Research Projects

2006 – 2011


Carl F. Pieper, D.P.H.- Project Leader
Tel: (919) 660-7525 FAX: (919) 684-8569 E-mail:
Funding Years: July 01, 2007 – June 30, 2009

The overall aims of this development project were to advance the analytic and statistical science of change and typologies of function and to apply these methods to the data sets and analytic issues in the Duke Pepper Center. The results of this project were intended to have utility in (1) analysis of secondary aims, and (2) advancement of analysis science. Results were distributed by the point when the proposed meta-analyses and final data reduction techniques would be required in future years.


Ricardo Pietrobon, M.D., Ph.D. – Project Leader
Tel: (919) 668-2054 FAX: (919) 681-8886 E-mail:
Funding Years: July 01, 2006 – June 30, 2009

This developmental project was aimed at modifying existing software applications for the purposes of creating a Database of Geriatric Databases. The end product will allow geriatric researchers to search efficiently, in a single Web-based repository, for information on a large number of geriatric databases of thematic relevance.

During the first year of funding relevant geriatric databases were reviewed, ranked and categorized for use in this study. We then developed web-based software that could house the databases and provide a search mechanism. The application has functionality for uploading new databases, as well as modifying content for any of the other databases, which are deployed in a wiki environment (see User Homepage and database information on the website): Upon locating the database, the researcher will find the contact information of the researcher who posted the database along with information about the database itself and edit functions. All databases have details regarding total number of observations, age group, data collection methods, database type, geographical area, interventions, longitudinal variables, pathogenesis, body system, time period, sampling, requirements for data use, and variable highlights/uniqueness in relation to other data sets. Register on this page for a username and password to view the site.

During the development phase, many informal website usability tests were conducted among researchers. The usability issues that were found in the initial prototype of the software were then resolved and further tests are currently being conducted. We will continue to improve the site as it matures over time and will conduct the formal tests during the year to come.

At the end of the development phase, the Database of Geriatric Databases was initially released within the Duke Pepper Center for internal testing and evaluation. After initial testing and modification, the Database of Geriatric Databases was released to the other participating Pepper Centers. After one year, a Web survey will be conducted across all centers. Support from the Pepper Center Coordinating Center at Wake Forest, headed by Dr. Kritschevski was ensured for this project.


Virginia B. Kraus, M.D., Ph.D. – Project Leader
Tel: (919) 681-6652 FAX: (919) 684-8907 E-mail:
Funding Years: July 01, 2006 – June 30, 2009

The purpose of this development project was to characterize and validate novel biomarkers of biological age that are based on protein racemization and isomerization. We hypothesized that racemization and isomerization of select proteins will be a predictive measure of functional status and morbidity with an emphasis on osteoarthritis (OA). To test these hypotheses there were three specific aims:

  1. To characterize levels of racemized and isomerized aspartate (Asp) residues in OA patient body fluids and tissue samples for correlation with chronological age, OA severity, and functional status by WOMAC index.
  2. To identify protein residue “hot spots” susceptible to Asp racemization and isomerization during aging within cartilage macromolecules, collagen II and aggrecan.
  3. To assess the utility of specific protein epitopes containing racemized/isomerized Asp as biomarkers predictive of functional decline through evaluation of the POP cohort samples at baseline and correlation with functional status at baseline and three years (radiographic severity of knee osteoarthritis and WOMAC index at baseline and three years; with grip strength, and ‘EPESE’ functional battery at three years).

Assay Development for Aim 1: We postulated that the characterization of both the quantity and state of post-translational modification of cartilage extracellular matrix proteins and fragments that appear in body fluids with aging are likely to be a better marker of cartilage damage compared with simple quantification of total protein epitope. This hypothesis relies upon evaluation of the proteins within cartilage having long half-life. Identification of these age-related changes may allow more specific biomarkers of OA to be developed. We have focused initially on the change of _-Asp to _-Asp (IsoAsp) which we and others have demonstrated increases with age. This modification causes a distortion in the peptide backbone which can be detected using the PIMT assay. As proof of concept, and to develop methods for the identification of modified epitopes within protein, we have investigated EDTA cartilage extracts containing high concentrations of cartilage oligomeric matrix protein (COMP). Using a commercial PIMT assay (Promega) for quantifying _-Asp, we find that COMP from aged human cartilage contains detectable levels of _-Asp. Preliminary experiments show there is approximately 2µM of _-asp/µg of COMP. While an EDTA extract of cartilage is mainly COMP, we modified the PIMT assay so that it would work in a sandwich ELISA, 96 well format, in which COMP is captured specifically by monoclonal antibody. Using this system we detected 123-243 dpm due to COMP over and above coating antibody background (437 dpm) could easily detect measurable PIMT activity above the background caused by the coating antibody. This has confirmed the presence of _-Asp in the COMP extracted from cartilage.

Assay Development for Aim 2: In the commercial PIMT assay the _-asp is labeled with a 3H-methyl group which is released as 3H-methanol when the _-asp is converted back to _-asp. The release of the 3H-methanol is used to determine the amount of the _-asp present within the protein. We have investigated and modified the buffer systems of the PIMT assay so that this methyl group is retained on the protein rather than being released in order to label and identify isoAsp hot spots within proteins of interest. We have increased the retention of the methyl groups so that even after 24 hr, over 50% of the methyl groups were retained on the protein. Preliminary data for Aim 3: Serum PIMT assays have been completed on 48 patient samples (half control, half knee OA). Serum PIMT was not associated with age overall (p=0.63) however serum PIMT was higher in knee OA versus control participants (p=0.048) and the association was unchanged after controlling for age (p=0.05). Surgical waste specimens have been collected from 20 patients at the time of knee joint replacement for OA. Cartilage extracts from lesion and non-lesioned areas have been produced and are being assayed for PIMT activity. Frozen sections are being prepared on which to perform in situ hybridization in the future for ßAsp localization by PIMT.

We have proceeded with a mass spectrometry approach to identify the _-asp hot spots in COMP as a paradigm. We have developed a urinary collagen II assay dependent upon monoclonal antibody 5109 to capture the 3/4 fragment of type II collagen (used in the OA-related biomarker uTIINE assay). This will afford the possibility of performing PIMT quantification of ß-Asp in collagen II epitopes for correlation with chronological age, OA severity, and functional status by WOMAC index. The mass spectrometry methods developed in Aim 2 using COMP will inform follow-up studies to identify the _-asp hot spots in collagen II.


James R. Bain, Ph.D. – Project Leader
Tel: (919) 479-2320 FAX: (919) 477-0632 E-mail:
Funding Years: July 01, 2006 – June 30, 2008

Comprehensive metabolomic profiling has the potential to provide novel insights into the processes of functional decline and degeneration associated with aging. There is growing evidence that aging and functional decline might involve alterations in lipid metabolism, but details of these changes remain to be established. The overarching purpose of this development project was to enhance our capabilities for analysis of lipid metabolism in bodily fluids and tissue samples from animal models and human subjects. Our three aims were directed at developing quantitative, high-throughput assays of several key classes of lipids that have heretofore been difficult and cumbersome to measure.

  1. To develop an assay of fatty acyl coenzyme As of different chain length and degree of saturation.
  2. To develop assays of ceramides and related sphingolipids.
  3. To apply these assays to tissues sampled by intra- and extramural Pepper investigators.

During the first year of funding the acyl CoA assay came on-line and was validated in rodent liver and skeletal muscle (Aim #1). Heptadecanoyl CoA was employed as an internal standard (IS), and the endogenous CoAs were quantified using calibrators prepared by spiking liver homogenates with authentic CoAs (Sigma, St. Louis , MO) with saturated acyl chain lengths C8 to C18 and unsaturated species of C16:1, C18:2, C18:1, and C20:4. Briefly, after IS spiking and solvent extraction, CoAs are further purified by solid-phase extraction (Oasis HLB cartridges, Waters Corporation, Milford, MA) and then loaded into 96-well plates. We have developed a high-throughput assay (20 samples per hour) by using flow-injection analysis with positive electrospray-ionization tandem mass spectrometry (Waters Quattro micro). This is a marked improvement over more conventional LC/MS/MS methods with throughputs of two per hour. Data are acquired in the multi-channel acquisition mode, monitoring neutral loss of the 507-amu fragment (phosphoadenosine diphosphate) while scanning from m/z 890-1060. Thirty-four acyl CoAs are measured in a typical rat liver, while nine species are measurable in a typical mouse gastrocnemius muscle. Results are expressed as moles analyte per unit mass wet tissue.

In one study (Monetti et al., 2007), an evaluation was made of the putative causal role of hepatic steatosis in the development of insulin resistance. Mice overexpressing human acyl CoA: diacylglycerol acyltransferase 2 (DGAT2) in the liver developed hepatic steatosis, with increased liver content of several lipids, including unsaturated long-chain fatty acyl CoAs as measured by our new method. Surprisingly, these mice with fatty livers remained retained normal glucose homeostasis and insulin sensitivity. Thus, our results indicated that hepatic steatosis can occur independently of insulin resistance.