High-Mannose Specific Lectin and Its Recombinants from a Carrageenophyta Kappaphycus alvarezii Represent a Potent Anti-HIV Activity Through High-Affinity Binding to the Viral Envelope Glycoprotein gp120

Abstract

We previously reported that a high-mannose binding lectin KAA-2 from the red alga Kappaphycus alvarezii, which is an economically important species and widely cultivated as a source of carrageenans, had a potent anti-influenza virus activity. In this study, the full-length sequences of two KAA isoforms, KAA-1 and KAA-2, were elucidated by a combination of peptide mapping and complementary DNA (cDNA) cloning. They consisted of four internal tandem-repeated domains, which are conserved in high-mannose specific lectins from lower organisms, including a cyanobacterium Oscillatoria agardhii and a red alga Eucheuma serra. Using an Escherichia coli expression system, an active recombinant form of KAA-1 (His-tagged rKAA-1) was successfully generated in the yield of 115 mg per liter of culture. In a detailed oligosaccharide binding analysis by a centrifugal ultrafiltration-HPLC method with 27 pyridylaminated oligosaccharides, His-tagged rKAA-1 and rKAA-1 specifically bound to high-mannose N-glycans with an exposed α1-3 mannose in the D2 arm as the native lectin did. Predicted from oligosaccharide binding specificity, a surface plasmon resonance analysis revealed that the recombinants exhibit strong interaction with gp120, a heavily glycosylated envelope glycoprotein of HIV with high association constants (1.48 - 1.61 × 10(9) M(-1)). Native KAAs and the recombinants inhibited the HIV-1 entry at IC50s of low nanomolar levels (7.3-12.9 nM). Thus, the recombinant proteins would be useful as antiviral reagents targeting the viral surface glycoproteins with high-mannose N-glycans, and the cultivated alga K. alvarezii could also be a good source of not only carrageenans but also this functional lectin(s).

Keywords: Alga; Antiviral lectin; Carageenophyta; HIV; Kappaphycus alvarezii.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

A diagram of recombinant KAA-1 (rKAA-1) expression vector pET28a-rKAA1. rKAA-1 was expressed as a 6-His- and thrombin cleavage site-fused protein

Fig. 2

Isolation of PE-KAA-2 and PE-ESA-2 by reverse-phase HPLC. PE-KAA-2 (a) and PE-ESA-2 (b) were purified by RP-HPLC on YMC-Pack PROTEIN-RP column. The eluate was monitored by absorbance at 280 nm, and the active fraction represented by a bar was recovered. Deconvoluted mass spectra of purified PE-KAA-2 (a) and PE-ESA-2 (b) using ESI-MS (LCQ) are also represented. SDS-PAGE of PE-KAA-2 (lane 1) and PE-ESA-2 (lane 2) (c). Five micrograms of protein was loaded in each lane. M molecular weight marker

Fig. 3

Peptide fragments generated by trypsin digestion of PE-KAA-2 and PE-ESA-2. Peptide fragments produced by trypsin digestion of PE-KAA-2 (a) and PE-ESA-2 (b) were purified by RP-HPLC on TSKgel ODS-120T column, and their enlarged images (c) are shown to compare the peak patterns. The peaks were numbered in serial order of retention time. Peaks 11, 12, 15, 16 and 23 and peaks 13, 20, and 21 were unique for PE-KAA-2 and PE-ESA-2, respectively

Fig. 4

The predicted amino acid sequence of native KAA-2. The results of sequence prediction by peptide mapping in KAA-2 are shown using ESA-2 sequence as a template. The peptide fragments produced by trypsin digestion of PE-KAA-2 are represented with arrows. The peak numbers examined by Edman degradation and their determined sequences are underlined. The sequences whose peptide masses were consistent with those of ESA-2 are represented in bold. Unidentified amino acids in KAA-2 are represented as lower case letter of corresponding sequence in ESA-2. Shaded regions in the sequence were used to design the degenerated primers for cDNA cloning

Fig. 5

Nucleotide and deduced amino acid sequences of KAA-1. a Nucleotide and deduced amino acid sequences of the cDNA encoding KAA-1. The stop codon is shown as an asterisk. The italicized and nonitalicized numbers represent the positions of nucleotides and amino acids, respectively. Initiating methionine is indicated in italic. The 20 N-terminal amino acid sequence elucidated in Kawakubo et al. (1999) is underlined. b Comparison among deduced amino acid sequences of KAA-1 and KAA-2, and the sequence predicted by peptide mapping for native KAA-2 (see Fig. 2). Lower case letters in native KAA-2 sequence represent the corresponding sequences in ESA-2. Amino acid substitutions among KAAs were boxed. The nucleotide sequence data of KAA-1 and KAA-2 appear in the DDBJ, EMBL, and GenBank databases under accession number LC007080 and LC007081, respectively

Fig. 6

Alignment of amino acid sequences from KAAs and related proteins. Identical and similar residues among all proteins are indicated with a black and gray background, respectively. Amino acids that interact with mannopentaose (Manα1-6(Manα1-3)Manα1-6(Manα1-3)Man) of M8/9 core unit in Burkholderia oklahomensis agglutinin (BOA) are indicated with asterisks. Data were cited from Eucheuma serra agglutinin-2 (ESA-2, GenBank accession No. P84331), Burkholderia oklahomensis agglutinin (BOA, AIO69853), Myxococcus xanthus agglutinin (Myxobacterial hemagglutinin (MBHA), M13831), Oscillatoria agardhii agglutinin (OAA, P84330), and Pseudomonas fluorescens lectin (PFL, ABA72252)

Fig. 7

Preparation of rKAAs. a Nonreducing SDS-PAGE for purified His-rKAA-1 (lane 1) and rKAA-1 (lane 2). Five micrograms of each protein was loaded. M molecular weight marker. ESI-MS multiply charged spectra of purified His-rKAA-1 (b) and rKAA-1 (c) were shown

Fig. 8

Structures of PA-oligosaccharides used in this study. Closed circle, galactose; open circle, mannose; closed triangle, glucose; open triangle, fucose; closed square, N-acetylglucosamine; open square, N-acetylgalactosamine; closed diamond, N-acetylneuraminic acid

Fig. 9

Interaction analysis between rKAAs and various oligosaccharides by a centrifugal ultrafiltration-HPLC method. a Binding activities of rKAAs to PA-oligosaccharides. Binding activity was expressed as a ratio (%) of the amount of a bound oligosaccharide to that of an added oligosaccharide. The assay was performed in duplicate for each PA-oligosaccharide, and the activity is expressed as the average value from duplicate assays. The activities less than 10 % were cutoff for their insignificance. b The detailed binding activities of rKAAs with high-mannose-type N-glycans. Dashed boxes represent the nonreducing terminal α1-2-linked mannose residue in the D2 arm

Fig. 10

Interaction of KAAs with an HIV envelope glycoprotein gp120. SPR analyses for interaction between gp120 and His-rKAA-1 (a) or rKAA-1 (b) were analyzed with BIAcore2000. Each sensorgram represents the binding activity of rKAAs to gp120 on sensor chip. Ninety microliters of rKAA solutions of each represented concentration was injected into the flow cells at 30 μl/min for 3 min. The binding response (black line) in resonance units is plotted against time (seconds). Binding kinetics of the interactions between rKAAs and gp120 (overlaid red line) were calculated by fitting the data to Langmuir model for 1:1 binding. c Comparison of binding activity to gp120 among KAAs including His-rKAA-1, rKAA-1, KAA-1, and KAA-2. SPR analysis was performed using 0.98 nM of each lectin solution by the same method as described above