Reliable and sensitive detection of glycosaminoglycan chains with immunoblots

Abstract

Complex glycans play vital roles in many biological processes, ranging from intracellular signaling and organ development to tumor growth. Glycan expression is routinely assessed by the application of glycan-specific antibodies to cells and tissues. However, glycan-specific antibodies quite often show a large number of bands on immunoblots and it is hard to interpret the data when reliable controls are lacking. This limits the scope of glycobiology studies and poses challenges for replication. We sought to resolve this issue by developing a novel strategy that utilizes an immunoreaction enhancing technology to vastly improve the speed and quality of glycan-based immunoblots. As a representative case study, we used chondroitin sulfate glycosaminoglycan (CS-GAG) chains as the carbohydrate target and a monoclonal antibody, CS-56, as the probe. We discovered that preincubation of the antibody with its antigenic CS-GAG chain distinguishes true-positive signals from false-positive ones. We successfully applied this strategy to 10E4, a monoclonal anti heparan sulfate GAGs (HS-GAGs) antibody, where true-positive signals were confirmed by chemical HS-GAG depolymerization on the membrane. This evidence that glycan-specific antibodies can generate clear and convincing data on immunoblot with highly replicable results opens new opportunities for many facets of life science research in glycobiology.

Keywords: chondroitin sulfate proteoglycan; glycosaminoglycans; heparan sulfate proteoglycan; immunoblot; western blot.

Fig. 1

Optimization of Immunoblot with CS-56 with a traditional method. (A) Amount of antigen required for CS-56 immunoblot. Different amounts of brain lysates obtained from 7-week-old mice prepared in SDS were separated and transferred to a PVDF membrane, followed by 5 h incubation with CS-56 (1:10,000 dilution with 10% CGS-1). Asterisk indicates nonspecific signal detected by anti-mouse IgM antibody since it was detected in the absence of CS-56 (B). (B) Specificity of CS-56 signals. Brain lysates obtained from 7-week old mice prepared in 8 M urea (20 μg) were separated on SDS-PAGE under reducing conditions and transferred to a PVDF membrane, followed by blocking with 10% skim milk in PBS-T. Immunoblot was performed as follows; No CS-56 (1), CS-56 diluted with 10% CGS-1, (2), diluted CS-56 was preincubated with either 10 ng/mL CS-C (3) or heparin (4) for 30 min at 22°C. The membranes were imaged as a single image and dotted lines indicate the border of individual membranes. Asterisks indicate nonspecific signal detected by anti-mouse IgM antibody. L, brain lysates; M, protein marker.

Fig. 2

(A) Conditions for ChABC digestion. ChABC activities in different buffers were compared using CS-A as a substrate. Blue; 100 mM Tris (pH 8.0), Black; 1.0 M urea in 100 mM Tris (pH 8.0), Red; PBS, Green; cOmplete™ Lysis-M, Brown; 0.1% Triton ×100 in 100 mM Tris (pH 8.0). (B) Disappearance/appearance of CS-56 and BE-123 epitopes upon ChABC digestion. Biotinylated CS-C treated with ChABC (0.0625 unit/mL) for indicated periods was immobilized and solid-phase binding experiments were performed with CS-56 (0.7 nM) and BE-123 (0.3 nM). Data are expressed as average ± SD. (C and D) ChABC digestion of brain lysates. Brain lysates (20 μg) obtained from P1 mice were prepared in 8 M urea. After reducing urea concentration to 1 M, samples were treated at 37°C for 16 h (C) and 36 h (D) in the presence (lane 1) or absence (lane 2) of ChABC. After acetone precipitation, proteins were separated by SDS-PAGE, followed by the optimized CS-56 immunoblot (left). After stripping the antibody, the membrane was reprobed by BE-123 (right). Asterisk indicates nonspecific signal detected by anti-mouse IgM antibody.

Fig. 3

Optimization of Immunoblot with CS-56 with a rapid method. (A) Amount of antigen required for CS-56 immunoblot. Different amounts of brain lysates obtained from 7-week-old mice prepared in SDS were separated and transferred to a PVDF membrane, followed by blocking with 10% skim milk in PBS-T for 22 h. CS-56 was diluted (1:5000) with 10% CGS-1 containing skim milk solution and incubated with PVDF membrane for 30 min with CDR method. (B) Incubation time required for CS-56 immunoblot. Brain lysates (10 μg/lane) prepared in SDS were separated and transferred to PVDF membrane, followed by blocking with 10% skim milk in PBS-T for 22 h at 4°C with agitation. (1) Traditional incubation (2 h) and CDR incubation for (2) 2 h, (3) 30 min and (4) 10 min were performed with CS-56 (1:5000 dilution with 10% CGS-1 containing skim milk solution). The membranes were imaged as a single image and dotted lines indicate the border of individual membranes. L, brain lysates; M, protein marker. (C) Effect of diluent for CS-56. Brain lysates (10 μg/lane) prepared in SDS were subjected to CS-56 immunoblot with CDR method. CS-56 was diluted (1:5000) in (1) 2.8 mL of 10% GCS-1 containing 0.2 mL of 5% skim milk in PBS-T, (2) 2.8 mL of 10% GCS-1 containing 0.2 mL of 5% gelatin from cold water fish in PBS-T, (3) 3 mL of 10% CGS-1 and (4) 3 mL of 100% CGS-1. The membranes were imaged as a single image and dotted lines indicate the border of individual membranes. L, brain lysates; M, protein marker. (D) Effect of blocking solution. Brain lysates (10 μg/lane) prepared in SDS were separated and transferred to a PVDF membrane. Membranes were blocked for 22 h at 4°C with agitation with (1) 10% skim milk in PBS-T, (2) 5% BSA in PBS-T, (3) PVDF Blocking Reagent for Can Get Signal, and (4) 5% gelatin from cold water fish in PBS-T. Immunoblot was performed as A. The membranes were imaged as a single image and dotted lines indicate the border of individual membranes. L, brain lysates; M, protein marker.

Fig. 4

Specificity of CS-56 signals. (A) Brain lysates (10 μg/lane) prepared in SDS were separated and transferred to a PVDF membrane. Membranes were blocked with 10% skim milk in PBS-T for 22 h at 4°C with agitation. CS-56 diluted (1:5000) with 10% CGS-1 containing skim milk solution was preincubated at 22°C for 30 min with either 500 ng/mL CS-C, heparin, or CS-C pretreated with ChABC. Immunoblot was performed as described in Figure 3A. The membranes were imaged as a single image and dotted lines indicate the border of individual membranes. (B) Aggrecan treated with or without ChABC was subjected to immunoblot with CS-56 as described in Figure 3A and BE-123 as described in Figure 2C.

Fig. 5

Effect of nitrous acid on 10E4 epitope. (A) Biotinylated heparin was treated with NaNO₂ ± acetic acid and immobilized to streptavidin-coated plate. Solid-phase binding assays were performed with 10E4. Data are expressed as an average ± SD. (B) HeLa cell lysates (10 μg/lane) was separated and transferred to a PVDF membrane. The membranes were treated with NaNO₂ alone or NaNO₂ together with acetic acid, followed by blocking with PVDF Blocking Reagent for Can Get Signal for 22 h at 22°C with agitation. 10E4 was diluted (1:3000) with 10% CGS-1 containing gelatin from cold water fish, and the incubation was performed with CDR method for 30 min. After acquiring the image, the bound antibody was stripped, and the membrane was reprobed by anti-ß actin antibody. The membranes were imaged as a single image and dotted lines indicate the border of individual membranes.