Displaying publications 1 - 20 of 22 in total

Abstract:
Sort:
  1. Perera D, Shimizu H, Yoshida H, Tu PV, Ishiko H, McMinn PC, et al.
    J Med Virol, 2010 Apr;82(4):649-57.
    PMID: 20166171 DOI: 10.1002/jmv.21652
    The VP4, VP2, and VP1 gene regions were evaluated for their usefulness in typing human enteroviruses. Three published RT-PCR primers sets targeting separately these three gene regions were used. Initially, from a total of 86 field isolates (36 HEV-A, 40 HEV-B, and 10 HEV-C) tested, 100% concordance in HEV-A was identified from all three gene regions (VP4, VP2, and VP1). However, for HEV-B and HEV-C viruses, only the VP2 and VP1 regions, and not VP4, showed 100% concordance in typing these viruses. To evaluate further the usefulness of VP4 in typing HEV-A enteroviruses, 55 Japanese and 203 published paired VP4 and VP1 nucleotide sequences were also examined. In each case, typing by VP4 was 100% in concordance with typing using VP1. Given these results, it is proposed that for HEV-A enteroviruses, all three gene regions (VP4, VP2, and VP1), would be useful for typing these viruses. These options would enhance the capability of laboratories in identifying these viruses and would greatly help in outbreaks of hand, foot, and mouth disease.
    Matched MeSH terms: Viral Structural Proteins/genetics*
  2. Nurulfiza I, Hair-Bejo M, Omar AR, Aini I
    Acta Virol., 2006;50(1):45-51.
    PMID: 16599185
    Three isolates of Infectious bursal disease virus (IBDV), designated UPM04178, UPM04190 and UPM04238, were obtained from severe outbreaks of infectious bursal disease (IBD) in Malaysia in 2004. The hypervariable region (HPVR) of VP2 gene of these isolates was sequenced. The obtained sequences were compared with those of other isolates. The highest similarity (98%) concerning both nucleotide and amino acid sequences was found to very virulent IBDV (vvIBDV) strains. Phylogenetic analysis revealed clustering of the three isolates with vvIBDV strains. Evolutionary relatedness of the three isolates to vvIBDV strains was demonstrated by three phylogenetic methods: bootstrap values of 100%, 95% and 90% for nucleotide sequences and those of 58%, 86% and 96% for amino acid sequences were obtained by the distance, maximum parsimony and maximum likehood methods, respectively. It is concluded that UPM04178, UPM04190 and UPM04238 are vvIBDV isolates of serotype 1, which originate from a common ancestor of IBDV strains present in Malaysia.
    Matched MeSH terms: Viral Structural Proteins/genetics
  3. Perera D, Yusof MA, Podin Y, Ooi MH, Thao NT, Wong KK, et al.
    Arch Virol, 2007;152(6):1201-8.
    PMID: 17308978
    A phylogenetic analysis of VP1 and VP4 nucleotide sequences of 52 recent CVA16 strains demonstrated two distinct CVA16 genogroups, A and B, with the prototype strain being the only member of genogroup A. CVA16 G-10, the prototype strain, showed a nucleotide difference of 27.7-30.2% and 19.9-25.2% in VP1 and VP4, respectively, in relation to other CVA16 strains, which formed two separate lineages in genogroup B with nucleotide variation of less than 13.4% and less than 16.3% in VP1 and VP4, respectively. Lineage 1 strains circulating before 2000 were later displaced by lineage 2 strains.
    Matched MeSH terms: Viral Structural Proteins/genetics
  4. Ong ST, Tan WS, Hassan SS, Mohd Lila MA, Yusoff K
    J. Biochem. Mol. Biol. Biophys., 2002 Oct;6(5):347-50.
    PMID: 12385971
    The coding region of the nucleocapsid (N) gene was amplified from the viral RNA and inserted into the bacterial expression vector, pTrcHis2, for intracellular expression in three Escherichia coli strains: TOP 10, BL 21 and SG 935. The N protein was expressed as a fusion protein containing the myc epitope and His-tag at its C-terminal end. The amount of the fusion protein expressed in strain SG 935 was significantly higher than the other two strains, and was detected by the anti-myc antibody, anti-His and swine anti-NiV serum. Hence, the N(fus) protein produced in E. coli could serve as an alternative antigen for the detection of anti-NiV in swine.
    Matched MeSH terms: Viral Structural Proteins/genetics
  5. Abdul Mutalib NE, Mat Isa N, Alitheen NB, Song AA, Rahim RA
    Plasmid, 2014 May;73:26-33.
    PMID: 24780699 DOI: 10.1016/j.plasmid.2014.04.003
    Plasmid DNAs isolated from lactic acid bacteria (LAB) such as Lactococcus lactis (L. lactis) has been gaining more interests for its positive prospects in genetic engineering-related applications. In this study, the lactococcal plasmid, pNZ8048 was modified so as to be able to express multiple genes in the eukaryotic system. Therefore, a cassette containing an internal ribosome entry site (IRES) was cloned between VP2 gene of a very virulent infectious bursal disease (vvIBDV) UPM 04190 of Malaysian local isolates and the reporter gene, green fluorescent protein (GFP) into pNZ:CA, a newly constructed derivative of pNZ8048 harboring the cytomegalovirus promoter (Pcmv) and polyadenylation signal. The new bicistronic vector, denoted as pNZ:vig was subjected to in vitro transcription/translation system followed by SDS-PAGE and Western blot analysis to rapidly verify its functionality. Immunoblotting profiles showed the presence of 49 and 29kDa bands that corresponds to the sizes of the VP2 and GFP proteins respectively. This preliminary result shows that the newly constructed lactococcal bicistronic vector can co-express multiple genes in a eukaryotic system via the IRES element thus suggesting its feasibility to be used for transfection of in vitro cell cultures and vaccine delivery.
    Matched MeSH terms: Viral Structural Proteins/genetics
  6. Phong SF, Hair-Bejo M, Omar AR, Aini I
    Avian Dis, 2003 Jan-Mar;47(1):154-62.
    PMID: 12713171
    The VP2 hypervariable region of P97/302 local infectious bursal disease virus (IBDV) isolate was amplified by the reverse transcriptase (RT) nested polymerase chain reaction (PCR) and cloned. This region of P97/302 local isolate was sequenced and compared with eight other reported IBDV sequences. The result showed that P97/302 IBDV was most identical to the reported very virulent IBDV strains because it has amino acid substitutions at positions 222, 256, 294, and 299, which encode alanine, isoleucine, isoleucine, and serine, respectively. This region can be digested with restriction enzymes of Taq1, Sty1, Ssp1 but not with Sac1. The P97/302 isolate was then used for the optimization of RT nested PCR enzyme-linked immunosorbent assay (ELISA). The RT nested PCR ELISA was able to detect 10(-4) dilution of the infected bursa homogenates and was 10 times more sensitive when compared with the agarose gel detection method. The RT nested PCR ELISA can detect up to 0.48 ng of the PCR product. The specificity of this nested PCR ELISA was also high (100%).
    Matched MeSH terms: Viral Structural Proteins/genetics*
  7. Yahiro T, Takaki M, Chandrasena TGAN, Rajindrajith S, Iha H, Ahmed K
    Infect Genet Evol, 2018 11;65:170-186.
    PMID: 30055329 DOI: 10.1016/j.meegid.2018.07.014
    A human-porcine reassortant rotavirus, strain R1207, was identified from 74 group A rotaviruses detected in 197 (37.6%) stool samples collected from patients who attended a tertiary care hospital in Ragama, Sri Lanka. This is the first report of a human-porcine reassortant rotavirus in Sri Lanka. The patient was a 12-month-old boy who had been hospitalized with fever and acute diarrhea with a duration of 6 days. The family had pigs at home before the birth of this boy. However, the neighbors still practice pig farming. The genotype constellation of R1207 was G4-P[6]-I1-R1-C1-M1-A1-N1-T1-E1-H1. This is based on the assignment of all the eleven gene segments a full genome-based genotyping system. R1207 showed a 4-2-3-2 genomic electrophoretic migration pattern, which is characteristic of group A rotaviruses. Our analyses revealed that five (NSP2, NSP4, VP1, VP2, and VP7) of the 11 genes were closely related to the respective genes of porcine strains. Although the remaining six genes (NSP1, NSP3, NSP5, VP3, VP4, and VP6) were related to human strains, with the exception of the gene sequence of NSP1, all of these human strains were human-porcine reassortants. With a genogroup 1 genetic backbone, this strain was possibly formed via multiple genetic reassortments. We do not know whether this strain is circulating in pigs, as no data are available on porcine rotaviruses in Sri Lanka. Surveillance should be strengthened to determine the epidemiology of this genotype of rotavirus in Sri Lanka and to assess whether the infection was limited or sustained by ongoing human-to-human transmission.
    Matched MeSH terms: Viral Structural Proteins/genetics*
  8. Jikal M, Mori D, Yusoff AF, Rai SB, Mukhsam MH, Ali I, et al.
    Am J Trop Med Hyg, 2021 07 12;105(3):777-782.
    PMID: 34255740 DOI: 10.4269/ajtmh.21-0036
    Foodborne outbreaks of hepatitis A virus (HAV) are most commonly associated with fresh and frozen produce and with various types of shellfish. Alcoholic beverage-borne outbreaks of hepatitis A are extremely rare. Here, we report an outbreak of hepatitis A associated with the consumption of a traditional wine at a funeral ceremony in the Sabah state of Malaysian Borneo. Confirmed cases were determined by serum anti-HAV immunoglobulin M and/or for fecal HAV by reverse transcription polymerase chain reaction (RT-PCR). The amplicons of RT-PCR were subjected to nucleotide sequencing followed by phylogenetic analysis. We conducted a 1:2 case-control study to identify the possible exposure that led to the outbreak. Sixteen patients met the case definition, they were 18 to 58 years old and 90% of them were males. The case-control study showed that the consumption of nipa palm wine during the ceremony was significantly associated (P = 0.0017) with hepatitis A infection (odds ratio, 5.44; 95% CI, 1.80-16.43). Untreated river water was used to dilute the traditional wine, which was assumed to be the source of the infection. Phylogenetically, these viruses belonged to genotype IA and formed an independent cluster with strains from Taiwan, Japan, and the Philippines. This strain might be an emerging HAV in Asian countries. Environmental assessments were performed and environmental samples were negative for HAV. The incidence of hepatitis A in Sabah was also determined and it was 0.795/100,000 population. Strict monitoring of traditional wine production should be implemented by the local authority to prevent future outbreaks.
    Matched MeSH terms: Viral Structural Proteins/genetics
  9. Lee SY, Park ME, Kim RH, Ko MK, Lee KN, Kim SM, et al.
    Vaccine, 2015 Jan 29;33(5):664-9.
    PMID: 25528521 DOI: 10.1016/j.vaccine.2014.12.007
    Of the seven known serotypes of foot-and-mouth disease virus (FMDV), type A has the most diverse variations. Genetic variations also occur frequently at VP1, VP2, VP3, and VP4 because these proteins constitute the viral capsid. The structural proteins of FMDV, which are closely related to immunologic correlations, are the most easily analyzed because they have highly accessible information. In this study we analyzed the type A vaccine viruses by alignment of available sequences in order to find appropriate vaccine strains. The matching rate of ASIA topotype-specific sites (20 amino acids) located on the viral surface, which are mainly VP1 and VP2, was highly related to immunologic reactivity. Among the available vaccines analyzed in this study, we suggest that A Malaysia 97 could be used as a vaccine virus as it has the highest genetic similarity and immunologic aspects to field strains originating in East Asia.
    Matched MeSH terms: Viral Structural Proteins/genetics
  10. Le VP, Nguyen T, Lee KN, Ko YJ, Lee HS, Nguyen VC, et al.
    Vet Microbiol, 2010 Jul 29;144(1-2):58-66.
    PMID: 20097490 DOI: 10.1016/j.vetmic.2009.12.033
    Foot-and-mouth disease (FMD) is a major cause of endemic outbreaks in Vietnam in recent years. In this work, six serotype A foot-and-mouth disease viruses (FMDV), collected from endemic outbreaks during January and February of 2009 in four different provinces in Vietnam, were genetically characterized for their complete genome sequences. Genetic analysis based on the complete viral genome sequence indicated that they were closely related to each other and shared 99.0-99.8% amino acid (aa) identity. Genetic and deduced aa analysis of the capsid coding gene VP1 showed that the six Vietnamese strains were all classified into the genotype IX from a total of 10 major genotypes worldwide, sharing 98.1-100% aa identity each other. They were most closely related to the type A strains recently isolated in Laos (A/LAO/36/2003, A/LAO/1/2006, A/LAO/6/2006, A/LAO/7/2006, and A/LAO/8/2006), Thailand (A/TAI/2/1997 and A/TAI/118/1987), and Malaysia (A/MAY/2/2002), sharing 88.3-95.5% nucleotide (nt) identities. In contrast, Vietnamese type A strains showed low nt identities with the two old type A FMDVs, isolated in 1960 in Thailand (a15thailand iso43) and in 1975 in the Philippines (aphilippines iso50), ranging from 77.3 to 80.9% nt identity. A multiple alignment based on the deduced amino acid sequences of the capsid VP1 coding gene of type A FMDV revealed three amino acid substitutions between Vietnamese strains and the strains of other Southeast Asian countries (Laos, Thailand, Malaysia, and the Philippines). Alanine was replaced by valine at residue 24, asparagine by arginine at residue 85, and serine by threonine at residue 196. Furthermore, type A FMDV strains recently isolated in Vietnam, Laos, Thailand, and Malaysia all have one amino acid deletion at residue 140 of the capsid VP1 protein compared with the two old type A FMDV strains from Thailand and the Philippines as well as most other type A representatives worldwide. This article is the first to report on the comprehensive genetic characterization of type A FMDV circulating in Vietnam.
    Matched MeSH terms: Viral Structural Proteins/genetics
  11. Sharma K, Hair-Bejo M, Omar AR, Aini I
    Acta Virol., 2005;49(1):59-64.
    PMID: 15929400
    Two Infectious bursal disease virus (IBDV) isolates, NP1SSH and NP2K were obtained from a severe infectious bursal disease (IBD) outbreak in Nepal in 2002. The hypervariable (HV) region of VP2 gene (1326 bp) of the isolates was generated by RT-PCR and sequenced. The obtained nucleotide sequences were compared with those of twenty other IBDV isolates/strains. Phylogenetic analysis based on this comparison revealed that NP1SSH and NP2K clustered with very virulent (vv) IBDV strains of serotype 1. In contrast, classical, Australian classical and attenuated strains of serotype 1 and avirulent IBDV strains of serotype 2 formed a different cluster. The deduced amino acid sequences of the two isolates showed a 98.3% identity with each other and 97.1% and 98.3% identities, respectively with very virulent IBDV (vvIBDV) isolates/strains. Three amino acids substitutions at positions 300 (E-->A), 308 (I-->F) and 334 (A-->P) within the HV region were common for both the isolates. The amino acids substitutions at positions 27 (S-->T), 28 (I-->T), 31 (D-->A), 36 (H-->Y), 135 (E-->G), 223 (G-->S), 225 (V-->I), 351 (L-->I), 352 (V-->E) and 399 (I-->S) for NP1SSH and at position 438 (I-->S) for NP2K were unique and differed from other IBDV isolates/strains. NP1SSH and NP2K showed highest similarity (97.8%) with the BD399 strain from Bangladesh as compared with other vvIBDV isolates/strains. We conclude that the NP1SSH and NP2K isolates of IBDV from Nepal represent vvIBDV of serotype 1.
    Matched MeSH terms: Viral Structural Proteins/genetics*
  12. Chong LK, Omar AR, Yusoff K, Hair-Bejo M, Aini I
    Acta Virol., 2001;45(4):217-26.
    PMID: 11885928
    The complete nucleotide sequences encoding precursor polyprotein (VP2-VP3-VP4) and VP5 of a highly virulent (hv) infectious bursal disease virus (IBDV), UPM97/61 was determined. Comparison of the deduced amino acid sequences with the published ones revealed 8 common amino acid substitutions, which were found only in the hv IBDV including the UPM97/61 strain. Three of the amino acid substitutions (222 Ala, 256 Ile and 294 Ile) were used as a marker for determining hv IBDV strains. The other five substitutions (685 Asn, 715 Ser, 751 Asp, 990 Val and 1005 Ala) were also conserved in hv IBDV strains isolated in various countries. UPM97/61 strain demonstrated also 8 unique amino acid substitutions of which 3 were in VP2, 4 in VP3 and 1 in VP4. There was 1 unique amino acid substitution in VP5 at position 19 (Asp-->Gly) not found in other strains. However, all the strains have a conserved 49 Arg. The amino acid sequence of UPM97/61 strain differed by 1.09% from the Japanese (OKYM) and Hong Kong (HK46) strains, and by 1.48% from the Israeli (IBDVKS) and European (UK661) strains. Hence, UPM97/61 is more closely related to the hv strains from Asia. However, phylogenetic analysis indicated that the origin of UPM97/61 might be the same as that of other hv strains isolated from other parts of the world.
    Matched MeSH terms: Viral Structural Proteins/genetics*
  13. Hoque MM, Omar AR, Hair-Bejo M, Aini I
    J. Biochem. Mol. Biol. Biophys., 2002 Apr;6(2):93-9.
    PMID: 12186763
    Previously we have shown that very virulent infectious bursal disease viruses (vvIBDV) that are SspI, TaqI and StyI positive (92/04, 97/61 and 94/B551) but not SspI and TaqI positive and StyI negative (94/273) cause high mortality, up to 80% in specific-pathogen-free chickens with significant damage of the bursal as well as nonbursal tissues. In this study, we sequenced the VP2 gene (1351 bp) of the 92/04, 94/273 and 94/B551 and compared them with other IBDV strains. All the isolates have the unique amino acid residues at positions 222A, 256I, 294I and 299S found in other vvIBDV strains. The deduced VP2 amino acids encoded by 92/04 is identical to the vvIBDV strains from Israel (IBDVKS), Japan (OKYM) and Europe (UK661), whereas the 94/273 and 94/B551 isolates have one to three amino acid substitutions. The 94/273 has two amino acid substitutions at positions 254 G to S and at 270 A to E that have not been reported before from vvIBDV strains. The 94/B551 also has one amino acid substitution at position 300 E to S, which is uncommon among other vvIBDV strains. However, phylogenetic analysis suggested that the isolates are very close to each other and all of them may have derived from the same origin as vvIBDV strains isolated from China, Japan and Europe. Even though antigenic index analysis of the 94/273 and 94/B551 indicated that the isolates are unique compared to other IBDV strains, their antigenic variation remain to be determined by monoclonal antibody study.
    Matched MeSH terms: Viral Structural Proteins/genetics*
  14. Cecilia D, Gould EA
    Virology, 1991 Mar;181(1):70-7.
    PMID: 1704661
    The Sarawak strain of Japanese encephalitis virus (JE-Sar) is virulent in 3-week-old mice when inoculated intraperitoneally. The nucleotide sequence for the envelope glycoprotein (E) of this virus was determined and compared with the published sequences of four other strains. There were several silent nucleotide differences and five codon changes. Monoclonal antibodies (MAbs) against the E protein of JE-Sar virus were prepared and characterized. MAb-resistant mutants of JE-Sar were selected to determine if mutations in the E protein gene could affect its virulence for mice. Eight mutants were isolated using five different MAbs that identified virus-specific or group-reactive epitopes on the E protein. The mutants lost either complete or partial reactivity with selecting MAb. Several showed decreased virulence in 3-week-old mice after intraperitoneal inoculation. Two (r27 and r30) also showed reduced virulence in 2-week-old mice. JE-Sar and the derived mutants were comparable in their virulence for mice, when inoculated intracranially. Mutant r30 but not r27 induced protective immunity in adult mice against intracranial challenge with parent virus. However, r27-2 did induce protective immunity against itself. Nucleotide sequencing of the E coding region for the mutants revealed single base changes in both r30 and r27 resulting in a predicted change from isoleucine to serine at position 270 in r30 and from glycine to aspartic acid at position 333 in r27. The altered capacity of the mutants to induce protective immunity is consistent with the immunogenicity changes predicted by computer analysis using the Protean II program.
    Matched MeSH terms: Viral Structural Proteins/genetics
  15. Kong LL, Omar AR, Hair-Bejo M, Aini I, Seow HF
    Arch Virol, 2004 Feb;149(2):425-34.
    PMID: 14745606
    The deduced amino acid sequences of segment A and B of two very virulent Infectious bursal disease virus (vvIBDV) isolates, UPM94/273 and UPM97/61 were compared with 25 other IBDV strains. Twenty amino acid residues (8 in VP1, 5 in VP2, 2 in VP3, 4 in VP4, 1 in VP5) that were common to vvIBDV strains were detected. However, UPM94/273 is an exceptional vvIBDV with usual amino acid substitutions. The differences in the divergence of segment A and B indicated that the vvIBDV strains may have been derived from genetic reassortment of a single ancestral virus or both segments have different ability to undergo genetic variation due to their different functional constraints.
    Matched MeSH terms: Viral Structural Proteins/genetics
  16. Tiong V, Lam CW, Phoon WH, AbuBakar S, Chang LY
    Jpn J Infect Dis, 2017 Jan 24;70(1):26-31.
    PMID: 27169942 DOI: 10.7883/yoken.JJID.2015.501
    The genes for Nipah virus (NiV) proteins were amplified from viral RNA, cloned into the plasmid pTriEx-3 Hygro, expressed, and purified using immobilized metal affinity chromatography. The recombinant N, F, and G NiV proteins (rNiV-N, rNiV-F, and rNiV-G), were successfully expressed in Escherichia coli and purified with a yield of 4, 16, and 4 mg/L, respectively. All 3 recombinant viral proteins reacted with all 19 samples of NiV-positive human sera. The rNiV-N and rNiV-G proteins were the most immunogenic. The recombinant viral proteins did not react with any of the 12 NiV-negative sera. However, serum from a patient with a late-onset relapsing NiV infection complication was found to be primarily reactive to rNiV-G only. Additionally, there is a distinctive variation in the profile of antigen-reactive bands between the sample from a case of relapsing NiV encephalitis and that of acute NiV infection. The overall findings of this study suggest that the recombinant viral proteins have the potential to be developed further for use in the detection of NiV infection, and continuous biosurveillance of NiV infection in resource-limited settings.
    Matched MeSH terms: Viral Structural Proteins/genetics
  17. Kong LL, Omar AR, Hair Bejo M, Ideris A, Tan SW
    J Virol Methods, 2009 Nov;161(2):271-9.
    PMID: 19591873 DOI: 10.1016/j.jviromet.2009.06.023
    A SYBR Green I based one-step real-time reverse transcriptase polymerase chain reaction was developed for the detection and differentiation of very virulent (vv) and classical strains of infectious bursal disease virus (IBDV). The assay showed high PCR efficiency >93% and high reproducibility with coefficient of variation less than 0.5%. When tested on characterized IBDV strains, the very virulent and classical-specific primers detected accurately only vvIBDV and classical IBDV strains, respectively. The diagnostic efficacy of the assay was also tested on 140 bursal samples from experimental infection and 37 bursal samples from cases suspected of IBD. The assay was able to detect IBDV from bursal samples collected at days 3 and 5 post-infection with the vvIBDV strain UPM94/273 and the classical IBDV strain D78. The assay was also able to detect bursal samples infected dually with D78 and UPM94/273. The melting temperature values of the amplification products from the classical and very virulent viral infection were statistically significant (P<0.05). The specificity of the assay for detecting IBDV from suspected cases was confirmed by sequence analysis of the VP2 gene. The assay showed high sensitivity since bursal samples which were negative for IBDV were confirmed by virus isolation and PCR amplification. Hence, the new assay offers an attractive method for rapid detection of strains of IBDV.
    Matched MeSH terms: Viral Structural Proteins/genetics
  18. Omar AR, Kim CL, Bejo MH, Ideris A
    J Vet Sci, 2006 Sep;7(3):241-7.
    PMID: 16871018
    The ability of a heat-inactivated whole virus from a highly virulent infectious bursal disease virus (hvIBDV) and VP2 protein from hvIBDV expressed in E. coli provided protection against a hvIBDV challenge in specificpathogen-free (SPF) chickens. Six out of seven chickens that were injected three times with crude VP2 protein developed significant antibody titer against IBDV. However, only four out of the seven chickens survived the hvIBDV challenge. Despite showing low antibody titer profiles, all chickens immunized with the heat-inactivated whole virus also survived the challenged with hvIBDV. However, all of these chickens had bursal atrophy and mild to moderate depletion of lymphocytes. Thus, antibodies raised against IBDV VP2 protein expressed in E. coli and denatured IBDV proteins induced some degree of protection against mortality but not against bursal damage following challenge with hvIBDV.
    Matched MeSH terms: Viral Structural Proteins/genetics
Filters
Contact Us

Please provide feedback to Administrator (afdal@afpm.org.my)

External Links