The interaction between various charmed mesons and charmed baryons are studied within the framework of the coupled channel unitary approach with the local hidden gauge formalism. Several meson-baryon dynamically generated narrow N * and Λ * resonances with hidden charm are predicted with mass above 4 GeV and width smaller than 100 MeV. The predicted new resonances definitely cannot be accommodated by quark models with three constituent quarks and can be looked for at the forthcoming PANDA/FAIR experiments.PACS numbers: 14.20. Gk, 13.30.Eg, 13.75.Jz Up to now, all established baryons can be ascribed into 3-quark (qqq) configurations [1], although some of them were suggested to be meson-baryon dynamically generated states [2][3][4][5][6][7][8] or states with large (qqqqq) components [9][10][11]. A difficulty to pin down the nature of these baryon resonances is that the predicted states from various models are around the same energy region and there are always some adjustable ingredients in each model to fit the experimental data. In this letter, we report a study of the interactions between various charmed mesons and charmed baryons within the framework of the coupled channel unitary approach with the local hidden gauge formalism. Several meson-baryon dynamically generated narrow N * and Λ * resonances with hidden charm are predicted with mass above 4 GeV and width smaller than 100 MeV. The predicted new resonances can be looked for at the forthcoming PANDA/FAIR experiments [13]. If confirmed, they definitely cannot be accommodated by quark models with three constituent quarks.We follow the recent approach of Ref.[12] and extend it from three flavors to four. We consider the P B → P B and V B → V B interaction by exchanging a vector meson, as shown by the Feynman diagrams in Fig. 1.The effective Lagrangians for the interactions involved are [12]:where P and V stand for pseudoscalar and vector mesons of the 16-plet of SU(4), respectively. Under the low energy approximation, the three momentum versus the mass of the meson can be neglected. We can just take the γ 0 component of Eq. (1). The threemomentum and energy of the exchanged vector are both much smaller than its mass, so its propagator is approximately g µν /M 2 V . Then with g = M V /2f the transition potential corresponding to the diagrams of Fig. 1 are given bywhere the a, b stand for different channels of P 1 (V 1 )B 1 and P 2 (V 2 )B 2 , respectively. The E is the energy of corresponding particle. The ǫ is the polarization vector of the initial or final vector. And the ǫ 0 1,2 component is neglected consistently with taking p/M V ∼ 0, with p the momentum of the vector meson. The C ab coefficients can be obtained by the SU(4) Clebsch Gordan Coefficients which we take from Ref. [14]. We list the values of the C ab coefficients for P B → P B with isospin and strangeness (I, S) = (1/2, 0) and (0, -1) explicitly in Table I and Table II, respectively. With the transition potential, the coupled-channel scattering matrix can be obtained by solving the coupledc...