Variable Surface Glycoprotein from Trypanosoma brucei Undergoes Cleavage by Matrix Metalloproteinases: An in silico Approach

In order to survive as extracellular parasites in the mammalian host environment, Trypanosoma brucei has developed efficient mechanisms of immune system evasion, which include the abundant expression of a variable surface glycoprotein (VSG) coat. VSGs are anchored in the parasite membrane by covalent C-terminal binding to glycosylphosphatidylinositol and may be periodically removed by a phospholipase C (PLC) and a major surface protein (TbMSP). VSG molecules show extraordinary antigenic diversity and a comparative analysis of protein sequences suggests that conserved elements may be a suitable target against African trypanosomiasis. However, the cleavage mechanisms of these molecules remain unclear. Moreover, in protozoan infections, including those caused by Trypanosoma brucei, it is possible to observe an increased expression of the matrix metalloproteinases (MMPs). To address the cleavage mechanism of VSGs, the PROSPER server was used for the identification of VSG sequence cleavage sites. After data compilation, it was observed that 64 VSG consensus sequences showed a high conservation of hydrophobic residues, such as valine (V), methionine (M), leucine (L) and isoleucine (I) in the fifth position—the exact location of the cleavage site. In addition, the PROSPER server identified conserved cleavage site portions of VSG proteins recognized by three matrix metalloproteases (gelatinases: MMP-2, MMP-3 and MMP-9). However, further biological studies are needed in order to analyze and confirm this prediction.


Introduction
Human African Trypanosomiasis (HAT) or sleeping sickness is a vector-borne disease caused by two hemoflagellate parasites named Trypanosoma brucei gambiense and Trypanosoma brucei rhodesiense [1]. Almost 60 million people are living in at risk areas in sub-Saharan Africa. There is no available vaccine and, therefore, rapid detection tests remain a priority for early treatment. Moreover, Trypanosoma brucei brucei parasites affect only animals-cattle and other ruminants [2]-causing important economic burden loss in endemic countries [3,4].
T. brucei have highly abundant variant surface glycoproteins (VSGs) expressed in the membrane, which provide numerous well-established functions on interaction with the mammalian host and even hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2). metalloproteinases (MMPs) in all subspecies sequences and that they include the two VSG domains, A and B. After the sequence compilation, 46 consensus sequences (Tables 1 and 2) were carried out. A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2).  (Tables 1 and 2) were carried out. A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2).  (Tables 1 and 2) were carried out. A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2).  (Tables 1 and 2) were carried out. A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2).  (Tables 1 and 2) were carried out. A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2).  (Tables 1 and 2) were carried out. A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2).  (Tables 1 and 2) were carried out. A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2).  (Tables 1 and 2) were carried out. A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2). A high conservation of hydrophobic residues (black letters) was observed, such as valine (V), methionine (M), leucine (L) and isoleucine (I), specifically in the fifth position-the exact location of the cleavage site. Additionally, in Table 1, a high frequency of polar neutral residues can be observed next to the fifth position (green letters). The sequences were recognized in the first 100 amino acids conserved in ascending order by MMP9 > MMP3 > MMP2 (Tables 1 and 2).

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions.

Discussion
This study underlines the importance of the in silico prediction of cleavage sites in all sequence regions of VSG molecules by matrix metalloproteinases (MMPs). The PROSPER server identified that all sequences have a conserved portion and specific amino acid cleavage site positions. Therefore, the activity of endogenous proteases would appear to be significant, essentially during microorganism infection. Previous studies, using a comparative analysis of different VSG sequences, suggest that a conserved element in this molecule could be an important target for an intervention strategy against African trypanosomiasis [6,10,[36][37][38].
It has been demonstrated by Cross that the VSG coat could be uniformly removed from the cells by proteases, such as trypsin or pronase, without causing any changes in parasite morphology [39]. Consequently, maybe proteolytic activity could play a role in the mechanism of protein surface replacement during T. brucei infection. However, LaCount and colleagues described three classes of conserved families of zinc metalloproteases (TbMSP-A, -B and -C) present on T. brucei as responsible for the release of ectopically expressed VSG molecules from the surface of procyclic trypanosomes [8,17]. All of these three families of proteins are expressed in bloodstream-stage trypanosomes, but only TbMSP-B is found in the procyclic stage. The sequences of the three TbMSP share approximately 33% identity, and the main difference is in their terminals. The TbMSP-A has an extended C-terminal region, rich in serine and glutamate amino acid residues, which was not seen in the other two, and finishes in a short hydrophobic segment. TbMSP-B has a hydrophobic tail in the C-terminal. TbMSP-C has a C-terminal region that is highly hydrophilic and rich in charged amino acids and proline, which indicates that it is not linked to a membrane via a GPI anchor, unlike TbMSP-A and -B [17]. Moreover, the hydrolysis of PLC results in the conversion of the hydrophobic membrane form of VSG (mfVSG) to a water-soluble VSG (sVSG), resulting in the shedding of the VSG from the parasite membrane [40].
We observed hydrophobic sites on VSGs as substrates for, at the least, one matrix metalloproteinase, MMP-9 protein. Due to the higher number of recognized portions in all T. brucei subspecies sequences, matrix metalloproteinase proteins maybe able to release the VSG coat from the surface of the parasite. Additionally, VSGs molecules are available for interaction with other molecules besides MSPs and PL-C [19]. Our speculation is that MMPs may use VSG molecules as substrates in parasite-host interactions during T. brucei infection.
In physiological processes, MMPs regulate the release or activation of chemokines, cytokines, growth factors, and other bioactive molecules [24]. Since VSGs are antigenically distinct, the conserved structure is probably necessary for their function. It is possible that the high numbers of protease cleavage sites identified by the PROSPER server may be important for VSG release and antigenic variation during infection by T. brucei. Thus, the protein-conserved positions identified in cleavage sites might have a significant function and are accessible for surface cleavage by MMPs in different sizes.
Some questions remain unanswered. Can MPPs directly cleave VSGs to provide the release mechanism of VSG coats during the phenomenon of antigenic variation? Does increasing MPP expression during T. brucei infection represent an escape mechanism of the parasite or host protection?
To answer these questions, the data predicted by the PROSPER server was associated with the interaction between the parasite and mammalian host, described so far. Based on several studies, MMPs represent a large family of secreted or membrane-bound endopeptidases, important in many physiological and pathological conditions, including cancer and protozoan parasitic diseases [24,41]. MMPs play a crucial role in leukocyte penetration in brain diseases [39]. Consistently, mice infected with T. brucei have a high expression of MMP-3 and MMP-12 at the mRNA level, followed by significant parasitemia increases [24,28]. Apparently VSG proteins from T. b. brucei have many conserved sequences, including both domains A and B, that have undergone cleavage by MMPs (Tables 1 and 2) and our results showed a significant number of cleavage sites recognized by MMP-3.
Besides, in the human infection context, some patients in the second stage of HAT (the neurological stage) present an increase in the expression of MMP-2 and MMP-9, correlated with the presence of parasites and leukocytes [20,24,42]. We could verify that MMP-2 and MMP-9 recognize several cleave sites of VSG domain A and B, present in T. b. gambiense and T. b. rhodesiense subspecies sequences. According to this finding, we believe that MMPs could be involved in VSG switch mechanisms and the release of soluble VSGs during infection caused by T. brucei. These data, together with the study model previously proposed by Moreno et al. [14], suggest that the mechanism of release of VSG in T. brucei may occur through the combination of host metalloproteinases and the intrinsic mechanisms of the parasite mediated by MSP and PLC enzymes.
Therefore, studies have shown that MMP inhibitors like antibiotics, such as tetracycline and minocycline, prolong the animal's survival and decrease the influx of parasites and leukocytes in the brain [20,27,28]. Our recent studies focus on the evaluation of potential new drugs in the context of trypanosomatid infections. Thus, the mechanism of the antigenic variation of T. brucei seems to be an interesting target for the search for potential new antiparasitic molecules [43,44]. Consequently, understanding the mechanism of VSG release and all the involved molecules represents an important strategy for the development of new drugs against infections caused by T. brucei. In addition, a detailed analysis of all identified VSG conserved residues may provide new insights into host-parasite interactions, taking into account that released soluble VSGs interact directly with the immune system. Nevertheless, more studies are needed to validate these cleavage sites by MMPs.

Materials and Methods
The conserved domains for VSG sequences were retrieved from the Conserved Domain Database of the NCBI [45]. All T. brucei sequences were recovered from the NCBI protein database, then those sequences were divided considering the three subspecies: T. b. brucei (Tbb), T. b. gambiense (Tbg) and T. b. rhodesiense (Tbr). Data were sourced using the Reverse Position-Specific BLAST (RPS BLAST) tool v2.6.0 [46]. During the search, the NCBI CDD accessions cl03014 (a-VSG) and cl26244 (b-VSG) were used. In addition, the fragmented domains were discarded, and the minimum e-value threshold was 1.0 × 10 −5 .
The sequences that contain VSG domains were submitted to the PROSPER server to predict protease cleavage sites [46]. As an exclusivity criterion, the cleavage sites with a score greater than 0.8 were considered as probable sites for MMPs/MSPs. To demonstrate that the MMP/MSP sites present in T. brucei are very conserved, the standard sites for these proteases were retrieved from the MEROPS database [47] (Supplementary Materials Figures S1-S3). The cleavage sites were searched in all extensions of the protein sequence. To illustrate the similarity of the cleave sites, an analysis in the WebLogo tool was performed [48]. Our approach can be summarized as follows in Figure 1.
mechanisms and the release of soluble VSGs during infection caused by T. brucei. These data, together with the study model previously proposed by Moreno et al. [14], suggest that the mechanism of release of VSG in T. brucei may occur through the combination of host metalloproteinases and the intrinsic mechanisms of the parasite mediated by MSP and PLC enzymes.
Therefore, studies have shown that MMP inhibitors like antibiotics, such as tetracycline and minocycline, prolong the animal's survival and decrease the influx of parasites and leukocytes in the brain [20,27,28]. Our recent studies focus on the evaluation of potential new drugs in the context of trypanosomatid infections. Thus, the mechanism of the antigenic variation of T. brucei seems to be an interesting target for the search for potential new antiparasitic molecules [43,44]. Consequently, understanding the mechanism of VSG release and all the involved molecules represents an important strategy for the development of new drugs against infections caused by T. brucei. In addition, a detailed analysis of all identified VSG conserved residues may provide new insights into host-parasite interactions, taking into account that released soluble VSGs interact directly with the immune system. Nevertheless, more studies are needed to validate these cleavage sites by MMPs.

Materials and Methods
The conserved domains for VSG sequences were retrieved from the Conserved Domain Database of the NCBI [45]. All T. brucei sequences were recovered from the NCBI protein database, then those sequences were divided considering the three subspecies: T. b. brucei (Tbb), T. b. gambiense (Tbg) and T. b. rhodesiense (Tbr). Data were sourced using the Reverse Position-Specific BLAST (RPS BLAST) tool v2.6.0 [46]. During the search, the NCBI CDD accessions cl03014 (a-VSG) and cl26244 (b-VSG) were used. In addition, the fragmented domains were discarded, and the minimum e-value threshold was 1.0 x 10 −5 .
The sequences that contain VSG domains were submitted to the PROSPER server to predict protease cleavage sites [46]. As an exclusivity criterion, the cleavage sites with a score greater than 0.8 were considered as probable sites for MMPs/MSPs. To demonstrate that the MMP/MSP sites present in T. brucei are very conserved, the standard sites for these proteases were retrieved from the MEROPS database [47] (Supplemental Materials 1, 2 and 3). The cleavage sites were searched in all extensions of the protein sequence. To illustrate the similarity of the cleave sites, an analysis in the WebLogo tool was performed [48]. Our approach can be summarized as follows in Figure 1. Methodologies overview: VSG domains A and B were analyzed by RPS BLAST after the VSG sequences were submitted to the PROSPER server to identify the cleavage sites. As a result, the protease cleavage sites were analyzed using the WebLogo tool to generate a consensus sequence. Figure 1. Methodologies overview: VSG domains A and B were analyzed by RPS BLAST after the VSG sequences were submitted to the PROSPER server to identify the cleavage sites. As a result, the protease cleavage sites were analyzed using the WebLogo tool to generate a consensus sequence.

Conclusions
The antigenic variation mechanism is a sophisticated phenomenon that contributes to chronicity in the context of infections caused by T. brucei. Trypanosomes manipulate their hosts by VSG switch mechanisms as a strategy to successfully elude the immune system. However, here, using VSG sequence analysis, the matrix metalloproteinases (MMPs) were able to recognize several VSG sequences from different subspecies of T. brucei. The interest in MMPs resulted from the observation that, besides their housekeeping role, they are involved in a wide range of physiological and pathological phenomena, including protozoa infections. Likewise, the VSG shedding mechanism by parasite protease is not fully understood. Therefore, we intended to analyze and identify VSG cleavage sites against protease present in the PROSPER server. Our study demonstrated that all the VSG sequences analyzed present conserved cleavage sites, which could be substrates for MMPs, such as MMP-2, MMP-3 and MMP-9.
T. brucei present thousands of surface VSGs that could hypothetically interact with MMPs. Thus, the mammalian host may contribute in the infection dynamics through the overexpression of MMPs. Nevertheless, more biological studies are required to determine the functions of MMPs in the context of T. brucei infection.
Supplementary Materials: The following are available online at http://www.mdpi.com/2076-0817/8/4/178/s1, List S1: Used sequences (a few of sequences were used two times because were found in more than one subspecies of Trypanosoma brucei), Figure S1: Cleavege sites for matrix metallopeptidase-2 retrieved from MEROPS database, Figure S2: Cleavage sites for matrix metallopeptidase-9 retrieved from MEROPS database, Figure S3: Cleavage sites for matrix metallopeptidase-3 retrieved from MEROPS database.