Peptidomics approach to elucidate the proteolytic regulation of bioactive peptides

Abstract

Peptide hormones and neuropeptides have important roles in physiology and therefore the regulation of these bioactive peptides is of great interest. In some cases proteolysis controls the concentrations and signaling of bioactive peptides, and the peptidases that mediate this biochemistry have proven to be extremely successful drug targets. Due to the lack of any general method to identify these peptidases, however, the role of proteolysis in the regulation of most neuropeptides and peptide hormones is unknown. This limitation prompted us to develop an advanced peptidomics-based strategy to identify the peptidases responsible for the proteolysis of significant bioactive peptides. The application of this approach to calcitonin gene-related peptide (CGRP), a neuropeptide associated with blood pressure and migraine, revealed the endogenous CGRP cleavage sites. This information was then used to biochemically purify the peptidase capable of proteolysis of CGRP at those cleavage sites, which led to the identification of insulin-degrading enzyme (IDE) as a candidate CGRP-degrading enzyme. CGRP had not been identified as an IDE substrate before and we tested the physiological relevance of this interaction by quantitative measurements of CGRP using IDE null (IDE(-/-)) mice. In the absence of IDE, full-length CGRP levels are elevated in vivo, confirming IDE as an endogenous CGRP-degrading enzyme. By linking CGRP and IDE, this strategy uncovers a previously unknown pathway for CGRP regulation and characterizes an additional role for IDE. More generally, this work suggests that this may be an effective general strategy for characterizing these pathways and peptidases moving forward.

Fig. 1.

Identifying endogenous CGRP cleavage sites. (A) Combination of peptidomics, biochemistry, and genetics provides a unique strategy to identify peptidases that process important bioactive peptides, such as CGRP. (B) Peptidomics analysis of mouse spinal cords reveals the largest number of CGRP fragments ever reported. These fragments were used to determine the endogenous CGRP cleavage sites. (Note: there is a disulfide between Cys₂ and Cys₇ for the full-length and N-terminal fragments). (C) Detection of the N-terminal CGRP fragments, CGRP_1–17 and CGRP_1–26, and the C-terminal fragments, CGRP_18–37 and CGRP_27–37, indicates that full-length CGRP_1–37 is cleaved at the Ser₁₇-Arg₁₈ and Asn₂₆-Phe₂₇ sites.

Fig. 2.

Biochemical assays with CGRP_1–37 in spinal cord lysates and plasma. CGRP_1–37 was incubated with spinal cord lysate and analyzed by (A) MALDI-TOF MS and (B) LC–MS/MS. MALDI-TOF MS analysis identified CGRP_1–17 and CGRP_18–37 as the major breakdown products of CGRP_1–37, whereas the more sensitive LC–MS also identified CGRP_1–26 and CGRP_27–37. Incubation of CGRP_1–37 with plasma and subsequent analysis by (C) MALDI-TOF MS and (D) LC–MS/MS identified the same fragments in plasma. (Error bars show SEM.)

Fig. 3.

Characterization of IDE as a candidate CGRP-degrading enzyme. CGRP-degrading activity was measured by quantitation of (A) CGRP_1–17 or (B) CGRP_18–37 produced from CGRP_1–37 by LC–MS (i.e., area under the curve for these peaks in the LC–MS chromatogram). (A) Biochemical assays using spinal cord lysates treated with class selective protease inhibitors (aspartyl, cysteine, metallo, or serine) showed that the peptidase responsible for CGRP proteolysis is a metallopeptidase. Control reactions included a vehicle-treated sample that retained full activity and a heat-treated sample that lacked any activity. Metallopeptidase inhibitors 1,10-phenanthroline and EDTA were the only two compounds tested that inhibited the CGRP degradation. (B) Identification of candidate peptidases was carried out by anion-exchange chromatography of the spinal cord proteome followed by proteomics of the fractions with the highest levels of CGRP-degrading activity. Proteomics identified three metallopeptidases in these fractions; THOP, NLN, and IDE. (C) IDE was identified as the only peptidase with CGRP-degrading activity by incubating CGRP_1–37 with recombinant murine THOP, NLN, and IDE followed by quantitation of CGRP_1–37, CGRP_1–17, CGRP_18–37, CGRP_1–26, and CGRP_27–37. THOP and NLN barely showed any activity against CGRP (two orders of magnitude less than IDE). (Error bars show SEM.)

Fig. 4.

IDE cleaves CGRP_1–37 in complex proteomes. (A) TT cell line naturally produces and secretes CGRP. By Western blot (Inset, Top) TT cells have very little or no IDE. To study the impact of IDE on CGRP levels in TT cells, we transfected these cells with an empty vector (mock) or a vector containing the IDE gene (IDE transfected cells). CGRP_1–37 levels were significantly lower in media from IDE transfected cells, whereas levels of the CGRP fragment, CGRP_18–37, were elevated in media expressing IDE, indicating increased CGRP_1–37 proteolysis. Incubation of CGRP_1–37 with (B) spinal cord lysate and (C) plasma revealed higher CGRP_1–17 in IDE^+/+ samples, demonstrating that IDE is responsible for the majority of the CGRP-degrading activity in these lysates and plasma. (D) Incubation of CGRP_1–37 with plasma revealed that CGRP_1–37 is more stable in IDE^−/− plasma, demonstrating that IDE processing can regulate CGRP_1–37 levels in vitro. Relative quantification was accomplished by using the ion intensity of the different peptides in the LC–MS chromatogram. These ion intensities were then normalized to the largest peak to enable comparison of different peptides on the basis of relative changes in their abundance. (*P value < 0.05, **P value < 0.01, ***P value < 0.001; P values were derived from the two-tailed Student t test, n = 3. Error bars show SEM.)

Fig. 5.

IDE regulates CGRP_1–37 in vivo. (A) IDE^+/+ and IDE^−/− mice were treated with vehicle or capsaicin, and absolute endogenous CGRP levels were measured by IDMS²⁹ using a synthetic d18-CGRP_1–37 internal standard. CGRP_1–37 levels were significantly higher in the IDE^−/− mice than in the IDE^+/+ mice under both treatment conditions. (B) A unique model for CGRP regulation includes the IDE-mediated proteolysis pathway discovered here. (**P value < 0.01; P values were derived from the two-tailed Student t test, n = 3. Error bars show SEM.)