Folia Medica 2014; 56(3): 182-186 Copyright © 2014 Medical University, Plovdiv doi: 10.2478/folmed-2014-0026
ORIGINAL ARTICLES
Experimental Investigations MICROMERITIC PROCEDURES IN ASSESSING ANTINUCLEAR ANTIBODY PATTERNS IN IMMUNOFLUORESCENT ASSAY Mariana A. Murdjeva1*, Emilia T. Milieva2, Rumen S. Stefanov3, Marian M. Draganov4, Petko B. Krastev5 of Microbiology & Immunology, 2Department of Physics, Biophysics & Mathematics, Faculty of Phar3 macy, Department of Social Medicine and Public Health, Faculty of Public Health, 4Department of Biology, Faculty of Medicine, Medical University, Plovdiv, Bulgaria, 5Institute for Nuclear Research and Nuclear Energy, Bulgarian Academy of Sciences, Sofia, Bulgaria 1Department
МИКРОМЕТРИЧЕСКАЯ ПРОЦЕДУРА В ЦЕЛЯХ АНАЛИЗА АНТИНУКЛЕАРНЫХ АНТИТЕЛ ПОСРЕДСТВОМ ИММУНОФЛЮОРЕСЦЕНТНОГО ТЕСТА Maриана A. Mурджева1*, Емилия T. Mилиева2, Румен С. Стефанов3, Mариян M. Драганов4, Петко Б. Крыстев5 1Кафедра Микробиологии и иммунологии, Фармацевтический факультет, 2Кафедра Физики, биофизики и математики, Фармацевтический факультет, 3Кафедра Социальной медицины и общественного здравоохранения, Факультет общественного здравоохранения, 4Кафедра Медицинской биологии, Медицинский факультет, Медицинский университет, Пловдив, Болгария, 5Институт ядерных исследваний и ядерной энергетики, Болгарская Академия наук, София, Болгария ABSTRACT AIM: The aim of this study was to introduce a micromeritic procedure (a statistical approach for small objects) in indirect immunofluorescence assay (IFA) to find objective quantitative parameters of antinuclear antibody (ANA) patterns which could support a diagnosis of auto-immune diseases. MATERIALS AND METHODS: Sera of patients with systemic autoimmune diseases, McCoy-Plovdiv serum-free cell line, goat anti-human immunoglobulin-G FITC-conjugate, fluorescent microscope, computer-assisted digital image processing, analysis using a micromeritic procedure, ANOVA. RESULTS: Three ANA fluorescent patterns (homogeneous, rim and speckled) were analyzed by the micromeritic procedure. Parameters for the image brightness of the pixels (pixel grey value) were obtained and discussed as objective characteristics of fluorescent patterns: maximum ANA-linkage volume and surface density were established for the objects with speckled localization pattern. CONCLUSION: The micromeritic method for getting objective quantitative values of ANA fluorescent patterns in indirect immunofluorescent assay might be a valuable tool aiding in immunological diagnosing if integrated in a laboratory software package. Key words: antinuclear antibodies, immunofluorescence, image analysis, FITC, pixel value distribution Folia Medica 2014; 56(3): 182-186 Copyright © 2014 Medical University, Plovdiv РЕЗЮМЕ ЦЕЛЬ: Работа ставит себе целью ввести микрометрическую процедуру (статистический анализ, применяемый для небольших объектов) в индиректный флюоресцентный анализ в целях получения объективных количественных параметров при анализе локализации антинуклеарных антител в диагностике аутоиммунных заболеваний. МАТЕРИАЛЫ И МЕТОДЫ: Применены: сыворотки пациентов с системными аутоиммунными заболяваниями; безсывороточная клеточная линия McCoy-Plovdiv; маркированный флюорохромом FITC иммуноглобулин; флюоресцентный микроскоп; цифровизирование и обработка изображений; микрометрический анализ; ANOVA статистическая обработка данных. РЕЗУЛЬТАТЫ: Три вида локализации антиген-антитела (ANA) реакции - гомогенная, периферическая и петнистая - анализированы микрометрическим методом. Считается, что яркость пикселов флюоресцентных изображений служит объективной характеристикой вида локализации. Максимальный объем и поверхностная плотность связанных с субстратом ANA получены для петнистого типа локализации. ЗАКЛЮЧЕНИЕ: Mикрометрический метод количественной характеристики типа ANA локализации в индиректной иммунофлюоресценции может оказаться ценным дополнительным инструментом для анализа в иммунологических диагностических лабораториях с возможностью для включения в пакет программного обепечения (sofware) компьютерной программы для обработки изображений. Ключевые слова: антинуклеарные антитела, иммунофлюоресценция, цифровой анализ изображений, FITC, распределение размера пикселя Folia Medica 2014; 56(3): 182-186 © 2014 Все права защищены. Медицинский университет, Пловдив Article’s history: Received: 06 February 2013; Received in a revised form: 16 June 2014; Accepted: 24 June 2014
182
*Correspondence and reprint request to: M. Murdjeva, Department of Microbiology & Immunology, Medical University, Plovdiv; E-mail:
[email protected]; Tel.: +359 882 038 807 15A Vassil Aprilov St., 4002 Plovdiv, Bulgaria Unauthenticated Download Date | 11/24/15 11:02 AM
Micromeritic Procedures in Assessing Antinuclear Antibody Patterns in Immunofluorescent Assay
INTRODUCTION
Antinuclear antibodies (ANA) are directed against various nuclear antigens, form complexes with them, and cause tissue damage in systemic lupus erythematosus and other non-organ specific autoimmune diseases.1 Indirect immunofluorescence assay (IFA) is the standard immunological test to screen ANA presence in sera of affected patients.2 Nuclear antigens, reacting with serum ANA, can be determined by secondary fluorochrome-conjugated anti-immunoglobulin antibodies. Fluorescein isothiocyanate (FITC) is the most widely used fluorochrome. The detected fluorescent patterns vary as homogeneous (diffuse), peripheral (rim), speckled, nucleoalar, centromere, etc. They correlate with different localization of antigen-antibody reaction and help to identify fine specificities associated with a particular disease or clinical manifestations. Detection of fluorescent type and intensity are highly dependent on observer’s qualification. Various procedures are used to improve the quality of ANA readings through sera screening dilution3, substrates4 and microscope technique5. The introduction of digital images and computer-assisted systems in IFA practice is a promising development that may help to achieve a more objective assessment of ANA patterns.5-7 AIM
The main goal of this study is to introduce the micromeritic procedure8 for obtaining objective quantitative parameters which could support fluorescent ANA determination and hence the diagnosis of auto-immune diseases. MATERIALS AND METHODS
IFA TESTING Procedures. The IFA of ANA testing from patients’ sera was performed as described previously using McCoy-Plovdiv serum-free cell line as a substrate for ANA detection.4,9 Ready-to-use goat anti-human immunoglobulin-G FITC-conjugate (Binding Site, UK) was used as a secondary antibody to visualize antigen-antibody reaction. The conjugate binds to those serum ANA immunoglobulins that react with the nuclear antigens of the cell line. Microscope examination of staining patterns. Slides prepared as previously described were examined on a fluorescent microscope (Optiphot 2, Nikon Corp., Japan) at 400x magnification to determine the staining pattern with planapochromate objective
lens – 40x. The reaction between ANA and nuclear antigens of McCoy-Plovdiv cells was visualized as apple-green staining of nuclear structures under UV light. The intensity and patterns of fluorescence were assessed and recorded. Selection of fluorescent patterns for analysis. Three nuclear fluorescent patterns, corresponding to different ANA localization in the nucleus were selected: homogenous (H) – diffuse staining of the interphase nuclei and staining of the chromatin of mitotic cells; rim (R) – peripheral solid staining around the outer region of the nucleus with weaker staining toward the center of the nucleus; speckled (S) – a fine or course granular nuclear staining of the interphase cell nuclei. Sera with subjectively assessed maximum fluorescent intensity (4+) were used for further analysis. Control sera from healthy individuals, negative for ANA, were used in the study. The non-specific fluorescent background, obtained in the negative controls, was subtracted from the fluorescent staining of ANA positive sera. DIGITAL IMAGE PROCESSING AND ANALYSIS The representative patterns were acquired by an operator with a charge-coupled device (CCD) camera Panasonic and transferred to a personal computer. The RGB color image was converted to a grey scale10 and 20 single objects (nuclei) were extracted from the image frames with different types of localizations. The brightness (grey level value) of the pixels were obtained from the line density plot profiles of the selected single nucleus by image software (Scion Image 4.0.32, Maryland, USA). The pixel grey value (PGV) distributions on a number basis (A) and on a weight basis (B) were obtained using the micromeritic procedure described elsewhere11 at different staining patterns. Distribution A corresponds to the number of pixels observed within a selected grey value range (Y-axis) plotted against mean PGV for ranges (X-axis). Distribution B is acquired from distribution A after calculation of the weight of the pixels. The number of pixels, having selected grey value or weight, are presented as percentage of the total number of pixels. Micromeritic parameters from distributions A and B (length, surface, volume, density) are calculated using the corresponding formulae. STATISTICAL ANALYSIS All measured values were validated and entered into a database for further analysis. Data were
Folia Medica 2014; 56(3): 182-186 © 2014 Medical University, Plovdiv
Unauthenticated Download Date | 11/24/15 11:02 AM
183
M. Murdjeva et al
statistically analyzed by descriptive statistics (mean ± SD values provided in the results) and one-way ANOVA. The ANOVA post hoc multiple comparison analysis were performed with the Tukey’s test in order to find which mean values are significantly different from one another. Statistical significance was set at p < 0.05. All statistical analyses were performed using SPSS v. 13. RESULTS AND DISCUSSION
The microscope images of the sera of three patients with different autoimmune diseases are presented in Fig. 1. The figure shows different patterns of fluorescent nuclear localization - homogeneous (H), rim (R) and speckled (S). These images correspond to (4+) semi-quantitative scale of ANA-sera patterns because this fluorescence intensity is easier for micromeritic testing. Data for PGV obtained after line density plot profile procedure were statistically analyzed and are presented in Table 1. The mean PGV among H, R and S localization differ significantly between H-localization of ANA
S (58.84 ± 2.50) and H (87.79 ± 2.81), as well as between S (58.84 ± 2.50) and R (78.42 ± 4.60) localizations (F=19.85, p < 0.001). Cumulative frequency plots with Distributions A (on a number basis) and B (on a weight basis) of PGV are shown in Fig. 2. Both distributions for the same samples put an emphasis on different asymmetry level of the antigenic nuclear object in terms of micromeritic parameters (length, surface, volume, density). These parameters are summarized in Table 2. The data in Table 2 further illustrate the importance of both distributions for better characterization of the digital image properties as a result of micromeritic approach: • the average PGV, calculated from the micromeritic parameter volume (Distribution A) of the antigen in the nucleus, has a minimum value for speckled localization pattern – 15.88 vs 113.99 (for H) and 94.71 (for R). This minimum value corresponds to maximum volume density for ANA-linkage to nuclear antigens11;
R-localization of ANA
S-localization of ANA
Figure 1. Three different ANA patterns detected on McCoy-Plovdiv cell line before image processing.
Table 1. Mean PGV of H, R and S-localizations of ANA Descriptive statistics
Localization H
R
S
N
366
363
409
MEAN
87.79
78.42
58.84*
SD
53.68
87.56
50.56
SEM
2.81
4.60
2.50
*statistically significant values (p < 0.001) for S vs. H or R localizations.
184
• the average PGV, calculated from the micromeritic parameter surface (Distribution B) has a minimum value again for speckled localization pattern – 75.07 vs 139.88 (for H) and 110.82 (for R). This minimum value is inversely proportional to the real surface11 of the nuclear antigens for ANA linkage and is, probably, an indication for maximum surface capacity of the nuclear antigens for ANA linkage. In this study we present quantitative parameters for objective determination of ANA fluorescent patterns by micromeritics – a statistical approach for small objects.8,11,12 The fluorescent intensity (brightness) data are represented by PGV for three distinct ANA fluorescent patterns – homogenous,
Folia Medica 2014; 56(3): 182-186 Unauthenticated Download Date | 11/24/15 11:02 AM
© 2014 Medical University, Plovdiv
Micromeritic Procedures in Assessing Antinuclear Antibody Patterns in Immunofluorescent Assay
Table 2. Average PGV calculated from Number distributions A and Weight distributions B Average PGV
Micromeritics parameters
H-localization
From distribution A length surface volume volume densitya From distribution B length surface surface densityb volume
R-localization
S-localization
87.80 102.90 113.99 ~ 1/113.99
78.42 87.56 94.71 ~ 1/94.71
58.84 63.30 15.88* ~ 1/15.88
120.61 139.88 ~ 1/139.88 151.77
97.77 110.82 ~ 1/110.82 120.94
68.11 75.07* ~ 1/75.07 81.15
a
volume density ~ 1/ volume; bsurface density ~ 1/surface; * statistically significant difference (p < 0.01).
rim and speckled. Homogeneous nuclear staining is a result of immune reaction between ANA and DNA in the nucleus; rim pattern appears in older cell cultures and represents ANA linkage to DNA too, whereas speckled pattern reflects reaction between ANA and extractable nuclear antigens (SS-A, SS-B, Sm, RNP). There have been a number of attempts to avoid subjective assessment of microscopic images in ANA readings by IFA. They focus either on sample preparation and adjustment of microscopes or applying computer-assisted approaches with image processing and analysis.6,7,10,13 Nevertheless, the relevant literature has little data about the positive aspects of micromeritic method and exploration of parameters in the field of ANA detection which have
80 60 40 20 0 50
100
150
200
S-localization of ANA
250
100
80
60
40
20
0 0
50
100
150
200
distribution B
distribution A
100 80 60 40 20 0 0
50
100
150
PGV
PGV
PGV distribution A
In conclusion, the obtained results suggest that the micromeritic method can be used to achieve objective quantification of ANA fluorescent patterns in indirect immunofluorescent assay. This approach supports the IFA test and is a valuable additional tool to immunological laboratory diagnostics, suitable to be included in software packages.
Cummulative % frequency undersize
100
0
CONCLUSIONS
R-localization of ANA Cummulative % frequency undersize
Cummulative % frequency undersize
H-localization of ANA
been introduced in the present study. The objective quantitative criteria used in this method could assist inexperienced observer in ANA detection pattern, especially in the differentiation between speckled fluorescence, on the one hand, and homogeneous and rim fluorescence, on the other.
distribution B
distribution A
distribution B
Distribution A is calculated from PGV on a number basis; Distribution B is received from the same pixels on a “weight” basis.
Figure 2. Cumulative frequency plots for three ANA patterns of localization. Folia Medica 2014; 56(3): 182-186 © 2014 Medical University, Plovdiv
Unauthenticated Download Date | 11/24/15 11:02 AM
185
M. Murdjeva et al
REFERENCES
1. Jost SA, Tseng LC, Matthews LA, et al. IgG, IgM, and IgA antinuclear antibodies in discoid and systemic lupus erythematosus patients. Scientific World Journal 2014:171028. 2. Mengeloglu Z, Tas T, Kocoglu E, et al. Determination of anti-nuclear antibody pattern distribution and clinical relationship. Pak J Med Sci 2014;30(2):380-3. 3. Ghosh P, Dwivedi S, Naik S, et al. Antinuclear antibodies by indirect immunofluorescence: Optimum screening dilution for diagnosis of systemic lupus erythematosus. Indian J Med Res 2007;126:34-8. 4. Zagorov M, Draganov M, Alimanska St, et al. Indirect immunofluorescent assay for antinuclear antibodies on McCoy-Plovdiv serum-free cell line substrate. Comp Immunol Microbiol Infect Dis 2007;30:153-62. 5. Hiemann R, Hilger N, Michel J, et al. Automatic analysis of immunofluorescence patterns of HEp-2 cells. Ann NY Acad Sci 2007;1109:358-71. 6. Nakabayashi T, Kumagai T, Yamauchi K, et al. Evaluation of the automatic fluorescent image analyzer, image titer, for quantitative analysis of antinuclear antibodies. Am J Clin Pathol 2001;115:424-9. 7. Rigon A, Soda P, Zennaro D, et al. Indirect immu-
186
nofluorescence in autoimmune diseases: assessment of digital images for diagnostic purpose. Cytometry Part B (Clinical Cytometry) 2007;72B:472-7. 8. Gupta MA. Micromeritics. [Internet] 2009 Aug 05. Available from: http://www.slideshare.net/ amitmgupta123/micromeritics-1812912. 9. Draganov M, Murdjeva M, Kamberov E. Development of a new serum-free cell culture system, McCoy-Plovdiv. In Vitro Cellular and Developmental Biology - Animal 2000;36:284-6. 10. Pavlova P. Quantitative assessment of fluorescence microscopic images. Automatics and Informatics. 2010;2:25-8 (Bulgarian). 11. Sinko PJ. Micromeritics. In: Sinko PJ (editor). Martin’s Physical Pharmacy and Pharmaceutical Sciences, 5th ed. Lippincott: Williams & Wilkins; 2006:533-40. 12. Ahmad MZ, Akhter S, Anwar M, et al. Colorectal cancer targeted Irinotecan-Assam Bora rice starch based microspheres: a mechanistic, pharmacokinetic and biochemical investigation. Drug Dev Ind Pharm 2013;39(12):1936-43. 13. Sack U, Knoechner S, Warschkau H, et al. Computerassisted classification of HEp-2 immunofluorescence patterns in autoimmune diagnostics. Autoimmun Rev 2003;2:298-304.
Folia Medica 2014; 56(3): 182-186 Unauthenticated Download Date | 11/24/15 11:02 AM
© 2014 Medical University, Plovdiv
