Rigidifying acyl carrier protein domain in iterative type I PKS CalE8 does not affect its function

Abstract

Acyl carrier protein (ACP) domains shuttle acyl intermediates among the catalytic domains of multidomain type I fatty acid synthase and polyketide synthase (PKS) systems. It is believed that the unique function of ACPs is associated with their dynamic property, but it remains to be fully elucidated what type of protein dynamics is critical for the shuttling domain. Using NMR techniques, we found that the ACP domain of iterative type I PKS CalE8 from Micromonospora echinospora is highly dynamic on the millisecond-second timescale. Introduction of an interhelical disulfide linkage in the ACP domain suppresses the dynamics on the millisecond-second timescale and reduces the mobility on the picosecond-nanosecond timescale. We demonstrate that the full-length PKS is fully functional upon rigidification of the ACP domain, suggesting that although the flexibility of the disordered terminal linkers may be important for the function of the ACP domain, the internal dynamics of the helical regions is not critical for that function.

Figure 1

(a) Domain organization of CalE8 and ribbon structure of meACP, with S971, A938C, and E1009C highlighted. (b) Thermal denaturation profiles of meACP and its mutant monitored at 222 nm by CD. (c) CD profiles of meACP and its mutant with and without 1.7 mM TCEP at 25°C.

Figure 2

Sequential- and medium-range NOE patterns of the meACP mutant. The NOE intensities are represented by the thickness of the solid bars. The secondary structure elements of the WT protein are indicated at the top of the sequence.

Figure 3

Comparison of ¹⁵N-{¹H} NOEs for WT meACP (•) and its mutant (○). The locations of two cysteine residues in the mutant are indicated by arrows on the secondary structure diagram.

Figure 4

Representative relaxation dispersion profiles of meACP (a–e) and comparison of experimental (δ_mi) and random coil (δ_random) ¹⁵N chemical shifts of the invisible minor form (f). The experimental data recorded by 500 and 800 MHz NMR are indicated by ○ and ^∗, respectively, and solid lines are fitting curves in panels a–e.

Figure 5

Comparison of P factors for WT (bar) and mutant (^∗) meACP. P factors were calculated from amide hydrogen exchange rates (k_ex) and random coil exchange rates. For the WT meACP, k_ex values were previously measured at pH 6.9 and 7.5 using an exchange spectroscopy (EXSY) scheme. For the mutant, k_ex values were obtained using either the EXSY scheme or the H-D exchange method. When P < 100 or log(P) < 2, k_ex could be measured using the EXSY scheme. When P > 1000 or log(P) > 3, k_ex could be measured using the H-D exchange scheme. When 100 < P < 1000, k_ex could not be measured by these two methods. The locations of two cysteine residues in the mutant are indicated by arrows on the secondary structure diagram.

Figure 6

Enzymatic activity of CalE8 and the double mutant. (A) Absorption spectra of the WT and mutant CalE8 expressed from E. coli. (B and C) Comparison of the enzymatic activity of CalE8 and the double mutant under (B) oxidizing and (C) reducing conditions.