Secretion and N-linked glycosylation are required for prostatic acid phosphatase catalytic and antinociceptive activity

Abstract

Secretory human prostatic acid phosphatase (hPAP) is glycosylated at three asparagine residues (N62, N188, N301) and has potent antinociceptive effects when administered to mice. Currently, it is unknown if these N-linked residues are required for hPAP protein stability and activity in vitro or in animal models of chronic pain. Here, we expressed wild-type hPAP and a series of Asn to Gln point mutations in the yeast Pichia pastoris X33 then analyzed protein levels and enzyme activity in cell lysates and in conditioned media. Pichia secreted wild-type recombinant (r)-hPAP into the media (6-7 mg protein/L). This protein was as active as native hPAP in biochemical assays and in mouse models of inflammatory pain and neuropathic pain. In contrast, the N62Q and N188Q single mutants and the N62Q, N188Q double mutant were expressed at lower levels and were less active than wild-type r-hPAP. The purified N62Q, N188Q double mutant protein was also 1.9 fold less active in vivo. The N301Q mutant was not expressed, suggesting a critical role for this residue in protein stability. To explicitly test the importance of secretion, a construct lacking the signal peptide of hPAP was expressed in Pichia and assayed. This "cellular" construct was not expressed at levels detectable by western blotting. Taken together, these data indicate that secretion and post-translational carbohydrate modifications are required for PAP protein stability and catalytic activity. Moreover, our findings indicate that recombinant hPAP can be produced in Pichia--a yeast strain that is used to generate biologics for therapeutic purposes.

Figure 1. Pichia-derived r-hPAP is secreted and catalytically active.

(A) Western blot of crude cell lysates (Lys) and supernatants (Sup) from P. pastoris X33 untransformed controls and from P. pastoris X33 expressing r-hPAP. Blot probed with anti-hPAP antiserum. (B) DiFMUP fluorometric enzyme assay with concentrated supernatants (secreted fractions) in comparison to native hPAP from human semen. 0.625 µg total protein used per reaction. Data are plotted as an average of duplicate trials ± standard deviation (SD).

Figure 2. r-hPAP is glycosylated when expressed in Pichia.

(A) Western blot of concentrated r-hPAP secreted fraction after incubation at 37°C for 24 h with or without 1000 U PNGase. Blot probed with anti-hPAP antiserum. (B) DiFMUP fluorometric enzyme assay using equal amounts of untreated and PNGase-treated r-hPAP. Data are plotted as an average of duplicate trials ± SD.

Figure 3. Expression and activity of N-linked glycosylation mutants.

Western blots of (A) crude cell lysates and (B) crude secreted fractions from the indicated P. pastoris X33 integrants. Blots probed with anti-hPAP antiserum. Equivalent amounts of total protein were loaded in each lane. (C, D) DiFMUP fluorometric enzyme assays of (C) crude cell lysates and (D) crude secreted fractions. Data are plotted as an average of duplicate trials ± SD.

Figure 4. Expression and activity of r-hPAP lacking signal peptide(-SP).

(A) Western blot of crude cell lysates from untransformed X33 cells and transformants expressing hPAP with (r-hPAP) or without (-SP) a signal peptide. Blot probed with anti-hPAP antiserum. Equivalent amounts of total protein were loaded in each lane. (B) DiFMUP fluorometric enzyme assay with the indicated crude cell lysates. Data are plotted as an average of duplicate trials ± SD.

Figure 5. r-hPAP has antinociceptive properties in vivo.

Antinociceptive properties of native hPAP, r-hPAP and r-hPAP (N62Q, N188Q) injected intrathecally into wild-type (n = 10) and A₁R ^−/− mice (n = 10). Equivalent unit amounts (250 mU/mouse) of native hPAP and r-hPAP were injected. Equivalent protein amounts (0.21 mg/mL) of r-hPAP and r-hPAP (N62Q, N188Q) were injected. Paired t-tests were used to compare responses at each time point to baseline (BL). *p<0.05, **p<0.005, ***p<0.0005. Data are plotted as means ± standard error of the mean (SEM).

Figure 6. r-hPAP does not affect balance or motor function in mice.

Rotarod tests with wild-type (n = 10) and A₁R^−/− mice (n = 10) 24 h before (3 iterations, separated by 40 s) and 24 h after (2 test iterations, separated by 40 s) intrathecal injection of r-hPAP (250 mU/mouse). No significant differences between genotypes or treatment. Data are plotted as means ± SEM. The same mice shown were also tested for thermal sensitivity, to confirm that hPAP injections were successful (as evidenced by a significant thermal antinociceptive effect in wild-type mice injected with hPAP, data not shown).

Figure 7. Antinociceptive effects of r-hPAP in chronic inflammatory pain model.

(A, B) CFA was injected into one hindpaw (CFA-arrow) of wild-type (n = 10) and A₁R^−/− mice (n = 10). r-hPAP (250 mU) was intrathecally injected 1 day later (r-hPAP-arrow). Inflamed and non-inflamed (control) hindpaws were tested for (A) thermal and (B) mechanical sensitivity. Data are plotted as means ± SEM. Paired t-tests were used to compare responses at each time point between genotypes, same paw comparisons. *p<0.05, **p<0.005, ***p<0.0005.

Figure 8. Antinociceptive effects of r-hPAP in neuropathic pain model.

(A, B) The sural and common peroneal branches of the sciatic nerve were ligated and then transected (injure-arrow) in wild-type (n = 10) and A₁R^−/− mice (n = 10). Six days later, r-hPAP (250 mU) was injected intrathecally. Injured and non-injured (control) hindpaws were tested for (A) thermal and (B) mechanical sensitivity. Data are plotted as means ± SEM. Paired t-tests were used to compare responses at each time point between genotypes, same paw comparisons. *p<0.05, **p<0.005, ***p<0.0005.

Figure 9. Location of N-linked asparagines residues relative to the active site of hPAP.

The x-ray crystallographic structure depicts the essential active site residue H12 and the three N-linked residues (highlighted in yellow) in one subunit of native hPAP. Distances were calculated in PyMOL between the alpha carbon of each amino acid. Structure coordinates from PDB #1ND6 .