[en] The estimation of the cross sections of certain dark matter interactions with nuclei requires a correct evaluation of the couplings between the scalar or pseudoscalar Higgs boson and the nucleons. Progress has been made in two aspects relevant to this study in the past few years. First, recent lattice calculations show that the strange-quark sigma term σ_s and the strange-quark content in the nucleon are much smaller than what are expected previously. However, in view of the conflict between lattice and experimental results for the pion-nucleon sigma term σ_πN, the quark sigma terms σ_u,d,s are still not well determined. Second, the pseudoscalar Higgs coupling with the nucleon is customarily expressed in terms of the axial-vector couplings g_A^a(a=0,3,8) or the quark spin components Δu,Δd and Δs. Lattice calculations, semi-inclusive deep inelastic scattering data, and the small ΔG/G obtained by RHIC, COMPASS and HERMES all indicate a smaller Δs=O(−0.02∼−0.03), which in turn implies sizable SU(3) breaking effects in the determination of g_A⁸ and g_A⁰. We re-evaluate the relevant nucleon matrix elements and compute the scalar and pseudoscalar couplings of the proton and neutron