NOTE: New end date is 11/20/2011 per NSSC information (Ed., 5/31/2011)

After successful completion of Phase I, preliminary data generated during Phase II indicated that the NASA criterion measure for bone loss during space flight (i.e., 24 hr urinary ionized calcium excretion), while related to calcium and T-CCL levels in sweat samples actively produced during defined exercise, were not predictive of 24 hr urine calcium excretion rates. After consultation and review of the preliminary Phase II results by representatives of the NASA-Johsnon Space Center-Human Research Program (JSC-HRP), the focus of Phase II was redirected to explore collection of a 24 hr sweat sample, rather than collection of a discrete, exercise-induced “active” sweat sample, to determine if a 24 hr integrated sweat sample was predictive of biomarker concentrations found in 24 hr urine samples. This redirection of effort required the identification and validation of additional commercially available absorbant materials which did not contain endogenous biomarker signal as well as a means of extracting the biomarkers from the absorbant material compatible with fluid handling limitations in the space flight environment. After identifying and developing such a collection method, this approach was then utilized to answer the question of whether or not biomarker levels in an integrated 24 hr sweat sample was predictive of those found in a concurrent 24 hr urine sample in a convenience of healthy individuals.

After successful completion of Phase II, the NASA-JSC-HRP program indicated that they wished to deploy the sweat monitoring technology in a NASA bed-rest campaigns being performed at the GCRC at University of Texas Medical Branch (UTMB) instead of in young and old subject populations as originally planned. Unfortunately however, during the last 9 months of the project NASA had to postpone bed-rest operations resulting in a joint decision by NASA-HRP and the Principal Investigator to utilize spinal-injured patients recruited from the Texas Medical Center (rather than NASA bed rest subjects) in which to test the 24 hr sweat monitoring technology as means of assessing bone loss. The resulting time delay surrounding availability of bed rest subjects and the subsequent decision to utilize spinal chord injury (SCI) patients, coupled with the additional requirement to seek Committee for the Protection of Human Subjects (CPHS) approval for testing in a new subject population resulted in NASA granting a one year no-cost extension to the project. Limited data gathered in the final year of the project provides evidence that biomarker levels in a 24 hr integrated sweat sample are predictive of those levels found in 24 hr urine samples in SCI patients. These data indicate that 24 hr sweat sample collection (using a collection and analysis scheme compatible with space flight operations) in a terrestrial human population undergoing active bone loss in a similar fashion to crew members during space flight is a non-invasive, time-efficient alternative to on-orbit 24 hr urine void collection as means of assessing biomarkers of bone loss. In addition, the ability to perform this type of sample collection using a microgravity compatible approach to liquid sample handling and the use of simple colorimetric based analysis techniques is of notable operational relevancy. Our data and validated methodologies suggest that sweat biomarker analysis (as an operationally compatible means of assessing and/or monitoring bone loss in crew members during space flight) should be considered for further development as a “real-time” analytical method for assessment of space flight-induced bone loss and a valid means of monitoring the efficacy of “in-flight” bone loss countermeasures.

[Editor's note 3/27/2013: No Task Book report received. Progress section and Bibliography compiled from PI's Final Technical Report submitted January 2013]

NOTE: New end date of 5/20/2012 is Pending as of 10/27/2011 (Ed.)

NOTE: New end date is 11/20/2011 per NSSC information (Ed., 5/31/2011)

As the overall goal of this project is to determine whether or not excretion of bone loss biomarkers in sweat is predictive of bone loss biomarker levels detected in 24 hr urine voids, we revisited the possibility of collecting a sweat sample over a 24 hr period in order to provide a more appropriate comparison. During Years 1 and 2 of this project we investigated the accuracy of the existing commercially available absorptive patch sweat collection technology with regard to analyte concentration relative to unadulterated liquid sweat samples collected in a contemporaneous fashion at the same anatomical site. These data indicated that the cellulose matrix of the commercially available absorptive patch posed significant analytical problems.

These issues were that the cellulose matrix: 1) contained significant amounts of endogenous calcium signal; 2) resulted in significant retention of sweat-derived T-CCL that could not be extracted from the matrix using the standard method (i.e. extraction in ddH2O); and 3) that extraction efficiency from the matrix was variable between identical samples. To address these analytical issues during Year 2 of this project we investigated the use of two new absorptive materials, a glass fiber material and a polypropylene woven material, for the collection of sweat over a 24 hr period. Our initial experiments focused on the extraction efficiency of four bone loss biomarkers of interest, namely calcium, T-CCL, NTx and osteocalcin (OST). We also modified the extraction protocol to utilize acidified ddH2O (pH 5.0) or a Tris buffer (pH 5.0) to enhance the solubility of these biomarkers in the extraction buffer and reduce ionic interactions within the matrix of the absorptive material. Unlike the original commercially available cellulose absorptive pad technology, efficient recovery of known amounts of all analytes was achieved using our modified extraction method from both the glass fiber and woven polypropylene materials.

During Year 2 of this project, utilizing glass fiber absorptive pads (7.5cm2 in area) and our modified extraction protocol, we collected sweat samples from a total of 18 normal healthy subjects over a period of 24 hr at three separate anatomical locations, the shoulder, hip and fore-arm. We also collected 24hr urine voids from these subjects during the same 24 hr period. When the total amount of calcium excreted in sweat over a 24 hr period into a 7.5cm2 glass fiber absorptive pad was compared to the total amount of calcium excreted in a 24hr urine void obtained from the same individual, our results indicate that there is a near linear correlation between the two values in sweat collected from the shoulder (R = 0.90) and fore-arm (R=0.86) and contemporaneous 24 hr urine voids. A similar relationship between 24hr sweat T-CCL and 24hr urine void T-CCL levels was observed at the shoulder (R = 0.99) in a sub-set of these subjects (N=6) for which this analysis has been completed. Our data indicate that calcium and T-CCL levels detected in 24hr sweat samples are highly correlated with levels detected in 24hr urine voids from healthy individuals, suggesting that bone loss biomarker concentration in the criterion sample (i.e. 24 hr urine void) can be accurately predicted from that found in a sweat sample collected over the same 24 hr period using an absorptive pad technology.

Year 3 of this project has focused on validating this approach to monitoring bone loss biomarkers in a terrestrial population undergoing active bone loss. Due to issues with gaining access to NASA bed rest subjects, and after agreement with HRP, we decided to switch our focus to a clinical population undergoing active bone loss, namely spinal cord injured patients. This change in focus and the subsequent time delay in obtaining CPHS approval for this study, added to difficulties in recruiting subjects based on our initial inclusion criteria (i.e. 6 – 18 month post injury, injury site T4 or below) necessitated a no cost extension to the project of 12 months (new project end date May 2012). We have now broadened our study inclusion criteria by including SCI patients who are more than 18 months post injury. To help in recruitment to our study, we have begun collaborating with an ongoing SCI rehabilitation study being carried out at UH which attracts SCI patients of all types. During Year 3 of the project we have also developed a new protocol suitable for deployment on-orbit for the extraction and collection of analytes from the absorptive sweat patch based on the Salivette™ technology previously validated for use on-orbit. We are also testing the ability of the iSTAT analysis technology (previously flown and validated on Shuttle and ISS) to perform Ca2+ analysis in sweat samples.

To overcome these difficulties we developed a standardized sweat collection protocol in which subjects performed a defined exercise protocol at a fixed level of intensity (i.e. 30 min of aerobic exercise at an intensity of 40% of their age-predicted VO2max) under similar environmental conditions and hydration status. Sweat samples were collected at the fore-head using micro-fabricated polymer capillary arrays which collect sweat as an unadulterated liquid sample. This approach allows both determination of the volume of sweat produced and has the advantage of minimal residual analyte retention within the array due to its polymer construction unlike the existing absorptive sweat collection technology. Sweat samples were initially collected at discrete time intervals to determine when during exercise-induced sweating did biomarker concentration reach “steady-state” levels. These experiments indicated that biomarker concentrations (i.e. Ca2+ and T-CCL) in “active” sweat reached a “steady state” after 15 min of exercise period regardless of the volume of sweat produced by a particular individual. In addition, analyte concentration (i.e. T-CCL) in actively produced sweat was found to be highly correlated (R = 0.92) with plasma levels of the same analyte during the period sweat was being collected. However, comparison of analyte concentration produced during actively produced sweat and 24 hr urine voids in the same individual indicated, that although related in a linear fashion (R = 0.61), that sweat analyte excretion during the 15 min period of exercise was not predictive of total 24hr urine analyte excretion. As 24 hr urine void collection provides an integrated value for whole body biomarker excretion rates, comparison of the amount of biomarker excreted during a 15 min period of “active” sweating from 7.5cm of skin surface at the fore-head may be inappropriate and thus account for the poor correlations observed between total calcium and T-CCL in sweat versus 24 hr urine voids.

As the overall goal of this project is to determine whether or not excretion of bone loss biomarkers in sweat is predictive of bone loss biomarker levels detected in 24 hr urine voids, we revisited the possibility of collecting a sweat sample over a 24 hr period in order to provide a more appropriate comparison. During Year 1 of this project we investigated the accuracy of the existing absorptive patch sweat collection technology with regard to analyte concentration relative to unadulterated liquid sweat samples collected in a contemporaneous fashion at the same anatomical site. These data indicated that the cellulose matrix of the absorptive patch posed significant analytical problems. These issues were that the cellulose matrix: 1) contained significant amounts of endogenous calcium signal; 2) resulted in significant retention of sweat-derived T-CCL that could not be extracted from the matrix using the standard method (i.e. extraction in ddH2O); and 3) that extraction efficiency from the matrix was variable between identical samples.

To address these analytical issues we investigated the use of two new absorptive materials, a glass fiber material and a polypropylene woven material, for the collection of sweat over a 24 hr period. Our initial experiments focused on the extraction efficiency of four bone loss biomarkers of interest, namely calcium, T-CCL, NTx and osteocalcin (OST). We also modified the extraction protocol to utilize acidified ddH2O (pH 5.0) which theoretically would enhance the solubility of these biomarkers in the extraction buffer and reduce ionic interactions within the matrix of the absorptive material. Unlike the original cellulose absorptive pad technology, efficient recovery of known amounts of all analytes was achieved using our modified extraction method from both the glass fiber and woven polypropylene materials.

Utilizing glass fiber absorptive pads (7.5cm2 in area) and our modified extraction protocol, we have now collected sweat samples from subjects over a period of 24 hr at three separate anatomical locations, the shoulder, hip and fore-arm. We also collected 24hr urine voids from these subjects during the same 24 hr period. To date we have collected and analyzed samples from 8 subjects and have an additional 12 subjects enrolled in the study. When the total amount of calcium excreted in sweat over a 24 hr period into a 7.5cm2 glass fiber absorptive pad was compared to the total amount of calcium excreted in a 24hr urine void obtained from the same individual, our preliminary results indicate that there is a near linear correlation between the two values in sweat collected from the shoulder (R = 0.94) and fore-arm (R=0.91) and contemporaneous 24 hr urine voids. A similar relationship between 24hr sweat T-CCL and 24hr urine void T-CCL levels was observed at the shoulder (R = 0.99) in these subjects.

Our data concerning the relationship between both calcium and T-CCL levels detected in 24hr sweat samples as compared to those detected in 24hr urine voids from the same individual suggests that bone loss biomarker concentration in the criterion sample (i.e. 24 hr urine void) can be accurately predicted from that found in a sweat sample collected over the same 24 hr period using an absorptive pad technology. The use of a glass fiber absorptive pad and modified extraction buffer ensures consistent and efficient analyte extraction from the pad. We are presently expanding the number of subjects in our data set and also including NTx analysis in that data set. From an operational perspective, we are also investigating a number of separate approaches to performing the liquid-based extraction procedure under microgravity conditions and testing the ability of the iSTAT analysis technology (previously flown and validated on Shuttle and ISS) to perform the sample analysis.