Olea europaea subsp. Europaea | 11 | |||
Tenebrio molitor | 20 | |||
Tick defensin OsDef1 | KGIRGYKGGYCKGAFKQTCKCY (Os) | Cicer arietinum | 26 | |
Lupinus albus | 28 | |||
Pisum sativum | 24 | |||
Helianthus annuus | 19 | |||
Tenebrio molitor | 17 | |||
Chicken cathelicidin-2 (CATH-2) | QITITVKPRFRRIKRLFRGFR | Tenebrio molitor | 11 | |
Olea europaea subsp. Europaea | 21 | |||
Helianthus annuus | 48 | |||
Lupinus albus | 12 | |||
Cicer arietinum | 13 | |||
Pisum sativum | 3 | |||
Plant | Lupin seeds (Lupinus angustifolius L.) | GPETAFLR | Cicer arietinum | 14 |
Pisum sativum | 1 | |||
Lupinus albus | 17 | |||
Helianthus annuus | 33 | |||
Olea europaea subsp. Europaea | 21 | |||
Tenebrio molitor | 22 | |||
Soybean (Glycine max) | YGGGGE | Tenebrio molitor | 20 | |
Helianthus annuus | 36 | |||
Olea europaea subsp. Europaea | 23 | |||
Cicer arietinum | 6 | |||
Lupinus albus | 15 | |||
Soybean (Glycine max) | SEGGFLE | Helianthus annuus | 31 | |
Olea europaea subsp. Europaea | 35 | |||
Lupinus albus | 15 | |||
Cicer arietinum | 13 | |||
Tenebrio molitor | 6 | |||
Foxtail Millet (Setaria italica) | EDDQMDPMAK | Tenebrio molitor | 13 | |
Olea europaea subsp. Europaea | 28 | |||
Helianthus annuus | 48 | |||
Cicer arietinum | 6 | |||
Lupinus albus | 8 | |||
Pisum sativum | 5 | |||
Foxtail Millet (Setaria italica) | QNWDFCEAWEPCF | Helianthus annuus | 50 | |
Olea europaea subsp. Europaea | 26 | |||
Cicer arietinum | 9 | |||
Lupinus albus | 14 | |||
Tenebrio molitor | 10 | |||
Fermented sorghum (Baijiu vinasse) | KLPDHPKLPK (VPH-1) | Cicer arietinum | 9 | |
Pisum sativum | 2 | |||
Lupinus albus | 22 | |||
Helianthus annuus | 45 | |||
Olea europaea subsp. Europaea | 22 | |||
Tenebrio molitor | 8 | |||
Fermented sorghum, Baijiu vinasse | VDVPVKVPYS | Lupinus albus | 28 | |
Cicer arietinum | 5 | |||
Olea europaea subsp. Europaea | 13 | |||
Helianthus annuus | 37 | |||
Tenebrio molitor | 27 | |||
Animal | Sturgeon muscle (Acipenseridae) | VHYAGTVDY | Tenebrio molitor | 14 |
Helianthus annuus | 40 | |||
Lupinus albus | 14 | |||
Cicer arietinum | 15 | |||
Olea europaea subsp. Europaea | 23 | |||
Sturgeon muscle (Acipenseridae) | KIWHHTF | Tenebrio molitor | 10 | |
Lupinus albus | 37 | |||
Pisum sativum | 15 | |||
Olea europaea subsp. Europaea | 29 | |||
Helianthus annuus | 39 | |||
Cicer arietinum | 16 | |||
Sturgeon muscle (Acipenseridae) | HLDDALRGQE | Tenebrio molitor | 21 | |
Lupinus albus | 24 | |||
Helianthus annuus | 31 | |||
Olea europaea subsp. Europaea | 14 | |||
Cicer arietinum | 5 | |||
Human (Homo sapiens) | SIFGKIFKRIIRVAWK (Hs02) | Tenebrio molitor | 27 | |
Lupinus albus | 18 | |||
Cicer arietinum | 16 | |||
Helianthus annuus | 26 | |||
Olea europaea subsp. Europaea | 20 | |||
Pisum sativum | 1 | |||
Salmon skin hydrolysates (Salmo salar) | QA | No significant similarity found | ||
Salmon skin hydrolysates (Salmo salar) | KA | No significant similarity found | ||
Chinese scorpion venom (Mesobuthus martensii) | HYGH | Tenebrio molitor | 11 | |
Lupinus albus | 28 | |||
Helianthus annuus | 33 | |||
Cicer arietinum | 19 | |||
Olea europaea subsp. Europaea | 24 | |||
Snake venom gland (Hydrophis cyanocinctus) | DEQHLETELHTHLTSVLTANGFQ (H-SN1) | Helianthus annuus | 29 | |
Olea europaea subsp. Europaea | 29 | |||
Lupinus albus | 10 | |||
Cicer arietinum | 4 | |||
Tenebrio molitor | 11 | |||
Black fly salivary glands (Simulium bannaense) | GKLTKDKLKRGAKKALNVASKV (SibaCec) | Helianthus annuus | 14 | |
Olea europaea subsp. Europaea | 29 | |||
Cicer arietinum | 13 | |||
Lupinus albus | 18 | |||
Tenebrio molitor | 14 | |||
Horsefly salivary glands (Tabanus yao) | RGQANILAGKNIKIRSGAAAGVGKTPQKANVEVLALGIW (Cecropin-TY1) | No significant similarity found | ||
Chicken Feather Meal (Gallus gallus domesticus) | SNPSVAGVR | Helianthus annuus | 20 | |
Olea europaea subsp. Europaea | 34 | |||
Lupinus albus | 21 | |||
Cicer arietinum | 5 | |||
Tenebrio molitor | 10 | |||
Salmon pectoral fin (Salmo salar) | PAY | No significant similarity found | ||
Salmon skin hydrolysates (Salmo salar) | WG | No significant similarity found | ||
Locusts (Schistocerca gregaria) | FDPFPK | Lupinus albus | 17 | |
Cicer arietinum | 14 | |||
Olea europaea subsp. Europaea | 25 | |||
Helianthus annuus | 38 | |||
Tenebrio molitor | 9 | |||
Egg | HC | No significant similarity found | ||
Egg | CR | No significant similarity found |
It is interesting that there are sequences very similar to synthetic peptides such as KKIRVRLSA (SET-M33D) and YKRWKKRWAKYWKKFRK inside the proteome of natural protein sources such as sunflower or the synthetic peptide GAKYAKIIYNYLKKIANALW (GW-A2) within the yellow mealworm proteome. In addition, the peptides VDVPVKVPYS and GPETAFLR, derived from sorghum and lupin, respectively, are also found in the protein of the insect yellow mealworm. Our study also found similarities between the synthetic peptide KGIRGYKGGYCKGAFKQTCKCY and the protein of legumes such as chickpea or pea, the peptide KIWHHTF from sturgeon muscle and lupin or pea protein or the peptide SNPSVAGVR isolated from chicken feather and the protein from olive. According to this in silico analysis, Tenebrio molitor, Helianthus annuus and Lupinus albus could be good sustainable sources of peptides with anti-inflammatory activity. These data confirm that peptide sequences that have been reported with great anti-inflammatory activity are present within the proteome of other organisms apart from its original source. This is an opportunity to obtain such bioactive sequences from alternative available and sustainable protein sources.
PreAIP | AIPpred | ||
---|---|---|---|
Peptide sequence | Score | Peptide sequence | Prob |
a High confidence. b Medium confidence. c Low confidence. d Negative AIP. e Non-AIP. | |||
LKWLKKLLKKL | 0.667a | LKWLKKLLKKL | 0.6488 |
SIFGKIFKRIIRVAWK | 0.639a | GKLTKDKLKRGAKKALNVASKV | 0.6419 |
KGIRGYKGGYCKGAFKQTCKCY | 0.634a | HAEGTFTSDVSSYLEGQAAKEFI | 0.6372 |
YKRWKKRWAKYWKKFRK | 0.634a | RGQANILAGKNIKIRSGAAAGVGKTPQKANVEVLALGIW | 0.6209 |
GAKYAKIIYNYLKKIANALW | 0.631a | KGIRGYKGGYCKGAFKQTCKCY | 0.6163 |
YKRWKKNWAKYWKIFRK | 0.625a | QITITVKPRFRRIKRLFRGFR | 0.6 |
QITITVKPRFRRIKRLFRGFR | 0.609a | KKIRVRLSA | 0.5953 |
GKLTKDKLKRGAKKALNVASKV | 0.57a | SEGGFLE | 0.5791 |
QNWDFCEAWEPCF | 0.548a | SIFGKIFKRIIRVAWK | 0.5535 |
KKIRVRLSA | 0.515a | GAKYAKIIYNYLKKIANALW | 0.5302 |
HAEGTFTSDVSSYLEGQAAKEFI | 0.501a | QA | 0.5116 |
DEQHLETELHTHLTSVLTANGFQ | 0.498a | GPETAFLR | 0.4907 |
GPETAFLR | 0.481a | HC | 0.4884 |
KIWHHTF | 0.476a | CR | 0.4884 |
RGQANILAGKNIKIRSGAAAGVGKTPQKANVEVLALGIW | 0.474a | WG | 0.4884 |
HLDDALRGQE | 0.466b | SNPSVAGVR | 0.486 |
VHYAGTVDY | 0.449b | DEQHLETELHTHLTSVLTANGFQ | 0.4837 |
KLPDHPKLPK | 0.424b | YKRWKKNWAKYWKIFRK | 0.4791 |
SNPSVAGVR | 0.42b | EDDQMDPMAK | 0.4698 |
EDDQMDPMAK | 0.383c | KA | 0.4465 |
YGGGGE | 0.374c | YKRWKKRWAKYWKKFRK | 0.4442 |
SEGGFLE | 0.364c | VHYAGTVDY | 0.4326 |
FDPFPK | 0.353c | PAY | 0.4326 |
VDVPVKVPYS | 0.34d | KIWHHTF | 0.4256 |
HC | 0.291d | FDPFPK | 0.4233 |
HYGH | 0.284d | QNWDFCEAWEPCF | 0.4209 |
QA | 0.283d | KLPDHPKLPK | 0.4116 |
CR | 0.265d | VDVPVKVPYS | 0.3791 |
PAY | 0.261d | HYGH | 0.3791 |
WG | 0.258d | HLDDALRGQE | 0.3605 |
KA | 0.258d | YGGGGE | 0.3209e |
This study found seven coincidences among the ten highest peptide sequences scored for both predictors. The synthetic peptide LKWLKKLLKKL scored the greatest anti-inflammatory potential according to both predictors. Peptide SIFGKIFKRIIRVAWK derived from the human myosin protein also has a high score in the ranking, ranking second position in PreAIP and ninth in AIPpred. Foxtail millet-derived QNWDFCEAWEPCF peptide has a high score in the PreAIP predictor but does not appear among the 10 best in the AIPpred predictor. Same happens with Cecropin-TY1 RGQANILAGKNIKIRSGAAAGVGKTPQKANVEVLALGIW, which has a high score in AIPpred but is not among the best PreAIP peptides. Among the worst rated are the dipeptides KA, WG and CR and the tripeptide PAY. This is because the predictors used are not compatible with sequences shorter than 5 amino acids, therefore, they are considered non-anti-inflammatory. Interestingly, peptide YGGGGE from soybean that was considered as a strong anti-inflammatory peptide since it inhibited a 56.78–75.76% NO production together with reducing TNF-α, IL-1β and IL-6 levels receives a low score in both predictors, not being considered an anti-inflammatory peptide by AIPpred but being considered a low-confidence anti-inflammatory peptide by PreAIP. However, SEGGFLE peptide also derived from soybean it is considered an anti-inflammatory peptide by AIPpred although for PreAIP it is a low confidence anti-inflammatory peptide.
The correlation coefficient of the scores obtained for each peptide by both predictors is 0.2337, which indicates the lack of relationship between the two selected anti-inflammatory predictors indicating that in vitro assays are still necessary. In addition, both predictors have some limitations, such as the sequence length they allow to analyse. In the case of PreAIP, the score is calculated based on a fixed sequence length of 25 residues, which can cause loss of information. As a consequence, PreAIP does not allow the analysis of small peptides while AIPpred requires a minimum chain length of 5 residues and a maximum of 25 residues. This is a limitation taking into account that a proportion of the biological active peptides reported previously have sequences of 2 to 3 amino acids. Therefore, a bioinformatic tool capable of predicting the bioactivity of short sequences is necessary since the selected predictors can not estimate the bioactivity of dipeptides or tripeptides.
Furthermore, there are differences between the scores obtained with the two predictors which may be due to the method implemented to measure the potential bioactivity and to the fact that the cut-off value of each predictor is different. There is still room to improve prediction performance and machine learning algorithms of the actual anti-inflammatory prediction tools in terms of precision and efficiency to support medical research. Nevertheless, these bioinformatic tools can be very useful in qualitative analysis of bioactive sequences previous to in vitro investigations, serving as a guide to provide valuable leads for proteome-wide prediction and classification of prospective anti-inflammatory candidates for experimental validations.
Considering that the anti-inflammatory peptides reviewed act through different mechanisms with no apparent relation to enzyme treatment or protein source, identifying specific physicochemical or structural features for their bioactivity its very challenging. However, there are indeed some common characteristics that seem to play an important role in their response to inflammation such as presence of hydrophobic (Val, Ile, Pro) and positively charged (His, Arg, Lys) amino acids. Nevertheless, these residues are not found in all anti-inflammatory peptides or molecular weight, where most of the bioactive peptides identified so far have 2–40 residues with a molecular weight less than 3 kDa. Other investigations confirm the positive effect on inflammatory response linked to the presence of aromatic residues or proline in terminal position.
Conventional animal and plant protein sources are undoubtedly the most used in bioactive peptides production, as evidenced by the significant body of research on the subject. However, identification of anti-inflammatory peptides in alternative protein sources, such as Tenebrio molitor, lupine (Lupinus albus) or sunflower (Helianthus annuus), has been possible thanks to BLASTp analysis. These findings pave the way to obtain biopeptides from more sustainable and nutritious sources than conventional meat or dairy proteins.
Bioinformatic approaches are drawing increasing attention as a tool to complement experimental trials and identify potential anti-inflammatory peptides prior to their synthesis. In our case, in silico bioactivity predicted by PreAIP and AIPpred computer tools showed little relation to in vitro experimental anti-inflammatory activity reported in literature. These predictors could be useful in hypothesis-based experimental design by classifying potential anti-inflammatory candidates, but the performance of the mentioned anti-inflammatory predictors still needs to be implemented in order to provide more accurate and reliable bioactivity prediction for the growth of new biopharmaceuticals in near future.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d2fo02223k |
This journal is © The Royal Society of Chemistry 2022 |