Seattle Structural Genomics Center for Infectious Disease


FUN_09: Evaluation of the Inhibitory Activity of Small Molecule Compounds Against Rickettsia prowazekii Methionine Aminopeptidase (MetAp) using in vitro & in vivo Assays.


SSGCID Functional Project Proposal:  PI, Tim Hagen, Ph.D., Northern Illinois University and Jon Audia, University of South Alabama. Co-investigator: James Horn, Northern Illinois University.


Newly synthesized proteins contain a N-terminal methionine that must be removed by a methionine aminopeptidase (MetAP). (1) This ubiquitous enzyme is essential in bacteria and gene deletion results in lethality.  Inhibition of MetAP is therefore a potential target for the development of novel antibacterial agents.  MetAPs are broadly divided into type 1 and type 2 classes. Both prokaryotes and eukaryotes contain type 1 MetAPs, while eukaryotes have an additional type 2 MetAP enzyme. 

SSGCID has solved the structure of Methionine aminopeptidase from Rickettsia prowazekii (RiprA.00039.a) in both an apo state and methionine-bound state (PDB ID: 3MR1 & 3MX6).  Dr. Hagen has developed a collection of 500 compounds based upon known MetAP inhibitor scaffolds.  Figure 1 presents representative structures of compounds within NIU’s MetAP library. Our studies characterizing inhibitor/MetAP interactions have progressed rapidly with several compounds demonstrating anti-rickettsial in vitro activity.  

For this project we propose to screen the NIU library of approximately 500 MetAP-focused compounds against R. prowazekii (RiprA.00039.a) using both in vitro and in vivo assays. In vitro enzymatic assays will be conducted at NIU by Drs. Tim Hagen and James Horn. In vivo assays will be conducted in the facilities of Dr. Jon Audia.  This effort will advance our chemical hypothesis to improve inhibitory activity and will lead to the synthesis of new molecules in the Hagen laboratory.


R. prowazekii is an obligatory intracellular parasitic bacterium that grows only within the cytoplasm of a eukaryotic host cell. R. prowazekii is transmitted in nature by the human body louse which defecates at the feeding site, depositing feces contaminated with rickettsiae-infected cells.  Scratching of the bite site results in sub-dermal inoculation followed by subsequent dissemination of the rickettsiae and infection of the vascular endothelium.  There the rickettsiae grow, filling the host cell cytoplasm causing cell lysis and systemic dissemination.  Widespread vascular ‘leak’ is a major contributor to the pathologies associated with R. prowazekii epidemic typhus fever.  Given that R. prowazekii is designated as a Select Agent pathogen by the Centers for Disease Control and Prevention (CDC) because of its potential use in bioterrorism/warfare, developing novel anti-rickettsial therapeutics is a priority.

Project Design: 

The proposed project includes both in vitro and in vivo assays to evaluate potential MetAP inhibitors. Inhibitor leads will be further developed using structure-guided approaches using X-ray structures.

Aim 1: Determine the in vitro inhibitory activity of the NIU metAP compound library against RpMetAP (NIU:Hagen/Horn)

We have established a plate-based assay to evaluate inhibitors targeting MetAp from B. pseudomallei (2) and P. falciparum.  This is a fluorescent-based assay that follows the enzymatic cleavage of the Met-AMC substrate. Using this assay we have confirmed that MetAP from R. prowazekii MetAP (SSGCID batch RiprA.00039.a) is functional using our published procedure.  To date, we have screened a total of 26 MetAP inhibitors from our MetAP collection to further validate the RpMetAP assay. These initial tests have revealed compounds capable of inhibiting R. prowazekii MetAP across a 1000-fold range in IC50 values (IC50 values greater than 100 mM to as low as 640 nM).  During the 1st and 2nd quarters, we will screen our entire library of MetAP inhibitors to determine IC50 values.  Chemical series that demonstrate rational SAR and submicromolar potency will be identified as leads for additional studies. This data will be used to prioritize compounds for in vivo and crystallization studies. 

Aim 2: Determine the in vivo activity of the NIU metAP compound library against Rickettsia prowazekii (USA: Audia)

By virtue of their intracellular lifestyle, rickettsiae cannot be grown in liquid culture medium or on solid agar in the Petri dish.  Thus, a unique growth environment necessitates that inhibitor screens consider: 1) use of a disease-relevant vascular endothelial cell phenotype; 2) potential adverse effects of the compound library (and solvent) on the eukaryotic host cell; 3) effects of the compound library (and solvent) on extracellular R. prowazekii prior to host cell infection; 4) effects of the compound library (and solvent) on R. prowazekii growing intracellularly; and 5) that all infection studies must be performed under BSL-3 safety conditions.  Dr. Audia has developed an infection model using primary cultured rat vascular endothelial cells and a rapid intracellular staining protocol to detect infected cells using a fluorescently labeled antibody and flow cytometry.  Moreover, the University of South Alabama has a long-standing BSL-3 research history, the facilities, and CDC certification to carry out the proposed studies with a Rickettsia Select Agent.  Together, this disease-relevant cell culture model and enumeration technique amenable to mid-high throughput represents high potential for success to identify lead compounds from Drs. Hagen/Horn’s in vitro screen that target the R. prowazekii MetAP in vivo as a novel therapeutic.

The proposed screening strategy was designed acknowledging that BSL-3 work is inherently expensive and hazardous and therefore, should be minimized wherever possible. It is important to note that the in vitro pre-validation of potential inhibitory compounds will be used to prioritize which compounds will be tested for effects on rickettsial growth in vivo.

Eukaryotic Toxicity Screen (Audia Phase 1): We have previously determined that rat vascular endothelial cells (ECs) tolerate the DMSO solvent used to dissolve Dr. Hagen’s MetAP-directed compound library.  Thus, we will first determine the toxicity of the in vitro-validated compound library on uninfected ECs grown in culture.  ECs grown in 96-well plates as adherent, confluent monolayers (which best approximates their in vivo function of forming tight barriers to the movement of fluid and proteins) will be incubated with a maximal dose (to be determined based on results of the in vitro screen) of each compound for 72 hours.  This time point is selected because of the slow generation time of R. prowazekii (8-10 hours for one rickettsia to replicate) to allow for proper assessment of growth effects.  Post-exposure, ECs viability will be assessed using the WST-1 cell proliferation reagent (Roche), which can be directly determined in the 96-well plate by spectrophotometry. Any compounds that kill the ECs will be eliminated from the screen, as adverse effects on the host will artifactually inhibit R. prowazekii growth. This simple technique will allow us to rapidly screen the compound library during the third quarter to determine which compounds should be tested in BSL-3 infection studies.

R. prowazekii Growth Inhibitor Screen (Audia Phase 2): The strategy to identify MetAP inhibitors that inhibit/limit R. prowazekii growth in ECs will be tailored to account for points 3 and 4 described above where compounds may target extracellular rickettsiae, intracellular rickettsiae, or both.  Thus, we will identify compounds that poorly permeate ECs but permeate rickettsiae and those compounds that permeate both ECs and rickettsiae. The difference may be subtle but important. Rickettsiae must adhere to a host cell, stimulate endocytosis, escape from the phagosome, grow in the cytoplasm, lyse the host cell, and repeat the cycle.  Thus, compounds that don’t readily enter ECs may still be efficacious in binding MetAP and inhibiting the early portion of the described lifecycle.  Compounds that permeate both host and rickettsiae would limit all portions of the lifecycle.  To accomplish the compound screen, ECs will be infected at a low multiplicity of infection that will inoculate only ~30-50% of the cells allowing rickettsiae to grow, lyse, and infect adjacent cells (and mimic the in vivo condition). At 72-hours post-infection, the ECs will be harvested and rickettsiae infected cells enumerated by intracellular staining with a FITC-labeled antibody and flow cytometry. This technique gives information on the numbers of ECs infected and their relative burden, allowing for sensitive measurement of infection and growth inhibition. This phase of the screen will be completed during the third quarter. We are currently testing a 96-well technique where intact cells are washed and stained with the antibody to determine if throughput could be increased using a fluorescent plate reader.

Testing Lead Compound Specificity (Audia Phase 3): This final phase will be critical to validate that the observed growth inhibition is due to direct effects of lead compounds on R. prowazekii MetAP. It is well established that rickettsial entry into a eukaryotic host cell is an energy-dependent process. Thus, any compounds that spuriously act as energy poisons (e.g., uncoupling proton motif force) will inhibit rickettsial entry into the host cell. Dr. Audia is an expert in the biochemisty and physiology of rickettsial transport systems and has devised a secondary screen to identify potential off-target effects. Purified, isolated rickettsiae can survive for short periods of time outside of the host cell and are metabolically active. These purified, isolated rickettsiae will be incubated with identified lead compounds (individually) and assayed for their ability to transport [14C]-Lysine which is an energy-dependent transport system.  Controls will include the assay of [32P]-ATP transport which is energy-independent. Inhibitors that uncouple rickettsiae and block energy-dependent transport will be excluded from further analysis. This phase of the screen will be completed during the third quarter.

Aim 3: Determine X-ray structures of lead compounds in complex with R. prowazekii MetAp


Chemical series that demonstrate rational SAR and submicromolar potency from Aims 1 and 2 will be identified as compounds of interest for structural studies. We anticipate identifying 2-5 different scaffolds that will be subjected to crystallization studies to determine the structure of MetAp/inhibitor complexes. These structures will serve to guide the next step of lead compound development.


Completion of the specific aims of this proposal (see timeline above) will identify compounds with lead attributes. These lead attributes include potency of 1 µM or less for R. prowazekii MetAP and have a discernible SAR. The SAR data will be a driving force in the design of new analogs. Fragment hit series that do not show lead attributes will quickly be identified and adjustments will be made to the design strategy. Finally, the outcomes of this work will serve as important preliminary data for future NIH grant submissions by Hagen/Horn and Audia. 


Functional Study 9: Rickettsia prowazekii MetAP inhibitors.

Project lead: Tim Hagen, Northern Illinois University and Jon Audia, University of South Alabama
Project collaborators: James Horn, Northern Illinois University
Status: Completed






Quarter 1


  • M1: Establish Rickettsia in vitro assay at. (Aim 1/Hagen/Horn)

  • M2: Establish Rickettsia in vivo assay (Aim 2/Audia)



Quarter 2


  • M3: Complete in vitro screen (Aim 1/Hagen/Horn)


Quarter 3


  • M4: Complete in vivo screen (Aim 2/Audia)


Quarter 4


  • M5: Complete co-crystallization of hit compounds (Aim 3/SSGCID)



Summary: The overall goal of this project was to identify lead compounds that inhibit in vitro activity of the R. prowazekii MetAP enzyme and determine if any inhibit rickettsial growth using a disease-relevant endothelial cell infection model. During the past six months the following results were obtained. At NIU the IC50 values were completed for the remainder of single point hits.  A total of 39 new compounds were discovered to have IC50 values ranging from 274 nM to 97 M.  In the past six months the Audia group finished screening 24 compounds in triplicate over a 10-fold dilution series.  Experiments were performed using triplicates per plate.  The entire experiment was repeated twice.  The Audia group also completed the fluorescent microscopy experiments over that same time frame (again, two biological replicate experiments).  Other experiments were also attempted including testing of a more narrow range MIC of some of the compounds that showed potential inhibitory efficacy.  However, it was determined that the infection efficiency was too low in these experiments to allow for reasonable interpretation.  The project came to a close before Dr. Audia's group could fully trouble-shoot and re-try experiments although it was most likely an issue with repeated freeze-thaw cycles of the rickettsial infection stocks.  However, significant data generated by the initial screen yielded important information that has poised the group to move into pre-clinical models of rickettsial disease as part of a larger grant project.  Purification of the enzyme using EDTA allowed removal of metal ions from the enzyme which has allowed for variation of the metal ions present in the enzyme.  Removal of metal ions using EDTA chelation was followed by attempted crystallization with Mn+2 and inhibitor.  The attempts resulted in 2.1A crystals that did not contain metal or inhibitor.

Specific Aim #1:  Determine the in vitro inhibitory activity of the NIU metAP compound library against RpMetAP (NIU:Hagen/Horn)

Completed. During the first quarter of this project Dr.’s  Hagen and Horn at NIU established and used a plate-based assay to evaluate inhibitors targeting R. prowazekii MetAP (SSGCID batch RiprA.00039.a), thus fulfilling Milestone 1. The Hagen lab confirmed that this enzyme was functional and the assay follows the enzymatic cleavage of a Met-AMC substrate During the first six months A total of 500 potential RpMetAP inhibitors from the Hagen compound collection has been screened, determining the percent inhibition at 10 mM.  During the first six months of the project compounds that demonstrated significant inhibition at 10 mM were followed up with IC50 determination. The remaining analogs had IC50 values determined during the 3rd quarter of the project.  The screen has  revealed a total of 39 new compounds that  inhibit R. prowazekii MetAP with IC50 values ranging from 273 nM to greater than 100 mM.

Milestone 3 is complete, all compounds have been screened in the RpMetAP enzyme inhibition assay and IC50 values were obtained on the single point hits. 


Specific Aim #2:  Determine the in vivo activity of the NIU metAP compound library against Rickettsia prowazekii (USA: Audia).

Completed. Rickettsiae are obligatory intracytoplasmic parasites that cannot be grown in liquid culture medium or on solid agar in the Petri dish.  Thus, all inhibitor-testing studies must be performed on rickettsiae-infected eukaryotic host cells.  This raises an important caveat that any compounds (or their DMSO vehicle) which inadvertently kill the eukaryotic host cell will also kill rickettsiae giving a ‘false positive’. In consideration of this caveat, The Audia group completed assessment of two disease-relevant, primary endothelial cell lines isolated from rats as candidate infection models for the in vivo inhibitor screening phase of the project.  As part of this assessment, Dr. Audia validated a moderate-throughput 96-well plate screening assay based on metabolism of the eukaryotic cell. The basic premise is that uninfected cells will exhibit a robust redox state measured using the WST-1 reagent in a colorimetric assay.  Cells infected with rickettsiae will lyse over time resulting in a reduced WST-1 signal.  If a compound inhibits rickettsial growth, then the WST-1 signal will increase comparable to uninfected cells. Dr. Audia has used this WST-1 screen to prove detection limits and assay linearity by seeding cells at 2-fold increasing densities (range 6,250 – 200,000 cells).

Using the WST-1 assay, pulmonary arterial endothelial cells (PAECs) and pulmonary microvascular endothelial cells (PMVECs) were vetted for growth rates and susceptibility to DMSO (the vehicle used to dissolve the inhibitor compounds). It was determined that the PAECs display the desired characteristics such as slow growth rate and a wider range of WST-1 metabolism compared to PMVECs making them most amenable to the assay.  Dr. Audia tested DMSO for effects on growth/viability over a range of 0.1 – 2% and determined that the PAECs could tolerate co-culture of ~0.3% DMSO for 48 hours before showing significant effects on WST-1 metabolism.  These data allowed the group to determine an appropriate concentration range to test inhibitors for off-target toxicity on PAECs prior to testing them on rickettsiae.  Finally, Dr. Audia has tested a range of infection conditions with and without 0.3% DMSO to optimize rickettsial infection and growth conditions. This work fulfills Milestone 2.

It should be noted that execution of the studies outlined in Specific Aim 2 were set back due to problems associated with completion of the Laboratory of Infectious Diseases BSL-3 select agent facility at the University of South Alabama.  All problems have been remedied and the facility is currently ‘hot’ and performing infection experiments.  However, development of the WST-1-based assay has increased the throughput capacity and will allow the Audia group to put the in vivo screening portion of the project back on schedule.

Dr. Audia’s group completed work described in Milestone 4 by testing compounds identified in Milestone 1 by Drs. Hagen and Horn.  The testing consisted of performing a crude solubility analysis for each of the compounds.  This led to the exclusion of 2 compounds that were insoluble in DMSO.  The remaining compounds were then tested for off-target toxicity on PAECs.  These data were used to redesign the screen using only compound concentrations (10-fold range) that showed little-to-no toxicity.  These compounds were tested for inhibition of rickettsial growth in PAECs and identified 10 compounds that potently inhibited rickettsial growth in vitro. 

In addition, during the last performance period we performed fluorescent microscopic analyses as validation of one of the compound identified as inhibitory to rickettsial growth.  

Specific Aim #3:  Determine X-ray structures of lead compounds in complex with R. prowazekii MetAp (SSGCID)

Ongoing. Work applied towards Milestone 5 has been initiated. Co-crystalliztion of furoic acid and oxine analogs that displayed in vitro potency for RpMetAP have been attempted.  Crystals have been obtained however the inhibitors were not bound. 

We have attempted to co-crystallize RiprA.00039.a.A4 and RiprA.00039.a.A6 with multiple compounds delivered by the project including HGN-0025, HGN-0026, HGN-0166 & HGN-0168. Crystals have grown and diffracted to 2.1 to 3.1 Å diffraction. However, no electron density is present for the inhibitor molecule. Metalloproteases, in particular Methionine Aminopeptidase's, have shown difficulty crystallizing metal bound inhibitors.  There are two metals in the active site. Unexpectedly, the human and E.coli enzyme have occasionally crystalized with three metals in the active site, with the third metal being chelated to the inhibitor molecule.  A hypothesis to explain the inherent difficulty of crystallizing MetAP with inhibitor bound is that the inhibitor compound chelates free floating excess metal in the crystallization experiment. The chelation likely reduces the affinity of the metal-bound compound for the protein molecule because there are already metals present in the active site which prevent the metal-bound inhibitor from entering the active site.  To overcome this, we repeated the purification of both RiprA.00039.a.A4 and RiprA.00039.a.A6 and added a EDTA dialysis step before the final SEC column. The dialysis removes metal from the protein and allows for the co-crystallization of inhibitor with exogenous metal added.  Purification of RiprA.00039.a.A4 was successful, however RiprA.00039.a.A6 precipitated during dialysis suggesting instability without metal to stabilize the shortened construct.  Unfortunately, repeat crystallization still resulted in lack of metal-bound crystals. Continued work is underway using core pipeline funds, to optimize co-crystallization conditions.

The Horn lab has expressed and purified the RpMetAP enzyme.  EDTA was used to remove the metal ions.  The metals Zn+2, Mg+2, Mn+2 and Co+2 were individually introduced into the enzyme.  The introduction of Co+2 resulted in an enzyme with substantially greater activity compared to Mn+2. The original lot of RpMetAP enzyme contained Mn+2.  Future studies on the role of the metal ion in the RpMetAP enzyme will be carried out and will be used to support grant applications.  In addition, the enzyme will be used to characterize the inhibitors using biophysical approaches.