Expression, purification of recombinant DEFA5 protein and generation of hybridoma cell lines

The synthetic gene encoding DEFA5 was successfully constructed with a size of 621 bp and inserted into pET30a. The recombinant plasmid was verified by DNA sequencing and restriction enzyme digestion. The prokaryotic expression plasmid, pET30a-DEFA5, was transformed into E. coli BL21 Star™. The bacterial lysate was centrifuged, and the supernatant was loaded onto an affinity column to obtain the purified target protein. The purity of the DEFA5 protein obtained using E. coli system was analyzed by SDS-PAGE and was found to be > 90%, with an apparent molecular weight of 13 kDa, which was consistent with the prediction (Fig. 1A). Western blot analysis showed that the expressed DEFA5 protein reacted with the antibody, indicating its biological activity (Fig. 1A). SDS-PAGE analysis demonstrated that the two affinity-purified anti-DEFA5 1A8 and 4F5 mAbs showed a heavy chain band at 50 kDa and a light chain band at 25 kDa, respectively (Fig. 1B, C). To develop anti-DEFA5 1A8 and 4F5 mAbs healthy Balb/c mice were immunized with purified DEFA5 (Fig. 2A). The blood sera of the mice were evaluated by ELISA to determine anti-DEFA5 antibody titers. The ELISA results indicated that recombinant DEFA5, as an immune antigen, successfully induced the generation of higher antibody titers. Spleen cells harvested from immunized mice with the highest serum titer were fused with Sp2/0 myeloma cells and two hybrid cell lines against DEFA5 were successfully obtained (Fig. 2B). Supernatants from positive cells were evaluated for antibody production using an indirect ELISA (Fig. 2C) (Supplementary Table 1). After primary and secondary cloning (Fig. 2D), two clones that produced larger amounts of antibodies were selected and designated 1A8 and 4F5. The isotyping results showed that both mAbs were of the IgG1 subtype and contained k light chains.

Fig. 1

SDS-PAGE analysis of purified recombinant DEFA5 and clones 1A8 and 4F5 mAbs. A Lane M: Protein marker, Lane 1: DEFA5; Lane 2: DEFA5 reacted with His tag antibody. B Lane M: Protein marker, Lane 1: 1A8 mAb. C Lane M: Protein marker, Lane 1: 4F5 mAb

Fig. 2

Production of clones 1A8 and 4F5 anti-DEFA5 mAbs. A The recombinant protein was immunized in Balb/c mice. B The spleen cells were fused with myeloma cells. C The supernatants were screened by ELISA to select anti-DEFA5 mAbs-producing hybridomas. D Clones 1A8 and 4F5 anti-DEFA5 mAbs were established after limiting dilution

Identification of clones 1A8 and 4F5 effective for immunohistochemistry

We used in-house 1A8 and 4F5 antibodies and sc-53997 antibody, to stain serial sections of normal ileum tissue to produce comparable results. All three antibodies produced a similar staining pattern in the same tissue and exhibited strong immunoreactivity for DEFA5 in the Paneth cells in a fine granular pattern (Fig. 3A–C), which was abolished in the negative control in the normal ileum (Fig. 3D). These DEFA5 positive cells were also observed in the crypt lumen. Thus, we identified two mouse monoclonal antibodies that effectively and specifically recognized DEFA5 through immunohistochemistry.

Fig. 3

Detection of DEFA5 in sequential sections of normal ileum by Immunohistochemistry assay. Thin sections of normal ileum were stained with two newly developed high-affinity anti-DEFA5 mAbs clones 1A8 and 4F5 compared to the commercially available Santa Cruz Biotechnology Inc antibody (sc-53997), showing ileal paneth cells immunoreactive for 1A8 (A); 4F5 (B); sc-53997 (positive control) (C); and negative control (D)

Western blot analysis

By using the optimized WB protocol as outlined above, our new mouse monoclonal 1A8 and 4F5 antibodies effectively and strongly detected DEFA5 in DV, UC, CC, IC tissue protein extracts (Fig. 4A, B), whereas the commercially available anti-DEFA5 antibody sc-53997 could detect faint DEFA5 bands in CC and IC tissue lysates only (Fig. 4C). The R&D#972207 antibody more specifically recognized DEFA5 in CC and IC tissue lysates (Fig. 4D). This suggests that R&D972207 antibody effectively detected DEFA5 only at high expression levels and did not detect the protein in DV and UC samples, which had lower DEFA5 expression, even with extended exposure time.

Fig. 4

Detection of DEFA5 in tissues from DV control and IBD tissue samples. A, B DEFA5 expression was detected by Western blotting using our novel DEFA5 antibodies 1A8 and 4F5 and C, D commercially available DEFA5 antibodies from sc-53997 and R&D972207. Protein extracts were prepared from tissues obtained from patients with DV, UC, CC, and IC tissue samples

Immunoprecipitation analysis

We evaluated the functionality of the antibodies during immunoprecipitation of tissue lysates prepared from active CC samples to identify DEFA5 capture antibodies. Commercially available sc-53997 and R&D972204antibodies, along with in-house 1A8 and 4F5 antibodies, were used to immunoprecipitate DEFA5 from CC tissue lysates. The presence of DEFA5 in the immunoprecipitates was detected by immunoblotting using mouse monoclonal 1A8, 4F5, R&D972207, and sc-53997 antibodies for the immunoprecipitates obtained with sc-53997, R&D 972204, 1A8, and 4F5 antibodies (Fig. 5A). Only the sc-53997 antibody successfully immunoprecipitated endogenous DEFA5, establishing the most robust enrichment of the protein compared to the starting material. Subfigures 5A1–A4 show DEFA5 detection by immunoblotting using 1A8, 4F5, R&D972207, and sc-53997 antibodies, respectively. In each subfigure, lanes 1, 2, 3, 4, and 5 represent immunoprecipitants using mouse IgG (negative control), sc-53997, R&D972204, 1A8, and 4F5 antibodies, respectively. Successful detection of DEFA5 is indicated by distinct bands corresponding to detection by the 1A8 (Fig. 5A1, lane 2) 4F5 (Fig. 5A2, lane 2), R&D972207 (Fig. 5A3, lane 2), and sc-53997 (Fig. 5A4, lane 2) antibodies, confirming the specificity and efficiency of these antibodies for DEFA5. Notably, the sc-53997 antibody was the most effective for capturing DEFA5 in tissue lysates, as shown by its strong band intensities in the corresponding subfigure. Antibodies 1A8, 4F5, R&D972207, and sc-53997 were the most effective for detecting DEFA5 in the immunoprecipitants obtained with the sc-53997 antibody in immunoblots, as evidenced by distinct bands (Fig. 5A1–A4, lane 2), suggesting that the sc-53997 antibody was a preferred candidate for capturing DEFA5 in tissue lysates. Additionally, the sc-53997 antibody successfully performed both immunoprecipitation and detection of DEFA5, as indicated by the presence of a specific band (Fig. 5A4, lane 2). The absence of bands in lanes 3–5 suggests that DEFA5 was not effectively captured by 1A8, 4F5 or R&D972207 or that the epitopes recognized by the detection antibodies were inaccessible in these immunoprecipitants. No band was detected for the mouse IgG, which served as a negative control, confirming the specificity of the antibodies used for detecting DEFA5 (Fig. 5A1–A4, lane 1).

Fig. 5

Specificity of antibodies for the capture of endogenous DEFA5 in complex protein extracts and detection of DEFA5 in control and IBD patient tissue lysates by immunoprecipitation and Western blotting. A 1 mg tissue extracts prepared from Crohn’s colitis (CC) sample for immunoprecipitation using mouse IgG control antibody (lane 1), sc-53997 (lane 2), R&D972204 (lane 3), 1A8 (lane 4) and 4F5 (lane 5) antibodies. A1–A4 Subfigures corresponding to immunoblots with detection using the 1A8, 4F5, R&D972207, and sc-53997 antibodies, respectively. Lane 1: Negative control IP with mouse IgG for CC lysate. Lanes 2, 3, 4, and 5: IP with sc-53997, R&D972207, 1A8, and 4F5 antibodies for CC lysate, respectively. The immune complexes were probed with 1A8 (A1), 4F5 (A2), R&D972207 (A3), and sc-53997 antibodies (A4) for standard Western blot analysis. B The commercially available mouse monoclonal DEFA5 antibody sc-53997 was used to immunoprecipitate DEFA5 protein from DV, UC, CC, and IC tissue lysates and the 1A8, 4F5, R&D972207 and sc-53997 antibodies were used for standard Western blot analysis. B1–B4 Subfigures show the detection of DEFA5 using 1A8, 4F5, R&D972207, and sc-53997 antibodies, respectively. Lanes 1, 3, 5, and 7: Negative control IP with mouse IgG for DV, UC, CC, and IC lysates, respectively. Lanes 2, 4, 6, and 8: IP with sc-53997 antibody for DV, UC, CC, and IC lysates, respectively. Immunoprecipitates were used in Western blot and each blot was incubated with 1A8 (B1), 4F5 (B2), R&D972207 (B3), and sc-53997 (B4) antibodies. C Densitometric analysis of the precipitated and detected DEFA5 protein by ImageJ. Bars represent the mean intensity of the DEFA5 band from two patient samples relative to the corresponding IgG control

Next, we assessed the specificity of the 1A8, 4F5, and 972207 antibodies for DEFA5. Immunoprecipitation studies were conducted using the sc-53997 antibody as the capture antibody on tissue lysates from DV, UC, CC, and IC. Following immunoprecipitation, Western blot analysis was performed with detection antibodies 1A8, 4F5, R&D972207, and sc-53997 to assess the specificity of these antibodies in recognizing endogenous DEFA5 at physiologically relevant levels to enhance our understanding of IBD colitis states and evaluate DEFA5 as a potential biomarker. Figure 5B illustrates the detection of DEFA5 using 1A8, 4F5, R&D972207, and sc-53997 antibodies in DV, UC, CC, and IC tissue lysates following immunoprecipitation with sc-53997 antibody. DV (lane 1), UC (lane 3), CC (lane 5), IC (lane 7) lysates represent negative control immunoprecipitations using mouse IgG, confirming the specificity of the assay by the absence of bands (Fig. 5B1–B4). Subfigures 5B1–B4 correspond to immunoprecipitation with the sc-53997 antibody from the DV lysate (lane 2), UC lysate (lane 4), CC lysate (lane 6), and IC lysate (lane 8), with each subfigure displaying detection by a specific antibody where 1A8 detects in 5B1, 4F5 in 5B2, R&D972207 in 5B3, and sc-53997 in 5B4, and distinct bands confirming successful detection of DEFA5 in the respective lysates. The relative intensity of bands between different tissue lysates highlights variations in DEFA5 expression. The absence of any signal with the mouse IgG further validates that the observed bands are specific to DEFA5 and not due to non-specific binding (Fig. 5B1–B4, lanes 1, 3, 5, 7). Quantification of the protein bands using ImageJ revealed that all detection antibodies effectively recognized DEFA5. The strongest DEFA5 signal was observed in CC, while the weakest signals were seen in DV and UC. For IC tissues, DEFA5 levels were higher than in DV and UC but lower than in CC (Fig. 5C). In-house antibodies, particularly 1A8, exhibited the highest sensitivity for detecting DEFA5 in CC compared to the other antibodies, highlighting their potential for further research in distinguishing IBD colitis subtypes.

Immunoprecipitation analysis to verify specificity of antibodies

The specificity of antibodies 1A8, 4F5, and R&D972207 for DEFA5 was verified in cell lysates of transiently transfected HEK293T cells with cDNA encoding DEFA1, DEFA5, and DEFA6, as well as a negative control consisting of cells transfected with cDNA encoding the vector only using immunoprecipitation and Western blotting (Fig. 6A). The 1A8 anti-DEFA5 monoclonal antibody was the most efficient in detecting a strong 15 kDa band in the HEK293T-DEFA5 cell extracts, which was weak or negligible in the HEK-293 T-DEFA1 and HEK293T-DEFA6 cell lysates, thus verifying the specificity of 1A8 to DEFA5 and exhibited minimal non-confounding cross-reactivity with DEFA6 (Fig. 6A1). However, clone 4F5 anti-DEFA5 monoclonal antibody recognized 15 kDa bands in all three α-defensin recombinant proteins with relatively more staining in DEFA1 (Fig. 6A2), but the 4F5 antibody was not able to recognize recombinant DEFA5 as effectively as it did the endogenous form, likely due to the protein complexes still be intact in the analyte, leading to masking of the antibody epitope. However, further studies are required to verify the specificity of 4F5. The anti-DEFA5 R&D972207 antibody displayed obvious cross-reactivity with DEFA1 and DEFA6 in the immunoprecipitation analysis. (Fig. 6A3). No band was detected for mouse IgG, which was used as a negative control (Fig. 6A1–A3, lanes 1, 3, 5). ImageJ analysis revealed that the 1A8 antibody is more specific to DEFA5 but has lower sensitivity compared to 4F5 and R&D972207. The 4F5 and R&D972207 antibodies showed equally reduced specificity, with 4F5 being more sensitive than R&D972207 (Fig. 6B). Based on these data, it could be worth to explore 1A8 as a detection antibody in clinical models, while 4F5 does not appear to be suitable for this purpose. Additionally, based on specificity experiments with the sc-53997 antibody this might be also a feasible option for detection. It is also demonstrated that the observed lack of detection of DEFA5 by the R&D972207 antibody in DV and UC, along with strong detection in CC and IC, suggests a strong specificity for conditions resembling Crohn’s-like colitis. No band was detected in the control lysates of HEK293T cells transfected with the expression vectors when 1A8 (Fig. 6C1), 4F5 (Fig. 6C2), and R&D972207 antibodies were used (Fig. 6C3), suggesting that the antibodies are not binding to a non-specific target in the transfected HEK293K cells, and the observed binding in the experimental conditions is specific. The negative control immunoprecipitates with mouse IgG in each subfigure (Fig. 6C1–C3, lanes 1, 3, 5) confirm the specificity of the assay by showing the absence of bands. The specificity of the R&D972207 antibody for recombinant DEFA5 was also verified by Western blotting of HEK293T cell lysates transiently expressing DEFA1, DEFA5, and DEFA6, or control cells expressing the vector only. The R&D972207 antibody detected a single band at 15 kDa in cell lysates containing DEFA5, but not DEFA1 or DEFA6, suggesting that R&D 927707 is specific for DEFA5 recombinant protein and does not cross-react with DEFA1 or DEFA6 (Fig. 6D). Although this antibody was able to detect DEFA5 when highly expressed in HEK 293 T cells and support as a specific tool for detecting DEFA5 in mammalian recombinant systems, it was unable to detect lower levels of endogenous DEFA5 protein when expressed in DV and UC tissues, indicating a lack of specificity for DEFA5, as shown in Fig. 4D. In contrast, no band was observed in the cells transfected with the vector used as a negative control.

Fig. 6

Specificity of the 1A8, 4F5, and R&D972207 antibodies for DEFA5. A Cell lysates from HEK293T cells expressing recombinant DEFA1, DEFA5, and DEFA6 α defensin family members were analyzed by immunoprecipitation and Western blotting. A1–A3 Subfigures represent the detection of recombinant DEFA1, DEFA5, and DEFA6 in HEK293T cell lysates using 1A8, 4F5, and R&D972207 antibodies, respectively. Lanes 1, 3, and 5: Negative control IP with mouse IgG for recombinant DEFA1, DEFA5, and DEFA6 HEK293T cell lysates, respectively. Lanes 2, 4, and 6: IP with sc-53997 antibody for DEFA1, DEFA5, and DEFA6 HEK293T cell lysates, respectively. Immunoprecipitants were used in Western blot and each blot was incubated with 1A8 antibody (A1), 4F5 antibody (A2), and R&D972207 antibody (A3). B Densitometric analysis of the precipitated DEFA5 protein. Bars represent the mean intensity of the DEFA5 band from two independent experiments relative to the corresponding IgG control. C (C1–C3) Subfigures represent the detection of control lysates from HEK293T cells transfected with empty expression vectors for DEFA1, DEFA5, and DEFA6. Lanes 1, 3, and 5: Negative control IP with mouse IgG for lysates from HEK293T cells transfected with empty DEFA1, DEFA5, and DEFA6 expression vectors, respectively. Lanes 2, 4, and 6: IP with sc-53997 antibody for lysates from HEK293T cells transfected with empty DEFA1, DEFA5, and DEFA6 expression vectors, respectively. Immunoprecipitates were used in Western blot and each blot was incubated with 1A8 (C1), 4F5 (C2), and R&D972207 (C3) antibodies. D Cell lysates from HEK293T cells expressing recombinant DEFA1, DEFA5, and DEFA6 α defensin family members were analyzed by standard Western blotting. Detection of GAPDH was used as the loading control

Immunohistochemistry to verify the specificity of antibodies

We further assessed the antibody staining of human tonsil tumor tissue sections and compared it with staining of human normal terminal ileal sections. The human α defensin 5 specificity of the 1A8 antibody (Fig. 7A, left image), 4F5 (Fig. 7B, left image), R&D972207 (Fig. 7C, left image), and sc-53997 (Fig. 7D, left image) was verified in sections of ileal tissues, where the antibodies strongly stained Paneth cells expressing DEFA5. Notably, the specificity of all four antibodies for Paneth-cell-specific DEFA5 was demonstrated by the absence of HNP-1, HNP-2, HNP-3, and HNP-4 immunoreactive cells in consecutive sections of tonsil tumor tissue (Fig. 7A–D, right images), confirming their specificity to DEFA5 in immunohistochemistry analysis. Hematoxylin and Eosin (H&E) staining of the tonsil tumor tissue revealed cells morphologically identical to neutrophils (Fig. 7E, right panel).

Fig. 7

Immunohistochemical staining of formalin-fixed paraffin sections of human normal terminal ileum and human tonsil cancer tissue with 1A8 (panel A), 4F5 (panel B), R&D972207 (panel C) and sc-53997 (panel D) antibodies. The ileal section stained without primary DEFA5 antibody and stained only with HRP conjugated secondary antibody as control (panel E, left image), and the tonsil tumor tissue section stained with H&E (panel E, right image) polymorphonuclear neutrophils can be seen

Clones 1A8 and 4F5 in ELISA assays

Next, we analyzed the binding capabilities of 1A8, 4F5, and R&D972207 to endogenous DEFA5 protein using Sandwich ELISA. When binding was detected using biotinylated antibodies, clones 1A8 and 4F5 displayed a stronger binding profile than the R&D972207 antibody did. A standard curve constructed with the human recombinant α defensin 5 (DEFA5) standards (0.62 -40 ng/ml) showed that the OD650 values were directly correlated to the diluted standard concentrations. A negative control sample (EV), a vector lacking α-defensin 5, was used to check for non-specific binding and false-positive results. Blank samples were used to subtract the background from the readings. The amount of DEFA5 protein in the DV, UC, CC, and IC samples was determined using a standard curve drawn using CurveExpert 1.4 Software. The DEFA5 levels in the DV, UC, CC, IC samples were 0.86 ng/ml (DV), 0.95 ng/ml (UC), 35.75 ng/ml (CC), 19.22 ng/ml (IC) detected with 1A8 antibody; 0.7 ng/ml (DV), 0.83 ng/ml (UC), 28.51 ng/ml (CC), 22.04 ng/ml (IC) with 4F5 antibody; and 0.71 ng/ml (DV), 0.89 ng/ml (UC), 20.68 ng/ml (CC), 19.36 ng/ml (IC) with R&D 972207 antibody (Fig. 8). This suggests that CC and IC samples led to a 20–34.9-fold increase compared to DV and UC samples, demonstrating a significant difference between UC and CC or IC (1A8; P < 0.01; 4F5; P < 0.01; R&D; P < 0.01) (Fig. 8). This analysis suggests that DEFA5 levels < 1 ng/ml may represent DV or UC, whereas DEFA5 levels > 20 ng/ml may denote CC. Both in-house antibodies, 1A8 and 4F5, showed similar trends in detecting DEFA5 levels across the DV, UC, CC, and IC disease states. In contrast, R&D972207 detected lower DEFA5 levels in CC and IC, which could suggest slightly less sensitivity or a narrower dynamic range. This specificity and binding strength make these in-house antibodies and ELISA platforms effective tools for quantifying DEFA5 levels, further supporting their utility in diagnostic applications for conditions like IBD colitis.

Fig. 8

Determination of DEFA5 levels in control and IBD tissue samples by Sandwich ELISA. The commercially available sc-53997 mouse monoclonal antibody was used as the capture antibody and our in-house biotinylated mouse monoclonal 1A8, 4F5, and the commercially available R&D972207 antibodies were used as the detection antibodies. Antigens for the sandwich ELISA were patient tissue lysates from Diverticulitis (DV), Ulcerative colitis (UC), Crohn’s colitis (CC) and Indeterminate colitis (IC) patient samples and vector only lacking DEFA5 (EV). Bars represent mean ± S.D (n = 3) values in ng/ml for the indicated antibodies. The statistical difference between UC and CC (1A8; P = 0.0017, **P < 0.01) or IC (P = 0.0029, **P < 0.01), UC and CC (4F5; P = 0.0023, **P < 0.01) or IC (P = 0.0034, **P < 0.01) and UC and CC (R&D972207; p = 0.0063, **P < 0.01) or IC (P = 0.0078, **P < 0.01) were significantly correlated