close

Вход

Забыли?

вход по аккаунту

?

Western blot band intensity analysis. Application to the diagnosis of lyme arthritis

код для вставкиСкачать
ARTHRITIS & RHEUMATISM Volume 37
Number 8, August 1994, pp 1206-121 1
0 1994, American College of Rheumatology
1206
WESTERN BLOT BAND INTENSITY ANALYSIS
Application to the Diagnosis of Lyme Arthritis
KRZYSZTOF KOWAL and ARTHUR WEINSTEIN
Objective. To determine the usefulness of quantitative band-intensity analysis of Western blots for the
diagnosis of Lyme arthritis.
Methods. IgG Western blots for antibodies to
Borrelia burgdorferi were performed on sera from 39
patients with Lyme arthritis, 30 patients with syphilis,
50 patients with connective tissue diseases, and 10
healthy individuals. Band positions and band intensities
were calculated using a computerized image analysis
system.
Results. Lyme arthritis patients had more bands
and higher-intensity bands than did non-Lyme patients.
The presence of at least 2 bands of moderate to high
intensity (>40 optical units) or at least 5 bands of lower
intensity (>20 optical units) was over 90% sensitive and
100% specific for the diagnosis of Lyme arthritis. A
60-kd band was present in all Lyme arthritis patients.
The presence of an 83-, 39-, 21-, or 18-kd band was
highly specific for Lyme arthritis.
Conclusion. Band intensity analysis increases the
objectivity and accuracy of Western blot interpretation
for the diagnosis of Lyme arthritis.
Lyme disease is often a diagnostic consideration in patients with acute and chronic articular
complaints, especially monarthritis and oligoarthritis
(1). The typical patient with untreated Lyme disease
has an erythema migrans rash and arthralgia prior to
the development of intermittent migratory arthritis
and/or chronic monarthritis (2). However, 20-30% of
Dr. Kowal’s work was supported by a grant from the
Kosciuszko Foundation.
Krzysztof Kowal, MD: New York Medical College, Valhalla; Arthur Weinstein, MD: New York Medical College.
Address reprint requests to Arthur Weinstein, MD, Division of Rheumatic Diseases and Immunology, New York Medical
College, Valhalla, NY 10595.
Submitted for publication November 29, 1993; accepted in
revised form March 2, 1994.
patients do not recall a rash, and the diagnosis of
Lyme arthritis can be difficult because routine laboratory testing, synovial fluid analysis, and radiographic
findings are nonspecific (3,4). Demonstration of Borrelia burgdorferi within the joints try culture is not
clinically useful because the organism is slow growing
and has only rarely been isolated from joint fluid (5,6).
Demonstration of borrelial nucleic acids by polymerase chain reaction may be a promising diagnostic
test, but is not yet standardized or commercially
available (7).
Because of these limitations, the diagnosis of
Lyme arthritis is generally confirmed by the demonstration of an antibody response to B burgdorferi
antigens (8). This has been buttressed by the early
production of IgM and IgG antibodies in patients with
borrelial infection and the rarity of antibody-negative
Lyme arthritis (9,lO). Antiborrelial antibodies are
most commonly detected by enzyme-linked immunoadsorption assays (ELISA) (11). This test is easily
automated and is thus well suited to clinical laboratories, but has been fraught with a high frequency of
false-positive results and interlaboratory error (12-14).
For this reason, immunoblotting (Western blot) has
become a method for confirming the diagnosis of
Lyme disease when the clinical picture or the ELISA
results are equivocal (15).
Unfortunately there are no accepted standards
for performing a Lyme Western blot test or for determining a positive result (16). As with ELISA, Western
blot testing may vary according to the strain of
B burgdorferi employed and whether early- or latepassaged organisms are used (17). Criteria for positivity have changed over the years, even within the same
laboratory (10,15). Furthermore, “positive” Lyme
Western blots may be seen in other illnesses that can
cause arthritis as well as in healthy individuals residing
in areas endemic for Lyme disease (16,18). It
1207
LYME ARTHRITIS AND WESTERN BLOT ANALYSIS
has become apparent that interpretation of Western
blots for the diagnosis of Lyme disease must take into
account both the number of positive bands (the number of borrelial antigens to which the antibody response is directed) and the position of the bands (the
response to Borrelia-specific as well as nonspecific
antigens) (15,16).
Two disadvantages in current Western blot
methodology are the subjectivity in determining band
position, especially where there are closely apposed
bands (e.g., 39-kd and 41-kd bands), and the lack of
quantitation of band intensity, which can lessen the
ability to discriminate true-positive from false-positive
results. There have been preliminary attempts to quantitate Lyme Western blots by reflective densitometry,
a cumbersome technique, and these have suggested
improved specificity in Western blot interpretation
(19,20).
In this study, we devised an image analysis
system for the objective interpretation of Western blot
band positioning and intensity and evaluated its utility
in the diagnosis of Lyme arthritis.
PATIENTS AND METHODS
Patients. Sera from 39 patients who had a typical
clinical presentation of Lyme arthritis (mono-, oligo-, or
polyarthritis confirmed by a physician) and who had a
positive Lyme ELISA result in our laboratory were studied
(3). There were 16 female and 23 male patients, with a mean
age of 41 (range 8-71 years). Twenty-one (53.8%) had
monarthritis; 18 (46.2%) had oligo- or polyarthritis.
As non-Lyme controls, we studied 10 sera from
healthy individuals living in a nonendemic area, 30 sera from
patients with syphilis (VDRL- or RPR-positive sera, provided by Westchester County Health Department), and 50
sera from patients with connective tissue diseases (CTD)
who had been seen in our Faculty Practice and had serum
samples stored in our serum bank. We selected the CTD
patients who had high reactivity for antinuclear antibodies or
rheumatoid factor. Sera were stored at -70°C until tested.
Lyme ELISA and Western blot. The antibody response to B burgdorferi was determined using a polyvalent
ELISA (21). Microtiter plates were coated with 5 pdml
sonicated B burgdorferi B31 (a Westchester tick isolate
originally donated by Dr. Charles Pavia, Division of Infectious Diseases, New York Medical College and propogated
in our laboratory) and incubated overnight at 4”C, washed
with phosphate buffered saline-Tween, and blocked with 1%
bovine serum albumin. Individual test sera, negative and
positive control sera, and standards were added at a 1:200
dilution.
Plates were developed with goat anti-human heavy
chain antisera conjugated with alkaline phosphatase and
phosphatase substrate reagent. Optical densities (ODs) were
determined using a microplate reader (Dynatech, Alexan-
dria, VA), and results were calculated in relative OD units
from a standard curve (twofold dilutions of a known highly
positive serum).
IgG reactivities to individual antigens were measured
by Western blot using commercial kits (MarDx, Carlsbad,
CA). Each individual kit was run at the same time. Incubation periods and development times were kept constant with
each run, and the same control sera were used for each run.
Band position and band intensity analysis. The position and intensities of the bands on Western blot were
determined using an image analysis system. The system
consists of 5 elements: the image source (NC-15 CCD Color
Camera; NEC, Wood Dale, IL), a video raster screen that
displays the image (Model PVM-1343 MD; Sony, Park
Ridge, NJ), a frame grabber (PCVISION Plus; Imaging
Technology, Woburn, MA), and a computer and monitor
that display the image for analysis (JAVA [Jandel Video
Analysis Software]; Jandel Scientific, Corte Madera, CA).
The basis of the technique was as follows: a video camera
produced an analog video signal containing image data
(Western blot strip) which the image processor converted to
a digital signal containing the image data in the form of digital
picture elements or “pixels.” The frame memory stored the
digital pixel data and this was accessed for display, feedback
processing, or additional computer processing after transformation back into an analog signal.
Band positions were determined on the computer by
running positive and negative reference sera provided by
MarDx and our own laboratory. For every Western blot
strip, the positions of individual bands were established by
overlapping the computerized results of the positive control
serum onto the investigated strips.
Band intensity was measured on a straight line drawn
along the center of each strip and was recorded in optical
intensity units (OU). The software was calibrated so that
white color corresponded to the lowest intensity (zero OU)
and black color to the highest intensity (100 OU). Determinations were made in duplicate and the mean result was
used. The results were stored on computer diskettes for later
analysis.
Statistical analysis. Sensitivity, specificity, and receiver operator characteristic (ROC) curves were calculated
by standard techniques (22). Categorical variables were
compared using chi-square analysis and continuous variables using the Mann-Whitney test.
RESULTS
Serum was available from 21 Lyme patients
pretreatment and from 18 posttreatment. The duration
of arthritis before treatment varied from 2 weeks to
several years (mean 14.8 months). The 50 CTD patients had the following diagnoses: rheumatoid arthritis in 16, systemic lupus erythematosus in 15, Sjogren’s syndrome in 4, and other CTDs in 15.
Lyme arthritis patients have stronger positive
Lyme ELISA results than do non-Lyme patients. The
polyvalent Lyme ELISA results in all 129 patients are
KOWAL AND WEINSTEIN
1208
Table 1. Lyme ELISA results in the study subjects with and without Lyme arthritis*
Subjects without Lyme
(n = 90)
ELISA
result
(OD units)
Patients
with
Lyme
(n = 39)
CTD
(n = 50)
Syphilis
(n = 30)
Healthy
(n = 10)
All
non-Lyme
Negative (<25)
Borderline (25-44)
Positive (45-100)
Strong positive (>loo)
0
0
6
33
44
4
2
0
3
14
9
8
2
0
0
55
20
4
11
4
* Values are the number positive. Enzyme-linked immunosorbent assay (ELISA) values are expressed
in relative optical density (OD) units. CTD = connective tissue disease (see Results for details).
shown in Table 1. In our laboratory, <25 OD units
(within 1 SD above the mean value for 50 healthy
controls) is negative, 2544 is borderline (1-3 SD above
normal mean), 45-100 is positive (3-10 SD above normal
mean), and >lo0 is strong positive (greater than 10 SD
above normal mean). In the non-Lyme group, 61% had a
negative Lyme ELISA result, 34% a borderline or positive result, and only 4% a strong-positive result.
By selection criteria, all the Lyme arthritis
patients had a positive Lyme ELISA result. None had
a borderline positive result, 15% were positive, and
85% had a strong-positive result. There was a significant difference in the degree of Lyme ELISA positivity between non-Lyme subjects and Lyme arthritis
patients (P < 0.0001).
Findings of band intensity analysis. Figure 1 is
an example of a Western blot strip from a patient with
Lyme arthritis and the corresponding intensity reading
83
45
66
75
60
39
31
41
34
29
MOLECULAR WEIGHT (kd)
25
18
21 15
Figure 1. Western blot of a serum from a patient with Lyme
arthritis and the corresponding intensity of each band.
for each band position. From the intensity readings, it
is apparent that the baseline is approximately 20 OU;
therefore, this value was adopted as the upper limit of
“normal” for determining a positive band. Band positions had previously been confirmed by monoclonal
antibodies for the 39-, 34-, 31-, and 25-kd bands and by
molecular markers and serum exchanges with other
laboratories for the other bands (Arthur Markovitz:
personal communication). In every strip, the highest
molecular weight band was 83 kd and the lowest was
15 kd. Many bands were detected in the 60-kd region,
and it was difficult to differentiate among them. The
30/31-kd bands also could not be distinguished from
each other because of overlapping; for instance, if a
strong 31-kd band was present, it was not possible to
determine if a 30-kd band was also present.
Band intensity reflects specific antibody concentration. To demonstrate that band intensity correlates
with specific antibody concentration, a strong-positive
serum was diluted, in doubling dilutions, from 1: 100 to
1:3,200. Western blots were run on these samples
simultaneously on the same kit, and the band intensities were calculated. The results are shown in Figure
2. In general, there was a linear decrease in the
intensity of individual bands over the range of dilutions studied.
Band intensity analysis is reproducible and reliable. Repeated image analysis of identical Western
blot strips gave essentially identical results each time,
with less than 5 OU difference in individual bands.
Repeated analysis of the same serum specimen run at
different times on different Western blot kits was less
reliable, given the variability of the Western blot
technique. Even so, there was less than 15 OU variation in individual band intensities.
Lyme arthritis patients have a greater number of
positive bands than do non-Lyme patients. The 14 most
prevalent bands were those of 83, 75, 66, 60, 45, 41,
1209
LYME ARTHRITIS AND WESTERN BLOT ANALYSIS
Table 3. Prevalence of individual bands >20 OU on Western blot
in Lyme arthritis and non-Lyme patients*
2
at
83 kd
75 kd
66 kd
60 kd
45 kd
41 kd
39 kd
34 kd
31 kd
29 kd
25 kd
21 kd
18 kd
15 kd
32
20
28
39
18
35
38
13
33
28
19
37
36
27
Subjects without Lyme
(n = 90)
CTD
(n = 50)
Syphilis
(n = 30)
Healthy
(n = 10)
All
non-Lyme
0
1
1
5
1
7
0
0
3
3
0
2
3
11
2
11
0
1
1
0
0
1
0
0
0
0
0
I
0
4
0
0
0
0
0
0
0
0
0
3
4
17
3
22
0
10
20
10
0
1:100
I
t200
1:1200
t:1600
?:no0
WOO
SERUM DILUTION
Figure 2. Intensity analysis of 5 bands from Western blots of one
serum sample run in doubling dilutions.
39, 34, 31, 29, 25, 21, 18, and 15 kd, and our analysis
is confined to these bands. As discussed above, only
bands with intensity of 20 OU or greater were considered positive.
The numbers of positive bands in the non-Lyme
and Lyme arthritis patients are shown in Table 2. Only
3 of the non-Lyme patients had more than 2 bands, and
none had more than 4 bands. All Lyme arthritis
patients had more than 2 bands, and 36 of the 39 had
more than 4 bands. The difference in numbers of bands
between these two groups is significant ( P < 0.0001).
Sera obtained pretreatment from 21 Lyme arthritis
patients had a greater number of bands (mean ? SD
12 ? 1.3) than did sera obtained postreatment from 18
patients (mean ? SD 8 -+ 2.9) ( P < 0.0001).
The most prevalent bands in Lyme arthritis, in
order of frequency, were 60, 39, 21, 18, 41, and 83 kd
(Table 3). The 60-kd band was present in all Lyme
Table 2. Number of bands >20 OU on Western blot in Lyme
arthritis and non-Lyme patients*
No. of
positive
bands
0
1
2
3
4
5-9
>9
Band
Patients
with
Lyme
(n = 39)
Patients
with
Lyme
(n = 39)
0
0
0
1
2
10
26
Subjects without Lyme
(n = 90)
CTD
(n = 50)
Syphilis
(n = 30)
Healthy
(n = 10)
All
non-Lyme
31
13
5
1
0
0
0
11
9
8
1
6
3
1
0
0
0
0
48
25
1
0
0
14
2
1
0
0
* Band intensity was expressed in optical intensity units (OU), and
calculated as described in Patients and Methods. CTD = connective
tissue disease (see Results for details).
1
0
0
3
1
4
3
I
I
0
3
* See Table 2 for details.
arthritis sera and the 39-kd band in 38 of the 39 Lyme
arthritis sera. In the non-Lyme patients, the 60-kd and
41-kd bands were the most prevalent individually and
were seen in combination in 9 patients.
Band intensity is greater in Lyme arthritis patients than in non-Lyme patients. The majority of the
positive bands in the non-Lyme group were relatively
faint, with readings between 20 and 40 OU. Very few
bands achieved an intensity >50 OU. In contrast,
patients with Lyme arthritis had many bands of higher
intensity (40-100 OU). The 83-, 39-, and 18-kd bands
were not present in the non-Lyme group. Sera from
the untreated Lyme arthritis patients had a higher
mean band intensity for each band than did sera from
the treated Lyme arthritis patients. The mean intensity
for the 83-, 39-, 21-, and 18-kd bands was >SO OU in
the untreated patients and 30-50 OU in the treated
patients.
The sensitivity and specificity of band numbers
and band intensity for Lyme arthritis is shown in Table
4. More than 4 bands of low intensity or as few as 2
bands of high intensity were 100% specific and -90%
sensitive for the diagnosis of Lyme arthritis.
For each band, the sensitivity and specificity
were calculated for 10-OU intervals (e.g., 20-29 OU,
30-39 OU, etc.) and ROC curves were computed. The
areas under the ROC curves, a measure of true- and
false-positive rates for a series of cut-off points, are
shown in Table 5. The highest values were seen with
the 83-, 60-, 41-, 39-, 31-, 21-, and 18-kd bands.
However, the minimal intensity for 100% specificity
1210
KOWAL AND WEINSTEIN
Table 4. Sensitivity and specificity for Lyme arthritis of numbers
of bands with specified band intensities*
No. of
bands
Band
intensity
(OU)
>20
2
2
4
4
5
>40
>20
>40
>20
Sensitivity
Specificity
(%I
(%)
100
94.8
97.4
82.1
92.3
81.1
100
98.8
100
100
* Band intensity was expressed in optical intensity units (OU), and
calculated as described in Patients and Methods.
was >40 OU for the 60-, 41-, and 31-kd bands, thus
diminishing their diagnostic value (Table 5 ) .
DISCUSSION
Our results are consistent with those of other
investigators, showing a high frequency (39%) of falsepositive Lyme ELISA results in patients without
Lyme disease, especially in patients with connective
tissue diseases and syphilis (12,16). However, a strong
positive result-> 10 SD above the normal mean-was
seen in 85% of Lyme arthritis patients and in none of
the CTD patients. Therefore a strong-positive Lyme
ELISA result is highly suggestive of Lyme disease.
Less positive and borderline results should be interpreted with caution. For this reason, Western blot analysis of the antibody response has become an important confirmatory diagnostic test for Lyme disease,
Table 5. Receiver operator characteristic (ROC) curve areas and
minimal band intensity for 100% specificity for Lyme arthritis, for
individual bands
Band
83 kd
75 kd
66 kd
60 kd
45 kd
41 kd
39 kd
34 kd
31 kd
29 kd
25 kd
21 kd
18 kd
15 kd
*
ROC
curve
area
Minimal intensity
for 100% specificity
(OU)
0.91
0.74
0.85
0.97
0.72
0.92
0.99
0.66
0.92
20
80
40
0.85
0.74
0.97
0.96
0.84
*
40
45
20
30
70
30
30
35
20
60
100% specificity was never achieved with the 60-kd band.
especially with inconclusive or low-positive ELISA
results (15).
We have overcome some of the limitations of
Western blot interpretation by devising a computerized image-analysis system to reproducibly define
band position and band intensity. We have shown that
band intensity is a reflection of antibody concentration. Using this semiquantitative analysis of Western
blots, very faint bands (<20 OU) were within background “noise level” and considered negative. This
eliminated many bands seen in the non-Lyme patients
and enhanced test specificity. The sensitivity of the
test was not diminished because of a high frequency of
many intense bands in the Lyme arthritis patients. In a
sense, we have developed a “normal range” for
Western blot band intensity, similar to the “normal
range” used in other tests such as ELISA. This
technique does not eliminate the problem of lack of
standardization of Western blot methodology. Although we chose one commercial product, there was
variation between test kits, resulting in up to 15 OU
differences in individual band intensity on the same
serum sample. However, this difference did not detract from the overall results.
Using this approach, there were 14 prevalent
distinct bands with reproducibly defined positions of
significant intensity in Lyme arthritis and non-Lyme
sera, at the 83-, 75-,66-, 60-, 45-,41-, 39-, 34-, 31-, 29-,
2 5 , 21-, 18-, and 15-kd regions. A similar pattern of
reactivity to borrelial antigens has been reported by
others (15,16,19,23). At least 1 distinct band was
observed in nearly 50% of the non-Lyme patients, but
the prevalence of reactivity to multiple antigens was
much lower. These results are similar to those obtained
by Fawcett et a1 (16). The most prevalent bands in this
group, singly and in combination, were at 41 and 60 kd.
The presence of any 5 of these bands at any
band intensity revealed very high specificity and sensitivity for the diagnosis of Lyme arthritis. This is
similar to the results described by Dressler et a1 (15).
When band intensity is taken into account, the presence of 2 or more bands, >40 OU, had a similarly high
sensitivity and specificity for Lyme arthritis. The
individual bands which exhibited the best combination
of sensitivity and specificity for Lyme arthritis were
83,39,21, and 18 kd. Bands in these regions have been
reported by others to have high diagnostic value for
Lyme disease (15,19,23). Interestingly, all Lyme arthritis patients had a 60-kd band. Its absence would
therefore virtually exclude that diagnosis. However,
the presence of antibodies to this antigen lacks speci-
LYME ARTHRITIS AND WESTERN BLOT ANALYSIS
ficity, since it is a homolog of the heat-shock protein 60
family, and antibody responses to it are seen with
many bacterial infections, some autoimmune diseases,
and even in healthy controls (24).
It has been shown that patients with Lyme
arthritis have a greater expansion of their antibody
response than those with earlier stages of Lyme disease (25). Therefore, it is likely that the results obtained in this study would not be generalizable to
patients with erythema migrans or neuroborreliosis.
Another caution in the interpretation of these results
relates to the lack of standardization of the Western
blot technique currently employed in commercial and
research laboratories that use different borrelial strains
and various methodologies. We are currently evaluating this Western blot technique in a large number of
sera from patients with early and late Lyme disease.
In summary, this system of image analysis of
Western blots is a reliable method to objectively
determine band position and band intensity. Lyme
arthritis can be diagnosed with great accuracy by the
presence of at least 2 high intensity bands, especially if
in the 83-, 39-, 21-, or 18-kd positions. Lyme arthritis
is unlikely if a 60-kd band is not present.
ACKNOWLEDGMENTS
The authors acknowledge the cooperation of Dr.
Piero Anversa, Department of Medicine, for allowing us to
use and adapt his computerized image analysis system and
the technical assistance of Ms Debra FitzDatrick.
REFERENCES
1. Steere AC: Lyme disease. N Engl J Med 321586596, 1989
2. Steere AC, Malawista SE, Hardin JA, Ruddy S, Askenase PW,
Andiman WA: Erythema chronicum migrans and Lyme arthritis: the enlarging clinical spectrum. Ann Intern Med 86:685498,
1977
3. Steere AC, Schoen RT, Taylor E: The clinical evolution of
Lyme arthritis. Ann Intern Med 107:725-731, 1987
4. Lawson JP, Steere AC: Lyme arthritis: radiologic findings.
Radiology 154:37-43, 1985
5. Snydman DR, Schenkein DP, Berardi VP, Lastavica CC,
Pariser KM: Borrelia burgdorferi in joint fluid in chronic Lyme
arthritis. Ann Intern Med 104:798-800, 1986
6. Schmidli J, Hunziker T, Moesli P, Schaad UB: Cultivation of
Borrelia burgdorferi from joint fluid three months after treatment of facial palsy due to Lyme borreliosis. J Infect Dis
158:905-906, I988
7. Nocton JJ, Dressler F, Rutledge BJ, Rys PN, Persing DH,
Steere AC: Detection of Borrelia burgdorferi DNA by poly-
1211
merase chain reaction in synovial fluid from patients with Lyme
arthritis. N Engl J Med 330:229-234, 1994
8. Craft JE, Grodzicki RL, Steere AC: The antibody response in
Lyme disease: evaluation of diagnostic tests. J Infect Dis
149:789-795, 1984
9. Dattwyler RJ, Volkman DJ, Luft BJ, Halperin JJ, Thomas J,
Golightly MG: Seronegative Lyme disease: dissociation of
specific T- and B-lymphocyte responses to Borrelia burgdorferi.
N Engl J Med 319:1441-1446, 1988
10. Grodzicki RL, Steere AC: Comparison of immunoblotting and
indirect enzyme-linked immunosorbent assay using different
antigen preparations for diagnosing early Lyme disease. J Infect
Dis 157:790-797, 1988
11. Russell H, Sampson JS, Schmid GP, Wilkinson HW, Plikaytis
B: Enzyme-linked immunosorbent assay and indirect irnmunofluorescence assay for Lyme disease. J Infect Dis 149:465-470, 1984
12. Magnarelli LA, Anderson JF, Johnson RC: Cross-reactivity in
serological tests for Lyme disease and other spirochetal infections. J Infect Dis 156:183-188, 1987
13. Luger SW, Krauss E: Serologic tests for Lyme disease: interlaboratory variability. Arch Intern Med 150:761-763, 1990
14. Bakken LL, Case KL, Callister SM, Bourdeau NJ, Schell RF:
Performance of 45 laboratories participating in a proficiency
testing program for Lyme disease serology. JAMA 268:891-895,
1992
15. Dressler F, Whalen JA, Reinhardt BN, Steere AC: Western
blotting in the serodiagnosis of Lyme disease. J Infect Dis
167:392-400, I993
Gibney KM, Rose CD, Dubbs SB, Doughty RA:
16. Fawcett IT,
Frequency and specificity of antibodies that crossreact with
Borrelia burgdorferi antigens. J Rheumatol 19582-587, 1992
17. Wilske B, Preac-Mursic V, Schierz G, Kuhbeck R, Barbour A,
Kramer M: Antigenic variability of Borrelia burgdorferi. Ann
N Y Acad Sci 539:126143, 1988
18. Huycke MM, D’Alessio DD, Marx JJ: Prevalence of antibody to
Borrelia burgdorferi by indirect fluorescent antibody assay,
ELISA, and Western immunoblot in healthy adults in Wisconsin and Arizona. J Infect Dis 165:1133-1137, 1992
19. Zoller L, Burkard S, Schafer H: Validity of western immunoblot
band patterns in the serodiagnosis of Lyme borreliosis. J Clin
Microbiol 29:174-182, 1991
20. Pachner AR, Ricalton NS: Western blotting in evaluating Lyme
seropositivity and the utility of a gel densitometric approach.
Neurology 42:2185-2192, 1992
21. Asch ES, Bujak DI, Weiss M, Peterson MGE, Weinstein A:
Lyme disease: an infectious and postinfectious syndrome. J Rheumatol21:454-461, 1994
22. Griner PF, Mayewski RJ, Mushlin AI, Greenland P: Selection
and interpretation of diagnostic tests and procedures. Ann
Intern Med 94553400, 1981
23. Ma B, Christen B , Leung D, Vigo-Pelfrey C: Serodiagnosis of
Lyme borreliosis by Western immunoblot: reactivity of various
significant antibodies against Borrelia burgdorferi. J Clin Microbiol 30:370-376, 1992
24. Girouard L, Laux DC, Jindal S, Nelson DR: Immune recognition of human Hsp60 by Lyme disease patient sera. Microb
Pathog 14:287-297, 1993
25. Craft JE, Fischer DK, Shimamoto GT, Steere AC: Antigens of
Borrelia burgdorferi recognized during Lyme disease: appearance of a new immunoglobulin M response and expansion of the
immunoglobulin G response late in the illness. J Clin Invest
78:934-939, 1986
Документ
Категория
Без категории
Просмотров
1
Размер файла
608 Кб
Теги
band, blot, intensity, application, western, lyme, arthritis, analysis, diagnosis
1/--страниц
Пожаловаться на содержимое документа