1.70 A ˚ X-Ray Structure of Human apo Kallikrein 1: Structural Changes Upon Peptide Inhibitor/Substrate Binding Gurunathan Laxmikanthan, 1,2 Sachiko I. Blaber, 1 Matthew J. Bernett, 1,2 Isobel A. Scarisbrick, 3 Maria Aparecida Juliano, 4 and Michael Blaber 1,2 * 1 Institute of Molecular Biophysics Florida State University, Tallahassee, Florida 2 Department of Chemistry and Biochemistry, Florida State University, Tallahassee, Florida 3 Program for Molecular Neuroscience and Departments of Neurology, and Physical Medicine and Rehabilitation, Mayo Medical and Graduate Schools, Rochester, Minnesota 4 Department of Biophysics, Universidade Federal de Sao Paulo, Escola Paulista de Medicina, Sao Paulo, Brazil ABSTRACT Human kallikreins are serine pro- teases that comprise a recently identified large and closely related 15-member family. The kallikreins include both regulatory- and degradative-type pro- teases, impacting a variety of physiological pro- cesses including regulation of blood pressure, neuro- nal health, and the inflammatory response. While the function of the majority of the kallikreins re- mains to be elucidated, two members are useful biomarkers for prostate cancer and several others are potentially useful biomarkers for breast cancer, Alzheimer’s, and Parkinson’s disease. Human tissue kallikrein (human K1) is the best functionally char- acterized member of this family, and is known to play an important role in blood pressure regulation. As part of this function, human K1 exhibits unique dual-substrate specificity in hydrolyzing low molecu- lar weight kininogen between both Arg-Ser and Met-Lys sequences. We report the X-ray crystal struc- ture of mature, active recombinant human apo K1 at 1.70 Å resolution. The active site exhibits structural features intermediate between that of apo and pro forms of known kallikrein structures. The S2 to S2 pockets demonstrate a variety of conformational changes in comparison to the porcine homolog of K1 in complex with peptide inhibitors, including the displacement of an extensive solvent network. These results indicate that the binding of a peptide sub- strate contributes to a structural rearrangement of the active-site Ser 195 resulting in a catalytically competent juxtaposition with the active-site His 57. The solvent networks within the S1 and S1 pockets suggest how the Arg-Ser and Met-Lys dual substrate specificity of human K1 is accommodated. Proteins 2005;58:802– 814. © 2005 Wiley-Liss, Inc. Key words: kallikrein; serine protease; substrate specificity; solvent structure; induced fit INTRODUCTION The kallikreins are a family of serine proteases that are distributed in a wide variety of tissues and biological fluids. The term kallikrein is derived from the Greek word “kallikreas” which means pancreas, and historically this organ was the main source of these proteases. 1 It was believed for many years that the human kallikrein family had only three members; however, relatively recent stud- ies have revealed that there are a total of 15 members that are co-localized at human chromosome loci 19q13.3- q13.4. 2–4 Sequence analysis indicates the members of the human kallikrein family share 40 – 80% amino acid iden- tity. 4 The catalytic triad residues His57, Asp102, and Ser195 are conserved in each case, however, residue positions that define the different substrate-binding pock- ets exhibit heterogeneity. For example, position 189, lo- cated at the base of the S1 binding pocket, is observed to be either Asp, Gly, Glu, Ser, or Asn. Thus, the substrate P1 specificity of the kallikreins may vary considerably. Histori- cally, rodents have been a focus for the study of the kallikreins. However, the rodent kallikrein gene locus appears to be uniquely different, and much more exten- sive, than the human. In addition to homologs of the 15 human kallikrein genes (i.e., KLK’s), rodents exhibit ap- proximately a dozen additional functional kallikrein genes known as “glandular” kallikreins (i.e., GK’s). 5 The human KLK family has gained attention in recent years due to the fact that most of the members appear to be differentially expressed in normal versus cancerous tis- sues, and this may prove to be a useful diagnostic in Abbreviations: Ni-NTA, nickel-nitriloacetic acid; EK, enterokinase; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophore- sis; EDDnp, N-[2,4-dinitrophenyl]-ethylenediamine; TFA, trifluoroace- tic acid; CAN, acetonitrile; MALDI-TOF, matrix assisted laser desorp- tion ionization time-of-flight; EDTA, ethylenediamine tetraacetic acid; RMS, root mean square; BPTI, bovine pancreatic trypsin inhibitor. The atomic coordinates and structure factors (code 1SPJ) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org). *Correspondence to: Michael Blaber, 406 Kasha Laboratory, Insti- tute of Molecular Biophysics, Florida State University, Tallahassee, FL 32306-3015. E-mail: blaber@sb.fsu.edu Received 3 August 2004; Accepted 7 September 2004 Published online 13 January 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/prot.20368 PROTEINS: Structure, Function, and Bioinformatics 58:802– 814 (2005) © 2005 WILEY-LISS, INC.