A novel recombinant antibody specific to full-length stromal derived factor-1 for potential application in biomarker studies

Abstract

Background: Stromal derived factor-1α (SDF-1α/CXCL12) is a chemokine that is up-regulated in diseases characterised by tissue hypoxia, including myocardial infarction, ischaemic cardiomyopathy and remote ischaemic conditioning (RIC), a technique of cyclical, non-injurious ischaemia applied remote from the heart that protects the heat from lethal ischaemia-reperfusion injury. Accordingly, there is considerable interest in SDF-1α as a potential biomarker of such conditions. However, SDF-1α is rapidly degraded and inactivated by dipeptidyl peptidase 4 and other peptidases, and the kinetics of intact SDF-1α remain unknown.

Methods & results: To facilitate investigation of full-length SDF-1α we established an ELISA using a novel recombinant human antibody we developed called HCI.SDF1. HCI.SDF1 is specific to the N-terminal sequence of all isoforms of SDF-1 and has a comparable KD to commercially available antibodies. Together with a detection antibody specific to the α-isoform, HCI.SDF1 was used to specifically quantify full-length SDF-1α in blood for the first time. Using RIC applied to the hind limb of Sprague-Dawley rats or the arms of healthy human volunteers, we demonstrate an increase in SDF-1α using a commercially available antibody, as previously reported, but an unexpected decrease in full-length SDF-1α after RIC in both species.

Conclusions: We report for the first time the development of a novel recombinant antibody specific to full-length SDF-1. Applied to RIC, we demonstrate a significant decrease in SDF-1α that is at odds with the literature and suggests a need to investigate the kinetics of full-length SDF-1α in conditions characterised by tissue hypoxia.

Fig 1. Representation of binding sites of HCI.SDF1α and commercial antibodies on full-length SDF-1α and SDF-1α after cleavage of its N-terminal di-peptide by DPP4.

Fig 2. SDF-1α sequence according to species.

Alignment of SDF-1α from humans, rats and mice demonstrates the N-terminal sequence is highly conserved between species. Accession numbers NP_001171605.1, NP_001029055.1 and NP_001012495.1 were used, respectively. Sequences in red represent the sequence of interest used to generate HCI.SDF1. Blue residues represent differences in the primary sequence between species; note that none of these are at the N-terminal end.

Fig 3. In vivo rat protocols.

Remote ischaemic conditioning (RIC) was applied to the rat hind limb in the following groups: A) Control: 35 min stabilisation; B) RIC: 5 min stabilisation, 3 cycles of 5 min ischemia and 5 min reperfusion. Blood samples were taken either immediately after RIC (sample 1) or 60 min thereafter (sample 2).

Fig 4. Laser Doppler blood flow assessment in the subject hind limb during RIC and reperfusion.

Panel A shows a colour photograph of the subject hind limb. Panel B demonstrates complete cessation of blood flow during RIC ischaemia (C) and restoration after reperfusion (D).

Fig 5. Representative HCI.SDF1 and MAB350 binding kinetics data.

Representative data obtained from label-free sensor evaluation of HCI.SDF1 and MAB350 with Amine Reactive Second-Generation (AR2G) Biosensors.

Fig 6. ELISA for intact and cleaved SDF-1α using HCI.SDF1 and MAB350.

An ELISA assay demonstrates that HCI.SDF1 is specific for active SDF-1α, with very low signal in the absence of intact SDF-1α, whereas there is substantial residual signal in the absence of intact SDF-1α when using the commercial antibody MAB350.

Fig 7. HCI.SDF1 used for Western blotting.

Western blots demonstrated a concentration-signal relationship when HCI.SDF1 was used to detect varying concentrations of rhSDF-1α.

Fig 8. The effect of sample type on SDF-1α concentration.

Serum samples, collected by allowing whole blood to clot for 10 min prior to centrifuging to remove the clot, had significantly more active SDF-1α than either platelet-free (PFP) or unfractionated plasma samples that were collected into a citrated tube and prepared by differential centrifugation, when measured using either capture antibody as indicated.

Fig 9. Effect of RIC on intact and cleaved SDF-1α.

Following RIC, ELISA using commercial antibodies detects an increase in plasma SDF-1α levels whereas HCI.SDF1α, which is more specific for full-length SDF-1α, measures a decrease immediately after RIC that is restored to baseline by 1 h.

Fig 10. Effect of RIC on intact SDF-1α in humans.

Plasma levels of full-length SDF-1α (quantified using HCI.SDF1α ELISA), decreased significantly in healthy male volunteers subjected to RIC.