My project week in the group of Prof. Dr. C. Griesinger at the MPI for Biophysical Chemistry was supervised by D. Lennartz, an associate member of the graduate school. The main topic was the investigation of residual dipolar couplings of proteins in orienting media, which can give additional information for structural refinements as well as molecular dynamics in solutions.The proteins investigated during the project week were Ubiqutin solved in the so called Helfrich-Phase as orienting medium and ProteinG3 solved in Purple Membrane (Bacterio Rhodopsin).
Apart from sample preparation, the "practical" part of the project week involved the handling of the NMR spectrometers. The magnetic field produced by the main magnetic coil was "shimmed," which means that it was homogenised by some extra coils, the so called shim coils. The power of the different shim coils was varied manually and optimised to a reference signal to which the spectrometer was locked. After the shimming the impedance of the probe head had to be adjusted to the impedance of the cable system to prevent a loss of signal within the cables. This is called wobbling.
In the case of Ubiqutin a 2D IPAP HSQC spectrum containing the residual dipolar HNC' and NC' couplings was recorded. The couplings were determined manually using the computer program FELIX2000 and compared to dipolar couplings obtained from an X-ray structure. The correlation between these two data sets was very high. One could not tell whether the discrepancies were due to dynamics within the solution or to systematic errors. For the ProteinG3 the dipolar NH couplings were determined semi-automatically from an IPAP HSQC spectrum by nmrPipe.
In addition I was introduced in peak assignment methods by M. Mukrasch, also an associate member of the graduate school. The protein he is investigating as part of his PhD thesis is an unfolded mutant of the τ-protein containing 130 amino acids. First a 2D [¹H, ¹⁵N] HSQC spectrum was recorded. The information about the connectivity of the protein was obtained by numerous 3D spectra (HNN, HN(C)N, HCNO, HN(CA)CO, HNCACB and HA(CA)NNH). The sequence of spin systems was matched to the amino acids sequence. In this way it was possible to assign each signal from the 2D [¹H, ¹⁵N] HSQC as well as the signals of the α-H, α-C and β-C atoms to a specific amino acid. This work was done by M. Mukrasch within the last months. My project week took place when he was about to finish the assignment.
I was also introduced to some maintenance work e.g. helium filling and SGU (signal generation unit) board exchange and to multi-user management of the spectrometers.
Finally, I would like to thank Dirk Lennartz for the time he spent doing the experiments with me and for his patience explaining the theory to me.

April 29 2004