Acquired immunodeficiency syndrome (AIDS) is caused by Human immunodeficiency virus (HIV) in which progressive failure of the human immune system causes life-threatening opportunistic infection or cancer. HIV-1 virus multiplies only inside the human body. HIV-1 entry into cells involves formation of a complex between gp120 of the viral envelope glycoprotein (Env), a host receptor (CD4), and a chemokine coreceptors usually CCR5 or CXCR4 [1]. gp120 contains the transmembrane protein gp41 and is derived from polyprotein gp160. Polyprotein gp 160 is encoded by env gene, which is found in all retrovirus [2]. Antiretroviral drugs, peptides have demonstrated potential to inhibit the Human immunodeficiency virus (HIV) [3]. Antipeptide or inhibiting peptide aim to preventing virus/coreceptor interaction by binding either virus envelope proteins or host proteins. The main cause of HIV infection starts with the interaction of exterior envelop of the viral protein gp120 with the chemokine receptors of CD4, the target cell. CD4 not only contributes in the viral attachment but also triggers some conformational changes in the HIV envelop that helps in the recognition of various chemokine receptors and leads to the membrane fusion. These conformational changes leads to the increase in the sensitivity of gp120 loops, release number of gp120 proteins and formation of chemokine receptor site along with its exposure of to gp120 molecules. It also leads to the formation of epitopes of neutralizing antibodies which blocks chemokine receptor binding [4].

Dataset for HIV antipeptide

HIV antipeptide dataset collected from HIPdb [3] database is freely available HIV antipeptide database. It is a manually curated database of experimentally validated HIV inhibitory peptides targeting various steps or proteins involved in the life cycle of HIV like fusion, integration, reverse transcription, etc. It is newly introduced database for HIV antipetides, which was used for collection of viral entry antipeptides. There are total 110 large dataset of antipeptide having different source and cell lines (Table 1) in this database. Out of 110 peptides, 47, 41, 5, 5, 4, 2, 2, 1, 1 inhibiting peptides were taken from GB virus, alpha-anti trypsin, Apelin, gp120, gp41, synthetic, RhoA, Tat; Dynein sources, respectively (Figure 1).

Table 1: HIV inhibiting peptide dataset in HIPdb database.

Multiple sequence alignment (MSA)

MSA were performed for all 110 inhibiting peptides using clustalW [5] and validate its result with T-Coffee [6] and Muscle [7]. Conserved patches of antipeptides for different cell lines and sources were collected from the clustalW result.

Prediction of antigenicity of conserved peptides and analysis

Low hydrophobic sequence evokes the production of peptide antibodies. The antigenicity analysis assists in the selection of low hydrophobic sequences from the one with the highest calculated antigenic potential. All the antipeptides were screened for their antigenicity using VaxiJen [8]. The threshold value for antigenicity is 0.4 in VaxiJen. Based on their conserved nature and antigenicity, few lists of peptides are proposed for their wide spectrum activity. The antigenicity is the ability of a compound to bind with antibodies and with cells of immune system.

Prediction of hydrophobicity of conserved peptides and analysis

Peptide property calculator [9] was used to calculate the percentage of hydrophobicity of all antipeptides. By using this tool, we got chemical formulae, molecular weight, hydrophobicity, hydrophilic, isoelectric point for analysis of inhibiting peptides. We analyzed our whole 11 inhibiting peptide with derived ID from HIPDB database inhibiting from Peptide Property Calculator.

Prediction of antiviral and antimicrobial activity of 11 peptides

Conserved Antimicrobial Peptide (CAMP) [10] is used for prediction of antimicrobial activity of our 11 peptides using the Support Vector Machine (SVM) [11] and Random Forest (RF) [12] algorithm. The category of AMPs has been analyzed from class AMP [13].

The results of MSA of 14 antipeptides were visualized in Jalview. Figure 2 is showing the decamers antipeptides in jalview. Then we predicted antigenicity of all 14 inhibiting peptides through VaxiJen. All the 14 HIV antipeptides’ antigenicity is shown in Table 2. The final step is to predict the best HIV antipeptide among these 14 as shown in Table 3. In our result, DRRPASCGTC, EESQNQQEKN, KYRRFRWKFK, SDRDTVVELS and KIRPSLHLVKC are showing hydrophobicity below 60%. These antipeptides have potential to inhibit HIV virus entry and they are able to induce humoral mediated immunity. In this study, DRRPASCGTC is showing minimum hydrophobicity with 20% and also having high antigenicity with 0.7480.

Table 2: All the 14 HIV antipeptides’ antigenicity.

Table 3: The best HIV antipeptide.

To explore AMPs as drugs, it is essential to understand sequencespecificity relationship of Anti Microbial Peptides (AMPs). Two algorithms have been used for prediction of antibacterial, antifungal and antiviral peptides based on their sequence composition (Tables 5-7). The individual inference about each 11 peptides is as follows:

Table 5: Conserved antimicrobial peptides.

Table 6: AMP analysis of Conserved antipeptides.

Table 7: Antipeptide and target organism.

EESQNQQEKN

For a peptide to be antigenic, it must be less hydrophobic and VaxiJen score should be more than 0.4. To our surprise, this peptide is predicted to be 0% hydrophobicity and has antigenic score of 0.5418. This peptide has a good probability of immunogenicity because it has molecular weight 1233.21 D, as per the rule the molecular weight has to be greater than 500D. As per CAMP analysis, this peptide is predicted to be AMP from both algorithms, SVM and Random Forest with probability of 0.69 and 0.55, respectively. This peptide has AMP category of antifungal activity from both algorithms predicted from classAMP and Random Forest with probability of 0.72 and 0.58 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “WQEWERKVDFLEENITALLEEAQIQQEKNMYELQK” sequence which has antiviral activity to Simian immunodeficiency Virus with 80% identity & E value of 1.3. Thus this conserved peptide has derived from gp41 sequence which is working as a potent viral entry inhibitor of HIV-1, SIVmac251 and SHIV89.6P [14]. This peptide might work against either CD4/coreceptor or gp120 to inhibit viral entry (Table 4).

Peptide Sequence	H** (%)	H*(%)	Others (%)	Chemical Formula	MW(Da)	pI	Piechart
KIRDSLHLVKC	45	36	18	C57H102N18O15S1	1311.60	9.67
KYRRFRWKFK	60	30	10	C74H111N23O12	1514.83	12.27
PNNNTRKSIH	30	20	50	C48H81N19O16	1180.28	11.66
PNNNTRRRLS	30	20	50	C48H86N22O16	1227.34	12.80
PTGERVWDRG	40	30	30	C50H77N17O16	1172.26	6.51
PWQGGRRKFR	40	30	30	C58H90N22O12	1287.48	12.81
RGNPVRSPLG	20	40	40	C44H77N17O13	1052.19	12.50
SDRDTVVELS	40	30	30	C45H77N13O20	1120.17	3.88
EESQNQQEKN	40	0	60	C47H76N16O23	1233.21	3.98
DRRPASCGTC	30	20	50	C39H68N16O15S2	1065.19	8.23
PYQNTRQRTP	20	20	60	C53H85N19O17	1260.36	11.15

Table 4: Hydrophobicity of 11 conserved anti-peptides.

PTGERVWDRG

This peptide is predicted to be 30% hydrophobicity and has antigenic score of 0.2602. This peptide has no probability of immunogenicity because VaxiJen score is less than threshold level i.e., 0.4 and it has molecular weight of 1172.26 D. As per CAMP analysis, this peptide is predicted to be non-AMP from SVM and AMP from random forest algorithms, with probability of 0.70 and 0.70 respectively. This peptide has AMP category of antiviral activity from SVM and antibacterial activity from random forest algorithms predicted from classAMP, with probability of 0.75 and 0.49 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “GLRKKFRKTRKRIQKLGRKIGKTGRKBWKAWREYGQIEYPCRI” sequence which has antibacterial activity to E.coli and S. aureus with 66% identity and E value of 24. This peptide is predicted to be antiviral and antibacterial activity with no immunogenicity. This conserved peptide has been derived from GB Virus C gE2, the peptides P6-2 and P4762 inhibits HIV-1 replication via interaction with HIV-1 particle and avoid the entry of virions [15]. This peptide is showing antiviral activity with no immunogenicity.

KYRRFRWKFK

This peptide is predicted to be 30% hydrophobicity and has antigenic score of 1.61. This peptide has a good probability of immunogenicity because it has molecular weight of 1514.83D. As per CAMP analysis, this peptide is predicted to be AMP from both algorithms, SVM and Random Forest with probability of 1 and 0.89 respectively. This peptide has AMP category of antibacterial activity from both algorithms predicted from classAMP and Random Forest with probability of 0.99 and 0.56 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “GRFKRFRKKFKKLFKKLS” sequence which has antibacterial and antifungal activity to E. coli, S. typhimurium, T. aeruginosa, S. marcescens, S. aureus, S. epidermidis, B. megaterium, C. albicans, C. neoformans with 60% identity & E value of 2.8. This conserved peptide is derived from synthetic peptides (SALPs- Synthetic anti-lipopolysaccharides peptides), bind to heparin sulphate moeties on the cell surface and inhibit the entry of HIV-1, HSV-1 and 2 both, HBV and HCV [16]. These peptides have high antiviral efficiency, no toxicity and adverse effects. This peptide is best as it is showing antibacterial as well as antifungal activity with good score of antigenicity and immunogenicity. This peptide has wide spectrum future prospective in HIV treatment.

PNNNTRKSIH

This peptide is predicted to be 20% hydrophobicity and has antigenic score of -0.155. This peptide has no probability of immunogenicity because VaxiJen score is less than threshold; it has molecular weight of 1180.28 D. As per CAMP analysis, this peptide is predicted to be AMP from SVM and non-AMP from random forest algorithms, with probability of 0.913 and 0.586 respectively. This peptide has AMP category of antifungal activity from both algorithms predicted from classAMP, and Random Forest with probability of 0.716 and 0.562 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “MSRSLKKGPFVFYSLIK KVDQMNSNRFKSVILTWSRSCTIIPIMIGNTIGVYNGKEHIPVL VSDQMIGHKLGEFVQPRNYRGHKKHDKKTKTKR” sequence which has antimicrobial activity to Simian immunodeficiency Virus with 50% identity and E value of 155. Similar to 2nd peptide, this peptide is predicted to be antifungal and antimicrobial activity with no immunogenicity. This conserved peptide has derived from gp120, V3 region of gp120 of T cell line trophic directly interact with CXCR4, a chemokine receptor of CD4+ cells, hence inhibiting T trophic HIV-1 infection [17]. This conserved peptide may be act as a good candidate to inhibit viral entry.

PNNNTRRRLS

This peptide is predicted to be 20% hydrophobicity and has antigenic score of -0.3115. This peptide has no probability of immunogenicity because VaxiJen score less than threshold; it has molecular weight of 1227.34. As per CAMP analysis, this peptide is predicted to be AMP from SVM and non-AMP from random forest algorithms, with probability of 0.589 and 0.598 respectively. This peptide has AMP category of antiviral activity from SVM and antifungal activity from random forest algorithms predicted from classAMP, with probability of 0.708 and 0.630, respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “SLSVEAKAKIVADFGRDANDTGSSEV QVALLTAQINHLQGHFSEHKKDHHSRRGLLRMVSTRRKLL DYLKRKDVASYVSLIERLGLRR” sequence which has antimicrobial activity with 60% identity and E value of 119. Similar to 4^th peptide, this peptide is predicted to be antifungal and antimicrobial activity with no immunogenicity. This conserved peptide has been derived from gp120, V3 region of gp120 of T cell line trophic directly interact with CXCR4, a chemokine receptor of CD4+ cells, hence inhibiting T trophic HIV-1 infection .This conserved peptide may act as a good candidate to inhibit viral entry.

PYQNTRQRTP

This peptide is predicted to be 20% hydrophobicity and has antigenic score of 0.112. This peptide has no probability of immunogenicity because VaxiJen score is less than threshold; it has molecular weight of 1260.36. As per CAMP analysis, this peptide is predicted to be non-AMP from SVM and AMP from random forest algorithms, with probability of 0.993 and 0.528 respectively. This peptide has AMP category of Antiviral activity from SVM and antibacterial activity from random forest algorithms predicted from classAMP, with probability of 0.858 and 0.636 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “DEKPKLILPTPAPPNLPQLVGGGGGNR KDGFGVSVDAHQKVWTSDNGGHSIGVSPGYSQHLPGPYGNS RPDYRIGAGYSYNF” sequence which has antibacterial activity to E. coli with 66% identity & E value of 70. This peptide sequence also has the capability to act as a AMP without having the property of inducing immune response. This peptide is predicted to be antiviral and antibacterial activity with no immunogenicity. This conserved peptide has derived from gp120, V3 region of gp120 of T cell line trophic directly interact with CXCR4, a chemokine receptor of CD4+ cells, hence inhibiting T trophic HIV-1 infection .This conserved peptide may be act as a good candidate to inhibit viral entry.

PWQGGRRKFR

This peptide is predicted to be 30% hydrophobicity and has antigenic score of 0.129. This peptide has no probability of immunogenicity because VaxiJen score is less than threshold; it has molecular weight of 1287.48. As per CAMP analysis, this peptide is predicted to be AMP from both algorithms, SVM and Random Forest with probability of 0.981 and 0.810 respectively. This peptide has AMP category of Antiviral activity from SVM and antibacterial activity from random forest algorithms predicted from classAMP, with probability of 0.951 and 0.654 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “RIRRPIALIWRGGRRLTEWL” sequence which has antimicrobial activity with 83% identity and E value of 4.8. This conserved peptide has derived from Apelin, inhibits the entry of some HIV-1 and 2 into CD4+ cells APJ receptor [18]. This is another best peptide candidate having antiviral activity with no immunogenic response.

RGNPVRSPLG

This peptide is predicted to be 40% hydrophobicity and has antigenic score of -0.2856. This peptide has no probability of immunogenicity because VaxiJen score is less than threshold; it has molecular weight of 1052.19. As per CAMP analysis, this peptide is predicted to be AMP from both algorithms, SVM and Random Forest with probability of 0.625 and 0.632 respectively. This peptide has AMP category of Antiviral activity from SVM and antibacterial activity from random forest algorithms predicted from classAMP, with probability of 0.953 and 0.772 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “RFRPPIRRPPIRPPFRPPFRP PVRPPIRPPFRPPFRPPIGPFP” sequence which has antimicrobial activity with 57% identity and E value of 53. This conserved peptide has derived from GB Virus C gE2, the peptides P6-2 and P4762 inhibits HIV-1 replication via interaction with HIV-1 particle and avoid the entry of virions. This peptide is showing high antiviral activity with no immunogenicity.

SDRDTVVELS

This peptide is predicted to be 30% hydrophobicity and has antigenic score of 0.572. This peptide has probability of immunogenicity because it has molecular weight of 1120.17. As per CAMP analysis, this peptide is predicted to be non-AMP from SVM and AMP from random forest algorithms, with probability of 0.959 and 0.624 respectively. This peptide has non AMP category of Antifungal activity from SVM and AMP antibacterial activity from random forest algorithms predicted from classAMP, with probability of 0.789 and 0.372 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “MKRNQRKQLIGTVVSTKNAKTATVKVT SRFKHPLYHKSVIRHKKYHVHNFGELVANDGDRVQIIETRPLSA LKRWRIVKIIERAK” sequence which has antimicrobial activity with 42% identity and E value of 346. This conserved peptide has derived from GB Virus C gE2, the peptides P6-2 and P4762 inhibits HIV-1 replication via interaction with HIV-1 particle and avoid the entry of virions. This peptide is showing low antiviral activity with immunogenicity.

KIRDSLHLVKC

This peptide is predicted to be 36% hydrophobicity and has antigenic score of 0.572. This peptide has a probability of immunogenicity because it has molecular weight of 1311.60. As per CAMP analysis, this peptide is predicted to be AMP from both algorithms, SVM and Random Forest with probability of 0.588 and 0.712 respectively. This peptide has AMP category of antifungal activity from SVM and antibacterial activity from Random forest, predicted from classAMP, and Random Forest with probability of 0.980 and 0.858 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “GLLSGILGAGKHIVCGLSGPCQSLNRKSSDVEYHLAKC” sequence which has antibacterial activity to Odorrana grahami with 80% identity and E value of 45. This conserved peptide has derived from GB Virus C gE2, the peptides P6-2 and P4762 inhibits HIV-1 replication via interaction with HIV-1 particle and avoid the entry of virions. This peptide is showing high antiviral activity with immunogenicity.

DRRPASCGTC

This peptide is predicted to be 20% hydrophobicity and has antigenic score of 0.748. This peptide has a good probability of immunogenicity because it has molecular weight of 1065.19 D. As per CAMP analysis, this peptide is predicted to be AMP from both algorithms, SVM and Random Forest with probability of 0.678 and 0.604 respectively. This peptide has AMP category of antiviral activity from SVM and antibacterial activity from Random forest, predicted from classAMP, with probability of 0.890 and 0.532 respectively. After AVBlast analysis of this peptide, it was found that this sequence is a part of “EQKQGQYGEGSLRPSECGQRCSYRCSATSHKKPCMFFCQKCC AKCLCVPPGTFGNKQVCPCYNNWKTQQGGPKCP” sequence which has antimicrobial activity to Simian immunodeficiency Virus with 66% identity & E value of 24. This conserved peptide has derived from GB Virus C gE2, the peptides P6-2 and P4762 inhibits HIV-1 replication via interaction with HIV-1 particle and avoid the entry of virions. This peptide is showing moderate antiviral activity with high immunogenicity.

In general, peptide sequences between 10 and 20 amino acids in length are recommended for ideal antigen [19]. So, our conserved antipeptide having 10 amino acid length which can be used as an inhibitor. In general, most ideal antigenic epitopes are hydrophilic, surface orientated and flexible. Hydrophilic residues are surface exposed so it has better affinity to bind with paratope as compared to hydrophobic residues.

In this study, we also compare hydrophobicity and antigenicity of our conserved antipeptide with antipeptide from HIPdb databases. Low hydrophobicity shows low affinity binding characteristics.

Many efforts are being made for the inhibition of HIV virus entry block by in silico, in vivo, in situ and in vitro approaches. We came to conclusion that PWQGGRRKFR is showing hydrophobicity of 30% and antigenicity of 0.129 with a molecular weight of 1287.48. So, according to this data it will definitely work against all the pathogens without interfering our immune system. Therefore, it acts as a broad spectrum antipeptide. KYRRFRWKFK is showing hydrophobicity of 30% and antigencity of 1.61 with a molecular weight of 1514.83D. So, all the above parameters prove that this peptide will inhibit the CD4- gp120 interaction with invoking immune system. Out of 11, we got 10 having less hydrophobicity percentage from derived HIPDB database, which proves that our inhibiting peptides will be effective inhibitor peptides than HIPdb antipeptides. These antipeptides have potential to inhibit HIV virus entry and they are able to induce humoral mediated immunity. These are the conserved patches taken from different sources and effective against different cell lines so these are wide spectrum antipeptides inhibiting the primary interaction of gp120-CD4 that is the major culprit for HIV pathogenesis.

We are very thankful to the Nthrys Biotech Labs, 1-8-747/11, Bagh Lingampally, near RTC Road, Hyderabad-44, where this work was performed. We again acknowledge the Nthrys Biotech Labs. as well as the scientific staff for providing infrastructural facilities.