CLIN. CHEM. 21/9, 1234-1237 (1975)
Application of Differential Light Scattering to the
Latex Agglutination Assay for Rheumatoid Factor
Philip Biume’ and Leonard J. Greenberg
When a suspension of particles is irradiated with a collimated beam of monochromatic polarized light of a
wavelength close to the particle size and the intensity of
the scattered light is measured as a function of angle,
the scatter intensity is characterized by a series of relative maxima and minima. The nature of the signal depends on several significant variables that are characteristic of the particles. We have constructed a differential light scatter photometer and have applied the tech-
nique to analysis of rheumatoid factor by using latex
particles coated with fraction II gamma-globulin. The results suggest that such a photometer may have potential applications in antigen-antibody assays based on
the use of sensitized
AddItional Keyphrases:
PHOTOOETECTcO
Fig. 1. A schematic depiction of the technique of differential
light scattering
A scatter angle of 00 refers to scattering In the same direction as the Incident beam. Althougl’ a moving photodetector has generally been used In DIS
studies, It Is possible to substitute a cttved piece of photographic film In an
appropriately designed apparatus
(7, 8)
particles.
laser
#{149}
antigen-antibody
reac-
tions
The technique known as differential
light scattering (DLS) involves the measurement
of scatter intensity from a particle or suspension
of particles as a
function of angle (Figure 1). In making such measurements it is essential that the light source be highly monochromatic,
because the parameters
of interest
are wavelength-dependent.
If the particle
size is
much smaller than the wavelength of incident light,
the scatter will not vary significantly
in intensity as a
function of angle, whereas particles much larger than
the wavelength of the light being used cause light to
be scattered in the forward direction, with the intensity decreasing
constantly
as the scatter angle increases. If, however, the diameter of the particle approximates
the wavelength of the incident illumination, a curious phenomenon
occurs. The scatter intensity, while generally more intense in the forward
direction, displays one or more sharply defined minima and maxima as the scatter angle passes from 00 to
greater values. Phillips et al. (1) constructed
a device
that actually allowed study of single particles electrostatically
levitated in a light beam. But if a suspension of many particles is used in place of a single
particle, there is signal averaging and the lobes are
somewhat smeared, although they remain sufficiently
sharp to provide both unequivocal
evidence of their
Department
of Laboratory
Medicine
and Pathology,
of Minnesota,
Minneapolis,
Minn. 55455.
‘Current
address:
Pathology
Department,
Good
Hospital and Medical Center, Portland, Ore. 97210.
Received Mar. 28, 1975; accepted May 12, 1975.
1234
ANGLE
CLINICAL CHEMISTRY, Vol. 21, No. 9, 1975
University
Samaritan
existence as well as a precise determination
of the angular location of both maxima and minima. The signal obtained depends on a number of significant variables, including particle size (relative to wavelength),
uniformity of particle size, concentration
of particles,
refractive indexes of both the particles and the suspending medium, and the structural
characteristics
of the particles. Wyatt has described the application
of DLS to problems in microbiology
(2), especially
the determination
of bacterial antibiotic sensitivities
(3-5).
He has produced
a commercially-available
DLS photometer2
and has shown that a definitive
change in the DLS pattern from an aqueous suspension of bacterial cells will occur within minutes after
an antibiotic to which the organisms are susceptible
has been added. Of course, as Wyatt (2) points out,
bacteria happen to be of a size comparable
with the
wavelength of light, making them especially suitable
for study by means of DLS.
Our interest in this area derives from the fact that
a number of antigen-antibody
reactions are assayed
by agglutination
of sensitized latex particles. With
such a reagent system, we need not rely upon the fortuitous similarity
between wavelength
and particle
size that occurs with bacteria, but might be able to
design reagent particles of the required size and characteristics.
We first looked at a number of commercially available latex-particle
reagents. Of those tried,
all but one failed to produce the desired lobular scatter pattern, largely because of the inhomogeneity
of
the size distribution
of particles or the presence of
aggregates
2
in the
Spectrum,
preparation.
Santa
Barbara,
This,
Calif.
we suspect,
93105.
is a
reversible
motor
drives
the
detector
arm
shaft
via
toothed pulleys and a timing belt. The arm moves
about 3.8 degrees per second. Microswitches
located
on the base limit the excursion of the arm, to prevent
damage.
A precision
potentiometer
(I) connected to
the bottom of the detector arm shaft divides the voltage from a regulated power supply (J) to provide a
position voltage connected
to the x-axis circuits of
both the oscilloscope and recorder. Apart from the
obvious
requirement
components
critical.
Fig. 2. The differential light scatter photometer constructed in
our laboratory
The components of the system are described In the text
situation
that might be corrected were there sufficient reason to do so. On the other hand, the assay of
rheumatoid
factor (anti gamma-globulin)
based upon
the use of sensitized 0.81-am (diameter) latex spheres
gave encouraging results and we chose to explore this
assay more fully as a model for other similar systems.
Materials and Methods
Apparatus
A simple DLS apparatus
was constructed
(Figure
2). The light source (A) is an inexpensive
helium!
neon laser3 operating at 632.8 nm. Because the output of this laser is not polarized, we placed at the output aperture a photographic
polarizer (B) in a holder
that permitted
its rotation.
The specimen
is contained in a standard
15-ml glass centrifuge
tube,
which is positioned by a holder (C) so that the beam
passes through the middle of the lower conical section of the tube. The use of a conical cuvette minimizes the effects of multiple internal reflections (2).
Through a bushing, located in the baseplate directly
below the center of the sample tube, passes a steel
shaft attached
above the base to an aluminum
bar
(D), upon which is mounted
a microscope
focused
upon the intersection
of the laser beam and the center of the sample tube. The microscope consists of a
12-cm tube equipped with a 32 mm, 0.1 NA objective
and a lOX ocular. A circular aperture with a diameter
of 1.6 mm is located in the ocular at the image plane
of the objective. The detector, a 1P21 photomultiplier tube, is located in a housing (E) equipped with a
red filter to block the noncoherent
radiation emitted
from the laser. The photomultiplier
is used in conjunction
with a photomultiplier
microphotometer
(F), and both are commercially-available.4
The photometer output signal was displayed on the y axis of a
conventional
flatbed x-y recorder (G) or a storage oscilloscope (H). Beneath the base of the apparatus,
a
‘ Helium-Neon
Gas Laser, Model
Mountain
View, Calif. 94043.
American
Instrument
Co., Silver
156, 1 mW;
Spring,
Md.
Spectra-Physics,
for the alignment
of the system,
of the various
the dimensions
are not
Procedures
Aliquots of specimens submitted to the clinical immunology
laboratory
for conventional
rheumatoid
factor assay were subjected to analysis with the DLS
system. In our conventional
latex-fixation
analysis we
used commercially-available
reagents5 and a protocol
similar to that described
by Singer and Plotz (6).
Results were obtained after overnight incubation
at
4#{176}C.
The DLS procedure
closely followed the conventional procedure. Buffer and fraction II gamma-globulin solutions were filtered through a O,45-Mm filter
(Millipore Corp., Bedford, Mass. 01730) before use.
The latex/gamma-globulin
reagent mixture was prepared as in the conventional
method. Fifty microliters of serum (heat inactivated
at 56 0 C for 30 mm)
was added to 1 ml of filtered buffer contained
in a
small test tube, followed by 1 ml of the latex/gammaglobulin
reagent
mixture.
The
contents
of the
tube
were mixed by means of a vortex-type
mixer and the
mixture was incubated at 56 #{176}C
for 15 mm, then 250
l of the reaction mixture was added to 2 ml of buffer
contained
in a 15-ml glass centrifuge
tube, which
serves
as the cuvette.
The contents
were mixed
with a
vortex-type
mixer and the tube was scanned in the
DLS photometer.
Each scan was recorded with either
the x-y recorder or storage oscilloscope or both. The
E-vector of the plane-polarized
light source was vertical for all assays.
Resufts
We first evaluated
the performance
of the DLS
photometer
by scanning a suspension of 0.81-tim (diameter) latex particles. It can be seen (Figure 3) that
with the E-vector of the plane-polarized
light source
vertically
oriented,
there are two clearly defined
peaks, at about 50#{176}
and 95#{176},
and a third at about
1350,
which was terminated
by the limited scan of
the instrument;
when the plane of polarization
was
rotated so that the E-vector was horizontal,
these
peaks disappeared.
In both cases the signal is relatively free of noise. Latex particles of different diameters produced strikingly different patterns as would
be expected. For example, Figure 4 illustrates the signals obtained with particles having diameters
of 1.3
and 1.9 zm.
‘
Difco
Laboratories,
Detroit,
Mich. 48232.
CLINICAL CHEMISTRY,
Vol. 21, No. 9, 1975
1235
20
60
90
SCATTER ANGLE.
120
g
0
Fig. 3. The DLS signal obtained from a suspension of 0.81,m latex spheres with plane-polarized light having a wavelength of 632.8 nm
Trace A: E-vector, vertical; trace B: E-vector. horizontal
20
40
60
60
00
20
40
SCATTER ANGLE, dog.
Fig. 5. Typical signal obtained with a specimen negative for
rheumatoid factor
Thisrepresentstwo scansplottedwith an x-y recorder.AlsoshownIs the
the slope of the Initial peak
line crawn to detern*e
I
a
30
60
90
SCATTER ANGLE.
d,Q
120
Fig. 4. Signals obtained from suspension of (A) 1.3- (B)
1.9-
rn diameter latex particles
Trace (B) has been displaced vertically,
20
40
60
60
ioo
20
40
SCATTER ANGLE, do.
to clarify the presentation
Fig. 6. Typical signal obtained with a specimen positive for
rheumatoid factor
When the technique was applied to assay of rheumatoid factor, we found that the reagent preparation
containing
either
no serum
or serum
lacking
rheuma-
toid factor according to the conventional
assay resulted in DLS signals that were essentially the same
as the pure O.8l-Mm latex suspensions
(Figure 5).
Specimens that were strongly positive by the conventional assay produced signals in which the peak at
about 50#{176}
was almost completely obliterated
(Figure
6). Between these two extreme conditions
we observed intermediate
degrees of sharpness of the peak.
While the differences in the patterns were readily observable in qualitative
terms, we sought some means
of quantitatively
assessing them. To do so, we drew a
line through the initial portion of the peak and determined its slope (Figures 5 and 6). When a strongly
positive serum was serially diluted and these slopes
were measured, we found that a plot of the slope vs.
the logarithm of the relative concentration
of rheumatoid factor produced
an approximately
straight
line (Figure 7).
Figure 8 summarizes
results of DLS analyses of 85
specimens; the slope of the initial portion of the first
peak is plotted vs. the rheumatoid
factor titer obtained with the conventional
assay. The data segregate into two modes: specimens with slopes greater
than or less than 0.2. If specimens with slopes greater
than 0.2 are considered to be negative, then most of
the data fall into two groups that are either negative
by both techniques
or positive by both. In generl,
there was a tendency for the slope to vary with the
titer of the positive specimens. Seven specimens clas1236
CLINICAL CHEMISTRY, Vol. 21, No. 9, 1975
sified
as positive
by the
conventional
method
ap-
peared negative by DLS. However, these samples
were all of relatively low titer. Furthermore,
13 specimens that had a positive DLS pattern were classified
as negative by the conventional
assay. Repeat analysis of these samples by the conventional
technique
indicated that 11 of the 13 were negative; the other
two gave titers of 1:80 and 1:160. Although the latex
reagent is generally suitable for DLS analysis, attempts to use reagents that were not freshly prepared
resulted in peaks of insufficient sharpness with negative sera. This problem
also occurred
with some
batches of reagent even when freshly prepared. Brief
sonication of the reagent mixture before serum was
added improved the sharpness of the signal obtained
with negative controls, but strongly positive sera then
failed to produce the profound change in signal that
was observed with other batches of reagent.
Discussion
Differential
light scattering is a technique that can
be performed rather easily and which may find some
interesting
applications
in laboratory
medicine. The
availability
of dependable
lasers that can provide a
light source of the required
intensity,
collimation,
and spectral purity at low cost should make this technique readily available to any who wish to experiment with it. Many assays used in the clinical laboratory are based on systems in which reactions occur on
the surfaces of sensitized particles or cells. We chose
+3
+2
+2
4’
0
+1
w
ILl
0.
0
0
In
-
-j
(I)
-
I
-2
M;
,zo
,
MC
‘
TI T ER
I
2
3
LOG2
RELATIVE
4
5
6
7
8
9
CONCENTRATION
Fig. 7. Effect on slope of the initial peak of 1/10 240 dIlution
of a specimen strongly positive for rheumatoid factor
to investigate the application
of DLS to the rheumatoid factor assay, not because this is the most significant assay of those that might be attempted
but simply because the present reagent system was readily
adaptable
to DLS, owing to the size, homogeneity,
and monodispersion
of the sensitized particles used.
Other reagents for assay systems-such
as that for
hepatitis-associated
antigen-failed
to produce acceptable signals because of the heterogeneity
of the
latex particles.
In the current study, the change in signal is almost
certainly due to rapid microaggregation
of the particles. While many variables contribute to the final signal, any effect that changed all of the particles but
left them homogeneous
and monodisperse
would result in a change in the location or intensity of the
peaks but would leave them sharply defined. The disappearance
of the peak in the region of 50 degrees is
characteristic
of the type of signal expected from a
mixture of particles of different sizes. Each particle
size individually
would produce peaks at different
angles but when mixed would result in an integrated
signal, where all potential
peaks are superimposed
and thus indistinguishable.
Microscopic examination
of such specimens confirmed the presence of microaggregates.
Although for the most part a correlation exists between results of the DLS assay and of the conventional latex fixation assay, a number of seemingly
“false positive” results were obtained by DLS. While
it is difficult to resolve this lack of agreement,
it is
highly probable that the reaction mixtures in question did indeed contain microaggregates
and that the
observed effect was not a measurement
artifact. In
fact, some representative
reaction mixtures were examined
microscopically,
and aggregates
could be
seen.
FIg. 8. Relation between the slope of the initial peak and the
rheumatoid factor titer obtained by means of the conventional
assay for 85 clinical specimens
Arrows Indicate specImens that on repeat analysis by the conventional assay
became positive, as noted In the text
This study suggests that further exploration
of the
application
of DLS to other assay systems is indicated. A first step in this direction would be the design
of particulate
reagents specifically tailored to the requirements
of this approach.
More interesting
than
the detection of microaggregation
would be the development of reagent particles having characteristics
that would be modified sufficiently during the assay
reaction so that a change in DLS signal might be observed in the absence of microaggregation.
Should
this be possible, one could use a reagent containing a
far lower concentration
of particles,
each of which
could react with a much larger proportion of the total
reactant present in the specimen.
Components
of the apparatus
used in this study were obtained
under a grant from the National
Institute
of General Medical Sciences, NIH (1-R01-GM-18020).
The work was supported
in part
by a grant from the Graduate School of the University
of Minnesota.
References
1. Phillips, D. T., Wyatt, P. J., and Berkman, R. M., Measurement
of the Lorenz-Mie
scattering of a single particle: Polystyrene
latex.
J. Colloid Interface Sci. 34, 159 (1970).
2. Wyatt, P. J., Differential
light scattering
techniques
for microbiology. In Methods
in Microbiology,
8, J. R. Norris and D. W.
Ribbons, Eds. Academic Press, New York, N. Y., 1973, p 183.
3. Wyatt, P. J., Automation
biotic susceptibility
testing.
Immunology,
C.-G. Heden
Sons, New York, N. Y., 1975,
of differential
light scattering
for antiIn Automation
in Microbiology and
and T. Illeni, Eds. John Wiley and
p267.
4. Stull, V. R., Clinical laboratory
use of differential
light scattering. 1. Antibiotic
susceptibility
testing. Clin. Chem. 19,883 (1973).
5. Wyatt,
P. J., Light scattering
in the microbial
world.
Inter! ace Sci. 39, 479 (1972).
6. Singer, J. M., and Plotz, C. M., The latex fixation
Med. 21,888(1956).
J. Colloid
test.
Am. J.
7. Koga, S., and Fujita, T., Anomalous
light scattering
by microbial cell suspensions.
J. Gen. AppI. Micro biol. 8, 223 (1962).
8. Brunsting,
A., and Mullaney,
P. F., A light scattering
photometer using photographic
film. Rev. Sci. Inst rum. 43, 1514 (1972).
CUNICAL
CHEMISTRY,
Vol. 21, No.
9, 1975
1237