| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Antibodies are the main effectors in the defense against blood stages of Plasmodium falciparum. Although there is some evidence that antibodies to blood stage antigens correlate with reduced parasitemia,1–3 the search for immune correlates of protection are ongoing. Evaluation of growth inhibitory activities and establishing positive correlations with reduced parasite burden and disease are required to establish in vitro growth inhibition as a suitable surrogate marker for predicting protective immunity. To this end, high-throughput assay formats are needed for testing large numbers of clinical samples. Such assay formats are currently under development in our laboratory and in laboratories around the world.4,5 Previously, a major impediment to the full analysis of these types of samples was the relatively time-consuming nature of these assays, which was further confounded by the limited sample volumes available, particularly from pediatric volunteers. To circumvent this problem, we developed a miniaturized and high-throughput method to evaluate large numbers of samples with small volumes. Successful miniaturization will require more than simple volume reduction: physical factors that may influence overall assay results include the effect of surface-to-volume ratio on the rate and extent of evaporation, the rate of target temperature achievement, the loss of critical solutes to binding on well surfaces, and the effects of surface tension on mixing techniques.
Previously, we reported on various methods for measuring growth and/or invasion inhibition and determined that only methods measuring viable, metabolically active parasites are capable of determining a wide array of anti-parasite activities in serum samples.6 In our "standard" assay, parasite growth and/or invasion inhibition is quantified by determination of parasite lactate dehydrogenase (pLDH)7,8 from a one-cycle assay performed in a 96-well plate (wp) with 1% hematocrit (hct), 0.3–0.4% starting parasitemia in a final assay volume of 100 µL. We describe here the conversion to a miniaturized pLDH-based growth inhibition assay (GIA assay) using 384 wp as culture and assay vessels. The 384 wp was chosen because culturing volumes as low as 10 µL/well results in a surface to volume ratio that does not carry the risk of partial evaporation compared with 96 wp with volumes < 50 µL.9 Starting parasitemia of synchronized schizont cultures was determined on the day of the assay by flow cytometric staining with hydroethidine (viable parasitemia).6 Matrix experiments were conducted to examine the relationship between culture volumes (three levels: 10, 20, and 30 µL); percentage hct (two levels: 1% and 2%); and starting parasitemia (four levels: 0.1%, 0.2%, 0.3%, and 0.4%). Finally, pLDH substrate volumes needed for optimal performance of the miniaturized assay were determined once the other culture and assay conditions were optimized.
Serum samples were first heat inactivated for 20 minutes at 56°C, cooled on ice, and pre-absorbed for 1 hour using 50% hct red blood cells (RBCs; 5 µL RBCs for every 100 µL of sera) at room temperature with intermittent mixing. After the pre-absorption, samples were centrifuged for 2 minutes at 14,000g to pellet RBCs. The required amount of serum needed was removed without disruption of the RBC pellet and transferred to and diluted in 96-well V-bottom plates, followed by a transfer with a multi-channel pipette (Rainin [Woburn, MA] 8- and 12- channel pipette, 5- to 50-µL range, LTS tips) into the 384 wp. This allowed for higher throughput loading and decreased sample handling time, which minimized dissolved gas exchange and pH shifts.10 To minimize evaporation, the outer two rows and columns of wells in the 384 wp were filled with 90 µL/well sterile, distilled water. Under these conditions, no fluctuation in parasite growth was observed for any assay well position. After loading the sample into wells, the 384 wp was placed into plastic bags (Thomas Scientific, Swedesboro, NJ) containing moistened blotting paper. The bags were gassed with 5% CO2, 5% O2, 90% N2, sealed, and allowed to equilibrate for 30 minutes at 37°C. Next, parasites were added to the plates with either a Rainin pipettor (see above) or a multi-channel pipettor specialized for 384 wp (Axypet-16; Axygen Scientific Tips, Axygen Union City, CA), and the plates were incubated for one cycle length (P. falciparum strain, 3D7 = 40 hours, FVO = 44 hours) gassed at 37°C. To harvest parasites, 85 µL of cold phosphate-buffered saline (PBS) was added to each well, and the plates were spun at 1,300g for 10 minutes at 4°C. Eighty-five microliters of the supernatant was removed from each well, and the plates were sealed with a self-adhesive plate sealer and stored frozen at –30°C until analysis. To develop the assay, plates were thawed for 30 minutes at room temperature with the plate cover in place (to avoid evaporation), and the substrate solution was prepared as previously described.7 Because the growth conditions were changed from our 96-well format conditions (volume reduced by a factor of five, and an increase in total cell number per well), we reevaluated the amount of substrate solution added to each well. In the 96-well format, we added 100 µL/well of substrate solution to 50 µL cell lysate (2:1). To optimize the ratio of substrate:lysate for the 384 wp, we tested ratios of 1:1, 2:1, 3:1, and 4:1 and identified the ratio of 2:1 as optimal with regard to signal:noise (data not shown). After addition of the substrate solution, plates were centrifuged at 1,800g for 1 minute and incubated on a plate shaker (Titramax 100; Heidolph Instruments, Schwabach, Germany) at 350 rpm. Plates were incubated for a total of 30 minutes with substrate at room temperature with the plate cover in place. Finally, the optical density in the wells was read at 650 nm (SpectraMax Plus; Molecular Devices, Sunnyvale, CA). Growth inhibition was calculated using the following formula: %GIA activity = {1 – [(ODimmune – ODRBC)/(ODpre-immune – ODRBC)]} x 100.
ODs of cultures grown at 1% hct with various starting parasitemias remained low and did not show a titratable effect (Figure 1, A and C
). Thus, only 2% hct was considered for subsequent experiments (Figure 1, B and D). The results from the various assay volumes tested showed that a linear relationship between assay volume and OD could be obtained even in the miniaturized assays. Although we did not detect significant differences in the assay readout across the range of tested volumes, a total volume of 20 µL (i.e., 10 µL diluted serum and 10 µL parasite suspension) was selected because we did not find a significant difference between 20 and 30 µL (P = 0.441 for 0.3% starting parasitemia and P = 0.987 for 0.4% starting parasitemia in 3D7 cultures, two-tailed t-test, df = 12). Based on these experiments, we identified the optimal assay conditions summarized in Table 1.
|
|
) performed by the same operator on three different days. We found that, although the calculated CV was generally higher for the smaller volume format, the median value for the 384 wp fell well below 10% (median CV: 96 wp = 4.4%; 384 wp = 6.7%) and remained below the generally accepted threshold of 20% for biologic assays.11
|
). The three horizontal reference lines depict the average difference and the average plus 2 SD above and below the average difference. The background inhibitory levels using control sera were higher in the 96 wp format than in the 384 wp format. This observation is consistent with both the positive mean percent GIA difference for the rabbit samples of 9.1% (23.0% upper limit, –4.8% lower limit) and the positive mean difference for the human samples of 4.9% (17.5% upper limit, - 8.4% lower limit). Nevertheless, these data support the conclusion of no difference between the two assay formats (96 wp versus 384 wp), because both confidence intervals overlap at 0%.12 Agreement in the test results by both methods was further confirmed by using the McNemar test to assess the significance in the differences for the proportions of responders and nonresponders. We considered anybody having a higher GIA activity than 15% to be a responder because this has been the maximal background with nave serum observed ï in > 300 experiments. Results from this analysis showed that the difference in proportions between the two methods for the rabbit samples was 8.3% and for the human sera was 5.7% (P = 0.625). These results are in general agreement with those obtained by the Bland-Altman plot discussed above. Thus, we conclude that results obtained from the miniaturized assay format (384 wp) are in agreement with results from the 96 wp format and that these two methods can be used interchangeably as necessary to assess parasite growth inhibition. The miniaturized assay format will be particularly useful for preclinical and clinical studies where serum volumes are limiting, i.e., Aotus monkey trials and pediatric trials. Application of this method will finally allow the large scale testing of samples from infants and young children living in endemic areas and may enhance our understanding of the development of natural immunity.
|
Special considerations were made to quality control for plate variations by positioning negative and positive plate controls at various locations on the plate. Because every assay plate contains their own set of assay controls, bridging between-experimental runs is possible.
Last, this method can easily be extended to other GIA readout methods such as flow cytometric analysis or microscopic analysis and may not be restricted to the study of malarial blood stage parasites.
Received November 1, 2007. Accepted for publication December 15, 2007.
Acknowledgments: The authors thank Ann V. Stewart for critical reading of the manuscript and helpful discussions.
Financial support: This work was supported by the US Agency for International Development, Project Number 936-6001, Award Number AAG-P-00-98-00006, Award Number AAG-P-00-98-00005, and the US Army Medical Research and Materiel Command.
Disclaimer: Research was conducted in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals, NRC Publication, 1996 edition. Serum samples from adults participating in the Phase 1/2a malaria vaccine trial at WRAIR (ClinicalTrials.gov identifier NCT00385047) were obtained under a human use protocol approved by the Walter Reed Army Institute of Research Human Use Review Committee and the US Army’s Human Subjects Research and Review Board of the Surgeon General. All volunteers gave informed consent before screening and enrollment into the study, and serum samples obtained from the volunteers were labeled with an alphanumeric code to maintain anonymity. The authors’ views are private and are not to be construed as official policy of the Department of Defense or the US Army.
* Address correspondence to Elke S. Bergmann-Leitner, 503 Robert Grant Ave. 3W76, Silver Spring, MD 20910. E-mail: elke.bergmannleitner{at}us.army.mil 
Authors’ addresses: Elke S. Bergmann-Leitner, Elizabeth H. Duncan, John R. Burge, Michele Spring, and Evelina Angov, WRAIR, 503 Robert Grant Ave. 3W76, Silver Spring, MD 20910. Telephone: 301-319-9278, Fax: 301-319-7358, E-mail: elke.bergmannleitner{at}us.army.mil.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
