MICROTITER ASSAY FOR ANTIVIRAL EFFECTS OF HUMAN AND MURINE INTERFERON UTILIZING A VERTICAL LIGHT PATH PHOTOMETER
FOR QUANTITATION
Lois B. Epstein Nancy H. McManus Samuel J. Hebert Judith Woods-Hellman
Diane G. Oliver
I. INTRODUCTION
Much effort has been made by many investigators to define the cellular origin of interferon produced in response to many diverse agents. Our own studies have demonstrated that human lymphocytes are the source of interferon produced in response to mitogens and antigens (1) and that macrophages enhance such interferon production by the lymphocytes (2). The type of in- terferon produced by stimulated lymphocytes was formerly called type II, or immune interferon and, now by the new nomenclature for interferon, IFN-γ (3). Human IFN-γ is antigenically dif- ferent from other types of interferons and distinguishes itself further by virtue of its greater lability on exposure to low pH and heat. The other types of interferon were formerly known as type I or classical interferons or as leukocyte or fibroblast interferon, and now are designated as IFN-α and IFN-ß, respec- tively. The comparative biology of the various types of inter- ferons has been reviewed (4). Macrophages, in addition to
METHODS FOR STUDYING Copyright © 1981 by Academic Press, Inc.
MONONUCLEAR PHAGOCYTES 6 1 9 All rights of reproduction in any form reserved.
ISBN 0-12-044220-5
lymphocytes, can produce interferon (5-9), but the interferon produced is usually of the a or $ variety.
The microtiter assay for the antiviral effects of human and murine interferon that we employ in our laboratory is rapid, sensitive, and economical. It represents a modification of previously reported techniques (10-13), combined with a totally new approach for quantitating the assay. Fibroblasts are ex- posed to interferon containing samples and the cells are then challenged with bovine vesicular stomatitis virus (VSV). In- terferon protection conferred upon the cells is reflected by the degree of inhibition of virus-induced cytopathic effect.
By staining the monolayer, qualitative and quantitative assess- ments of protection can be made. Qualitative results can be obtained within 72 hr of cell seeding. Quantitative results are obtained within this same period by utilizing the Titertek Multiskan, an automated vertical light path photometer that permits the reading and recording of the optical density of a 96-well flat bottom microtiter plate in less than 1 min. The Titertek Multiskan also makes it possible for the entire assay to be performed in the original microplate.
II. REAGENTS
A. Cells
The fibroblasts employed in this assay for human interferon are derived from human skin biopsy or fetal tissue specimens that we possess in our own laboratory. Alternatively, normal human fibroblasts may be obtained from HEM Research, Inc., Rockville, Maryland and Microbiological Associates, Bethesda, Maryland, or human foreskin fibroblasts may be prepared as de- tailed by Epstein and McManus (14). As fibroblasts of patients with trisomy 21 (Down syndrome) have been shown to be three to ten times more sensitive to the antiviral effects of interferon as compared to matched diploid cells (15,16), we prefer to use these in our assay, especially when low levels of interferon are anticipated. Cells from such individuals may be obtained commercially from the Human Genetic Mutant Cell Culture Facili- ty, Camden, New Jersey or prepared directly, as we do, from tissue specimens.
L 929 cells used for the assay of murine interferon were kindly supplied by Dr. Werner Rosenau. These cells may be ob- tained commercially from American Type Culture Collection
(Catalog No. CCL-1), from Flow Laboratories, Inc., McLean, Virginia, or from microbiological Associated, Bethesda, Mary- land.
B. Tissue Culture Reagents
Dulbecco's modified Eagle Medium (3 gm glucose/liter) (DME) is supplemented with fetal calf serum (FCS), as indicated in the procedure, and 100 U/ml penicillin and 100 yg/ml strepto- mycin. For subculturing, the cells are trypsinized with a solution of saline A containing 0.05% trypsin and 0.02% ethyl- enediaminetetraacetate [Cell Culture Facility, University of California, or Grand Island Biological Co., (GIBCO), Santa Clara, California]. The DME is stored at 4°C; trypsin solution is stored at -20°C until use, then at 4°C.
C. Laboratory and International Interferon Reference Standards A human internal laboratory standard for interferon is pre- pared as outlined in Epstein and McManus (14) and is caîibrated against the International Reference Standard for leukocyte in- terferon (IFN-α), (G-023-901-527), which is available from the Viral Resources Branch, National Institutes of Allergy and In- fectious Diseases (NIAID), Bethesda, Maryland 20205. Our inter- nal laboratory standard for mouse interferon was obtained from the laboratory of the late Dr. Kurt Paucker and is calibrated against the International Reference Standard for mouse inter- feron (G-002-904-511), also available from NIAID.
To date no International Reference Standard for human or mouse immune interferon (IFN-γ) is available, and the other available human and mouse interferon standards are used as sub- stitute reagents. The use of such international reference standards and the reporting of interferon in international units allows for the comparison of results from numerous labo- ratories.
D. Virus
The Indiana strain of bovine vesicular stomatitis virus is propagated on chick-embryo fibroblasts, and plaque-forming units (PFU) per milliliter are determined for human and mouse cells as described by Epstein and McManus (14).
E. Stain and Reagent for Elut ion
Qualitative results of cytopathic effect (CPE) are visual- ized by staining the monolayer with a dye-fixative solution containing 0.5% crystal violet (MC/B Manufacturing Chemists, Norwood, Ohio), 5% formalin (v/v), 50% ethanol (v/v) and 0.85%
NaCl. Quantitation of CPE is accomplished by eluting the stain
with ethylene glycol monomethyl ether (Sigma Chemical Co., St. Louis, Missouri, (10). This is stored at room temperature in the dark; inhalation of fumes should be avoided.
III. PROCEDURE
A. Maintenance and Culture of Cell Lines
Culture and maintain the human fibroblasts in 75-cm2 Falcon flasks (Becton, Dickerson & Co., Oxnard, California) in Dul- becco's modified Eagle Medium (3 gm glucose/liter) (DME) sup- plemented with 10% FCS in a water-jacketed incubator (NAPCO, National Appliance Co., Scientific Products, Menlo Park, Cali- fornia) at 37°C in a humidified atmosphere containing 5% C02- Culture and maintain L 929 cells in the same manner using DME + 5% FCS.
B. Microassay for Interferon
Seed the cells used in the assay into a 96-well tissue culture plate (Linbro Division, Flow Laboratories, Inc.,
McLean, Virginia) in a 0.2-ml DME + 10% FCS-well, using a 1-ml serological pipette. The cell concentration for human fibro- blasts is 2.5 x 10^ cells/well; the concentration for L 929 cells is 1 x 10^ cells/well. Warm the microplates and medium to 37°C before use. To facilitate obtaining a homogeneous monolayer for uniform staining qualities, place the microplates on a Micromix (Hyperion, Inc., Cordis Corp., Miami, Florida) for 5 min immediately after seeding, after which they are incu- bated at 37°C in a 5% CO2 humidified atmosphere for approxi- mately 24 hr. At this time the monolayers should be greater than 95% confluent. Make serial dilutions of the samples to be tested in DME + 5% FCS and, in parallel, make serial dilutions of an interferon preparation of known concentration that has been calibrated against the appropriate international standard.
Remove the medium from each well by careful aspiration with a finely tapered Pasteur pipette that has been bent to a 90° an- gle. Add the interferon samples in a 0.1-ml volume/well.
Eight wells, reserved for both cell and virus controls, receive fresh medium. Incubate the plates under the above standard culture conditions for 20-24 hr.
Aspirate the interferon samples and control media. Add 0.1 ml DME (containing 2% FCS at 37°C for human cells, and 5% FCS for mouse cells) to each monolayer and reincubate the plates.
Dilute bovine vesicular stomatitis virus (VSV), containing ap- proximately 0.5 PFU/cell, in cold DME with FCS concentrations
for cells as indicated above, and add 0.1 ml to each test and virus control well after aspiration of the wash medium. Aspi- rate the medium from wells that serve as cell controls and re- apply 0.1 ml DME + 2% FCS at this time. Reincubate the cul- tures under the above standard culture conditions for 20 hr to obtain 3+-4+ CPE.
At the conclusion of the VSV challenge, aspirate the super- natant fluids into a virus decontamination waste container, gently wash the monolayers twice with 0.85% NaCl and stain for 20 min with 0.1 ml dye-fixative solution. Remove excess stain by aspiration, wash the monolayers three times with distilled water and allow the plates to thoroughly air dry. At this stage in the assay, the degree of protection can be qualita- tively assessed by comparing the staining intensity of the cell and virus controls with the interferon-treated monolayers.
This is facilitated by the use of a microtiter plate reading mirror (Flow Laboratories, Inc., McLean, Virginia). In addi- tion, if desired, the plates may be stored at room temperature in the dark for several weeks prior to quantitation.
To quantitate the amount of interferon protection, elute the stain by the addition of 0.2 ml ethylene glycol monomethyl ether to each well, using a Titertek 12-channel variable pi- petter (Flow Laboratories, Inc., McLean, Virginia). Place the plates on a vertical vibrator (A. H. Thomas Co., Philadelphia, Pennsylvania) to hasten total stain elution. Elution time is empirically determined and ranges from 10-40 min. Elution time should be long enough to elute dye from the darkest wells. The samples should be read immediately after total elution of dye, to minimize the progressive fading of the dye with time. After blanking the instrument with a Linbro plate, in which the eight wells in the first row contain 0.2 ml eluting fluid, the optical density of the eluate contained in each well is quantitated in a Titertek Multiskan, a vertical light path photometer (Flow Labo- ratories, Inc., McLean, Virginia) at 550 nm. The optical densi- ty of the 96 wells is automatically read and printed out within 1 min.
IV. CALCULATION OF DATA
Convert the optical density of the eluted stain to percent- age dye uptake by use of the following equation:
OD test - OD virus control
% Dye uptake = X 100 OD cell control - OD virus control
Plot the percentage dye uptake against the log of the inter- feron concentration. Using linear regression analysis of
points that fall between 15-85% dye uptake, calculate the 50%
protection point for each unknown sample. Do the same for the interferon standard run in parallel. The concentration of In- terferon contained in the dilution of the unknown sample pro- ducing 50% protection is equated with the concentration (in International Units per milliliter) of the standard producing 50% protection. By taking into account the dilution factor of the unknown sample, the concentration of interferon in the original unknown sample is calculated. A sample calculatiion is provided below.
The mean OD of eight cell controls (CC) from a given assay is 0.627, and the mean OD of eight virus controls (VC) is 0.329. The mean OD of three wells exposed to the human inter- feron standard known to contain 10 U/ml is 0.602. Thus, using the above formula,
0.602 - 0.329 0.273
% Dye uptake = 0.6 2 7 . 0.3 2 9 = ^ £ - 92%
Similar calculations made for other known concentrations of the human interferon standard give the following data (see the tabulation below):
Concentration of human interferon
standard (IU/ml OD Dye uptake (%) 10.0 0.602 92
5.0 0.538 70 2.5 0.464 45 1.0 0.400 24 0.5 0.365 12
Using the values between 15 and 85%, the linear regression is calculated. The IFN^Q is obtained from the linear regression data and is 2.6 IU/ml. Let us assume that one unknown sample at a 1/100 dilution gave 50% dye uptake. Then the original concentration of the unknown sample would be 260 IU/ml.
V. CRITICAL COMMENTS
One of the most crucial aspects of this assay, especially to obtain excellent reproducibility when quantitated on the Titertek Multiskan, is a homogeneous and well-stained mono- layer in each well. The cells employed in our laboratory for this assay have excellent sensitivity to interferon. The mean
±SE of IFN^Q from seven experiments with the human trisomy 21
fibroblasts was 2.9 ± 0.3 IU/ml; that of diploid human fibro- blasts in eight experiments was 7.3 ± 0.8 IU/ml; that of mouse L cells in 15 experiments was 5.3 ± 0.9 IU/ml. However, the fact that these human cell lines can become extremely dense and that the mouse L cell line does not exhibit contact inhi- bition posed potential problems in the sloughing off of the monolayers. This was overcome by paying careful attention to keeping constant the conditions of cell maintenance (i.e., cell splitting, time of cell splits, and percentage of fetal calf serum employed). Also, the choice of a 20-hr virus challenge minimizes time for overgrowth, and careful aspiration of fluids during the assay minimizes disturbance of the cell layer. The most reproducible monolayers well-to-well and assay-to-assay were obtained by the use of either a 1.0-ml or 2.0-ml serologi- cal pipette for initial seeding and were far superior to those obtained by seeding with an automatic pipetter. The coefficient of variation of the optical density of cell controls was £10%
for human fibroblasts and <13% for L cells using the serological pipettes.
We also observed that there is an unevenness of monolayer growth within the wells around the periphery of the microplate as compared with the interior of the plate. Thus we pay parti- cular attention to obtaining uniform air flow, temperature, and high humidity in our CO2 incubator, so that this uneveness is avoided. The reader is referred to the work of Martuza et al.
(17) and Duvall et al. (13) for further discussion of the causes and suggestions for correction of uneven humidity and gas ex- change .
The Titertek Multiskan allows the quantitation of a 96-well flat bottom microplate in 1 min, an incredibly short time as compared with other methods of automatic quantitation that em- ploy horizontal light path spectrophotometers (10-12), espe- cially those which involve manual transfer of each sample to a cuvette for reading optical density.
As the Titertek Multiskan is a vertical light path photo- meter, measurement of the optical density is in cuvettes with their longitudinal dimension parallel to the light beam. As the path length is dependent on the total volume, variation in the volume changes the light path by the same amount, but the absorbance is constant (18). Thus the OD readings are indepen- dent of the volume of eluate, and this feature counteracts pos- sible pipetting errors in the elution of stain from the mono- layers. These are minimal, however, as use of the 12-channel Titertek pipetter for adding eluate permits a coefficient of variation of <2% in replicate sampling.
As we recognized that the absorbance is independent of the volume of eluate, we naturally explored the possibility of read- ing the plates dry, without eluate, in the Titertek Multiskan,
to eliminate the time necessary for dye elution. However, we found that this was not feasible for the reason given below.
In the Titertek Multiskan with a 550-nm filter, the con- centration of eluted crystal violet is linear within the OD range of 0.2 - 1.7, and our assay conditions are designed to fall within the linear range. In the case of the virus control wells and wells containing samples in which there are low levels of interferon and thus minimal antiviral protection, the OD readings frequently fall below an OD of 0.2. This is because in such wells there is more central than peripheral destruction of each well by virus, and a rimming phenomenon occurs. The light beam sees only the central portion of each well. Despite at- tempts to minimize the rimming phenomenon by using Linbro plates, which give less rimming than Nunc or Falcon microtiter plates, the rimming phenomenon cannot be eliminated. Therefore in order to get accurate results it is necessary to use solvent to elute all the dye from each well.
For many years in our laboratory we employed a macro· virus plaque reduction technique for the assay of interferon (19).
We find the present microassay to be far superior because of considerable economy of time and materials and ease in quanti- tation of the assay.
Acknowledgments
This work was supported by NIH grant CA 27903 and
grant 6-126 from the March of Dimes Birth Defects Foundation.
The authors are indebted to Flow Laboratories, McLean, Virginia for making available a Titertek Multiskan and 550-nm filter for use in the development of this assay. We also thank Mary Evelyn Rose for typing the manuscript.
REFERENCES
1. L. B. Epstein. Mitogen and antigen induction of interferon in vitro and in vivo. Tex. Rep. Biol Med. 35:43-56, 1978.
2. L. B. Epstein. The ability of macrophages to augment in vitro mitogen- and antigen-stimulated production of interferon and other mediators of cellular immunity by lymphocytes. In "Immunobiology of the Macrophage"
(D. S. Nelson, ed.), pp. 201-234. Academic Press, New York, 1976.
3. W. E. Stewart II et al. Interferon nomenclature.
J. Interferon Res. l:vi, 1981.
4. L. B. Epstein. The comparative biology of classical (Type I) and immune (Type II) interferon. In "Biology of the Lymphokines" (S. Cohen, J. Oppenheim, and E. Pick, eds.), PP. 443-514. Academic Press, New York, 1979.
5. D. 0. Lucas and L. B. Epstein. Interferon and macro- phages. In "The reticuloendothelial System: A Compre- hensive Treatise, Vol. 4: Physiology" (J. P. Filkens and S. M. Reichard, eds.). Plenum, New York, 1981.
6. N. Maehara and M. Ho. Cellular origin of interferon
induced by bacterial lipopolysaccharide. Infect. Immun.
15:78-83, 1977.
7. H. B. Fleit and M. Rabinovitch. Production of interferon by in vitro derived bone marrow macrophages. Cell.
Immunol. 57:495-504, 1981.
8. C. Sorg, C. Neumann, V. Klimetzek, and D. Hannich.
Lymphokine-induced modulation of macrophage functions.
In "Mononuclear Phagocytes, Functional Aspects" (R. Furth, ed.), pp. 539-567. Martinus Nijhoff Publishers, The Hague, 1980.
9. R. W. Fulton and B. D. Rosenquist. In vitro interferon production by bovine tissues: Induction with infectious bovine rhinotracheitis virus. Am. J. Vet. Res. 37:1497- 1502, 1976.
10. J. A. Armstrong. Semimicro dye binding assay for rabbit interferon. Appl. Microbiol. 21:723-725, 1971.
11. N. H. McManus. Microtiter assay for interferon: Micro- spectrophotometric quantitation of cytopathic effect.
Appl. Environ. Microbiol. 31:35-38, 1976.
12. E. C. Borden and P. H. Leonhardt. A quantitative semi- micro, semiautomated colorimetric assay for interferon.
J. LaJb. Clin. Med. 59:1036-1042, 1977.
13. J. Duvall, A. Khan, N. 0. Hill, M. Ground, D. Muntz, and R. Lanius. A computer-assisted micro-dye uptake human interferon assay. In "Interferon: Properties and Clini- cal Uses" (A. Khan, N. 0. Hill, and G. L. Dorn, eds.), pp. 529-539. Leland Fikes Foundation Press, Dallas, 1980.
14. L. B. Epstein and N. H. McManus. Macro and microassays for the antiviral effects of human and mouse interferons.
In "Manual of Clinical Immunology" (N. R. Rose and H.
Friedman, eds.), pp. 275-283. American Society of Microbiologists Publications, Washington, 1980.
15. L. B. Epstein and C. J. Epstein. Localization of the gene (AVG) for the antiviral expression of immune and classical interferon to the distal portion of the long arm of chromosome 21. J. Infect. Dis. I3J;A56-A62, 1976.
16. J. Weil, L. B. Epstein, and C. J. Epstein. Synthesis of interferon-induced polypeptides in normal and chromosome 21-aneuploid human fibroblasts: Relationship to relative sensitivities in antiviral assays. J. Interferon Res.
1:111-124, 1980.
17. R. L. Martuza, M. R. Proffitt, M. B. Moore, and C. F.
Dohan, Jr. Evaporation as a cause of positional differ- ences in cell plating and growth in microtiter plates.
Transplantation 21:271-273, 1976.
18. H. G. Eisenweiner and M. Keller. Absorbance measurement in cuvettes lying longitudinal to the light beam.
Clin. Chem. 25:117-121, 1979.
19. L. B. Epstein. Assay of human interferon from lymphocyte macrophage cultures by a virus plaque reduction method.
In "Manual of Clinical Immunology" (N. R. Rose and H.
Friedman, eds.), pp. 120-128. American Society of Microbiologists Publications, Washington, 1976.
