Researching of resistance to etravirine in some HIV-1 low-level viremia strains by in-silico methods

Introduction

Human immunodeficiency virus (HIV) is a pathogen that targets the immune system of the human host, specifically the CD4+ T lymphocytes, which play a critical role in the body’s defense against infections. HIV primarily spreads through sexual transmission, but it can also be transmitted through blood-borne routes such as contaminated needles or blood transfusions. HIV is a chronic condition for which there is currently no cure, but it can be managed through antiretroviral therapy (ART). ART is a combination of medications that can suppress the viral load and reduce the risk of transmission to others. When administered consistently, ART can enable individuals living with HIV to lead long and healthy lives. If left unchecked, HIV can progress to acquired immunodeficiency syndrome (AIDS), the most advanced stage of HIV infection, characterized by severe damage to the immune system and increased susceptibility to opportunistic infections [1].

The clinical significance of low levels of viremia, also known as low-level viremia (LLV), remains a topic of debate and investigation within the medical community. A major contributing factor to the lack of consensus regarding the clinical implications of LLV is the inconsistent definitions and measurement methods employed across studies, as well as the discrepancies in the findings of these studies. Nonetheless, a relatively broad definition of LLV, encompassing viral loads between 200 and 1,000 copies/mL, has consistently been associated with virologic failure, viral evolution, and the emergence of drug resistance in multiple investigations. As such, the clinical significance of LLV warrants further examination in order to better understand its potential impact on the management of HIV [2, 3].

The effectiveness of narrow-spectrum non-nucleoside reverse transcriptase inhibitors (NNRTIs) is limited due to the low genetic barrier to resistance, which allows for the easy selection of single mutations that often confer complete cross-resistance. Etravirine (ETR) is a modified NNRTI that exhibits conformational flexibility, allowing it to bind to HIV reverse transcriptase (RT) in the presence of most mutations that would normally confer resistance to other NNRTIs [4]. Mutations have the potential to alter the three-dimensional structure and biochemical properties of proteins. Notably, changes occurring in the ligand-binding region can trigger resistance to drugs. Alterations in amino acid residues involved in protein-ligand interactions can modify physicochemical properties, binding affinity, number, and type of bonds [5].

Several in vitro studies have demonstrated that multiple mutations are required for the development of high-level resistance to etravirine (ETR) [6, 7]. Clinical trials have shown that ETR is effective in rapidly reducing plasma viral loads in both treatment-naive patients [8] and those with NNRTI-resistant viruses. Studies have identified several ETR resistance-associated mutations (RAMs), including V90I, A98G, L100I, K101E/P, K101H, E138A, V179T, V106I, V179D/F, Y181C/I/V, M230L, and G190A/S (http://hivdb.stanford.edu) [9].

The clinical implications of LLV remain unclear and have been a focus of interest in clinical research. One of the difficulties in understanding the significance of LLV is the discrepancy between the results of various studies, as well as the lack of consistency in the definitions of LLV and virologic failure used in these studies. While some research has consistently shown that LLV in the range of 200–1,000 copies/mL is associated with virologic failure, viral evolution, and the development of drug resistance [2, 3], it is not yet clear whether LLV in the range of 50–199 copies/mL (below the current threshold defined by the Department of Health and Human Services as persistent viremia ≥200 copies/mL) is associated with virologic failure [10, 11]. Some studies have found that LLV in this range is not associated with virologic failure, while others have found that it is associated with virologic failure and the emergence of drug resistance [12].

In our study, mutations that are located close to the binding pocket of etravirine (ETR) and that have the potential to alter the location of amino acids within the binding pocket were analyzed for their effect on ETR resistance using in-silico methods. The mutations (S191F, L228R, K101R-K104R, K104E – H235L, K102Q, K104Q, A98S, V179I) analyzed in this study were obtained from the National Reference Laboratory of HIV/AIDS and Viral Hepatitis, Ministry of Health, Turkey.

Materials and methods

Results

Homology modeling

The homology models of the wild-type (Figure 1A) and mutant strains were compared, and conformational changes were analyzed. When considering the homology model of the wild-type strain, it was observed that there were changes in physicochemical properties (such as hydrophobicity, electrical charge, polarity, etc.) at the atomic and amino acid level due to the amino acid substitutions in the mutation regions of the mutant strains. However, the alpha helices and/or beta sheets did not change, or only minor conformational changes occurred. The analysis revealed that the wild-type and K10Q, K101R-K104Q, S191F, K104Q, L228R, I94L-K104Q, and K104E-H235L mutant strains all have a similar hydrophobic NNRTI pocket. However, in the V179I and A98S mutant strains, there were changes in the hydrophobic pocket due to the mutations, and in the A98S mutant, there was also a change in hydrophobicity. A distance analysis was performed to identify the amino acid locations associated with ETR in the hydrophobic ETR pocket. An average change of 0.3 Å was observed in the A98S mutation (Table 1), while the V179I (Figure 1B) mutation was found to affect the entry point of ETR into the pocket by increasing the volume at the pocket’s entrance.

Figure 1:

Shows the hydrophobicity surface analysis of (A) wild-type model and (B) V179I model. Hydrophilic regions are indicated by blue, hydrophobic regions are indicated by red, and neutral amino acids are indicated by white.

Table 1:

This statement refers to the distances between the 98th amino acid and the amino acids ETR (average distance), L100, K101, and Y318 in both wild-type and mutant models.

Distance, Å	ETR	L100	K101	Y318
A 98	9.8	5.8	8.7	5.7
S 98	9.5	5.7	8.4	5.4

A98S mutant type

According to the molecular docking analysis results, when the wild type (Figure 2A) and A98S (Figure 2B) models are compared, hydrogen bonding interaction between Lys101 and ETR has increased, the π-interaction regions of Leu100 with ETR have changed and increased, the π-interaction regions of Val179 with ETR have changed and increased, the π-interaction region of Val106 with ETR has changed and decreased, the π-interaction region of Trp229 with ETR has changed, the π-interaction region of Lys103 with ETR has changed, the interaction between Tyr188 and ETR has been lost, the interaction between Leu234 and ETR has been lost, a π-interaction has been established between Tyr181 and ETR, the interaction between Tyr318 and ETR has been lost.

Figure 2:

2D Diagram representation of molecular docking results. (A) Wild-type and (B) A98S.

V179I mutant type

According to the molecular docking analysis results, when the wild type (Figure 3A) and V179I (Figure 3B) models are compared, the interaction region of Ile179 amino acid with ETR has changed and decreased. A new donor-donor interaction has formed, the interaction region of Val106 with ETR has changed, the interaction region of Leu100 with ETR has changed, the interaction region of Trp229 with ETR has increased, the interaction region of Lys103 with ETR has changed, the hydrogen bond is not broken, and Ile101 replaced Val179, the π-interaction between Tyr181. ETR has formed, the π-interaction between Pro95 and ETR has formed, the interaction between Tyr318 and ETR has disappeared, the interaction between Leu234 and ETR has disappeared, and the interaction between Tyr318 and ETR has disappeared.

Figure 3:

2D Diagram representation of molecular docking results. (A) Wild-type and (B) V179I.

Discussion

According to the results of homology modeling, molecular docking, and protein-ligand interaction analysis, it is believed that the mutations S191F, K104E – H235L, K102Q, and K104Q identified in the HIV-RT protein of low-level viremic patients do not confer resistance to ETR. LLV is not caused by these mutations [2, 15, 16]. Instead, it is thought that LLV may be due to virological failure [11].

In the literature, it has been reported that the S191F mutation alone did not confer resistance [17], and the K104 polymorphisms were suggested to be associated with resistance in a study by Sushama et al. in 2016 [18]. The K102Q mutation was not found to be associated with resistance, but the K102R mutation was reported to be resistant in combination with other mutations in a study by Hachiya et al. in 2004. There is no study suggesting that the H235L mutation is associated with resistance. Based on the literature and our analysis, we believe that the mutations S191F, K104E – H235L, K102Q, and K104Q in the HIV-RT protein do not confer resistance to ETR and, therefore cannot cause LLV.

The A98S mutation is believed to lead to low-level resistance due to the formation of a new π-stacking interaction with the 181TYR amino acid and certain ETR-amino acid interactions, which affect the ETR-protein binding score. According to a study conducted by Vingerhoets et al. in 2010 [19], the A98S mutation can lead to resistance in combination with another NNRTI mutation in the same strain. Additionally, it has been reported that the A98G mutation can lead to resistance in combination with other mutations, according to the Stanford HIV Drug Resistance Database [19, 20].

It is believed that the V179I mutation leads to low-level resistance due to the disappearance of a hydrogen bond and the appearance of two new hydrophobic interactions, which cause changes in the ETR-protein binding region. According to previous reports, the region of V179 is highly prone to mutations, and V179D/E/F/L/T mutations have been suggested to cause resistance [21]. In a cell culture study conducted by Dwivedi R. and colleagues in 2022, the V179I mutation was found to have drug susceptibility similar to the wild type, but it showed resistance to rilpivirine, one of the NNRT inhibitors when combined with Y181 mutations. Additionally, in a bioinformatics analysis conducted by Marcelin A. and colleagues in 2012, changes at the V179 position were found to be important for ETR resistance, and V179I/L mutations were suggested to be associated with resistance with other mutations, reducing phenotypic susceptibility.

Our in-silico analyses, parallel to the evaluation of literature data, imply that A98S and V179I mutations may alter the conformational structure of the ETR bound hydrophobic pocket, potentially resulting in low-level resistance against ETR. Based on literature studies and our analysis results, it is considered that A98S and V179I mutations detected in HIV-RT protein can show low-level resistance to ETR alone and can lead to high resistance in a combined situation.