0 1986 Alan R. Liss. Inc.

Cytometry 7101-103 (1986)

TECHNICAL NOTE
A Protocol for Papanicolaou Staining of Cytologic
Specimens Following Flow Analysis
Todd K. Berkan, Jay E. Reeder, Peter A. Lopez, Jr., Kevin M. Gorman, and Leon L. Wheeless, Jr.
Analytical Cytology Unit, Department of Pathology and Laboratory Medicine (T.H.B., J.E.R., L.L.W.), Department of
Biology (K.M.G.), University of Rochester Medical Center, Rochester, New York 14642, and Fox Chase Cancer Center,
Cell Sorter Facility (P.A.L.), Philadelphia, Pennsylvania 19111
Received for publication April 1, 1985; accepted August 7, 1985
A protocol has been developed for restaining
cytologic specimens that have been analyzed
on a multidimensional slit-scan flow system,
The technique involves Papanicolaou stain-
ing of cells on a membrane filter that has
been previously stained with acridine orange
and fixed with glutaraldehyde buffer. The
specimen and staining solutions were sequen-
tially added to a 5-micrometer pore size, 47-
mm diameter Gelman Metricel filter while
it remained in a glass filtration apparatus.
The practice of retaining the filter in the
filtration apparatus throughout the staining
procedure minimizes cell loss and eliminates
specimen cross contamination when com-
pared with conventional filter dip staining.
The availability of this postflow specimen Pa-
panicolaou staining protocol permits accu-
rate determination of the performance char-
acteristics of a multidimensional slit-scan
flow system and should be useful whenever
staining of a limited number of cells with
minimal cell loss is desired.
Key terms: Multidimensional slit-scan, Pa-
panicolaou staining, postflow specimens, ac-
ridine orange, cell loss
A procedure has been developed for evaluating post-
flow analysis aliquots of cytologic specimens analyzed
on a multidimensional slit-scan (MDSS) flow system (1,8).
This procedure is used in assessing the performance of
the flow system in screening cytologic specimens from
the female genital tract and urinary bladder for carci-
noma or its precursors,
The postflow aliquot is collected through the exit port
of the MDSS flow chamber. Because of the number of
cells contained in this aliquot (approximately 50,000
cells) and the large collection volume (30-60 ml distilled
water), the utilization of a technique that results in high
cell recovery is essential. I n addition, the cellular mate-
rial must be presented so that accurate cytological eval-
uation can be accomplished using light microscopy, This
requires that the cells, previously stained with acridine
orange (AO) and fixed in glutaraldehyde, be restained
with conventional Papanicolaou stains. To satisfy these
criteria, a protocol has been developed to Papanicolaou
stain cells on a cellulosic filter without removing the
filter from the filtration apparatus. In this protocol, the
staining solutions and washes are sequentially added
through the filter to restain the cells while the filter
remains in the apparatus.
The use of membrane filters for preparing cytologic
samples with limited cellularity is common in cytopa-
thology laboratories. Typically, a suspension of cells is
filtered onto a &micrometer pore size cellulosic filter
with subsequent fixation and staining of the cells by
dipping the filter through a series of solutions (5). Al-
though useful, this procedure can result in a portion of
the sample being lost to the staining and fixation solu-
tions as well as possible cross contamination of the sam-
ple. Improved cell recovery was reported by maintaining
the filter in the filtration apparatus during fixation and
Papanicolaou staining. However, unsolved technical
staining problems were reported (4).
This report presents a protocol for Papanicolaou stain-
ing cytologic specimens previously stained with A 0 and
fixed in glutaraldehyde buffer. Cell loss and overall
staining characteristics are compared with results ob-
tained by alcohol fixation with subsequent Papanicolaou
staining of cells on a membrane filter.
MATERIALS ANT) METHODS
A total of 20 gynecological samples containing cells
derived from a broad spectrum of abnormality present
in the female genital tract was collected. Both normal
and abnormal samples were obtained by scraping the
uterine ectocervix using a plastic Ayer-type spatula. The
This work was supported by the National Cancer Institute under
Grant R01 CA 33148.
102 BERKAN ET AL.
FIG. 1. Normal cells (left) and abnormal cells (right) previously
stained with AO, fixed with glutaraldehyde and subsequently Papani-
colaou stained.
cells were suspended in a mucus-dissolving solution
(Mucosol (R), Lerner Labs). Each sample was syringed
for cell dispersal and divided into two equal aliquots.
Syringing was accomplished by a pneumatically driven
automatic syringer (6,7).
A 5-micrometer pore size, 47-mm diameter Gelman
Metricel filter (Gelman Sciences f60003) was pre-
pared for each aliquot by expanding the filter in 95%
ethanol for 30 s in a glass petri dish. Each filter was
then clamped into an analytical filter holder (Millipore
Corporation #XX10047-30), and secured on a filter man-
ifold. Low vacuum (2-5 cm Hg) was applied to the filter
as it was hydrated.
Papanicolaou Staining and Evaluation
The first specimen aliquot of ten samples was pro-
cessed according to a poststain fixation protocol devel-
oped for preparing gynecological specimens for analysis
on the X-Y-Z mutidimensional slit-scan flow system (2,3).
Following low-speed centrifugation (2,000 r pd20 min),
the Mucosol was removed and the cells stained in sus-
pension using 10 cc of 0.01% A0 solution, washed, and
resuspended in 10 cc of Millonigs glutaraldehyde buffer
(3.8% aq.). The sample was refrigerated for 24 hr. Fol-
lowing low-speed centrifugation, the glutaraldehyde was
removed and the cells washed twice with 10 cc phos-
phate bufTered saline and stored under refrigeration.
The suspension of cells was diluted with 50 cc of distilled
water and the cells placed on a hydrated (previously
alcohol expanded) Gelman filter.
The second specimen aliquot was transferred to a filter
holder containing an alcohol expanded filter. Mucosol
was removed by washing the cells with water through
the filter while applying low vacuum. The cells were
then step dehydrated to 95% ethanol and fixed by allow-
ing them to stand in 95% ethanol for 15 min. Finally,
cells were step hydrated, and the filter was kept moist
with water.
The Papanicolaou staining procedure for this study is
a modification of the routine dip staining protocol used
PAPANICOLAOU STAINING POSTFLOW SPECIMENS 103
by the Cytopathology Laboratory at the University of
Rochester Medical Center. Both specimen aliquots were
handled in a similar manner for Papanicolaou staining.
Harris Alum hematoxylin (Harleco #638) staining so-
lution (filtered two times through Whatman #2 filter
paper) was added to cover the filter and remained in
contact with the cells for 4 min. The hematoxylin was
removed by washing with water, and the filter flooded
with 0.025% (aq.) hydrochloric acid solution, and subse-
quently washed with water. The filter was then flooded
with 0.01% (aq.) ammonium hydroxide solution and im-
mediately washed with water. Following step dehydra-
tion to 95% ethanol, Orange G (OG-6, Harleco #7052X)
stain was added to the filter and allowed to stand for 1.5
min, before washing with 95% ethanol. Eosin (EA-65,
Harleco #7054X) stain was added to the filter and al-
lowed to stand for 3.5 min. The sample was then washed
with 95% ethanol, 99% ethanol, and finally xylene
(Fisher X-5). The xylene acted to clear the filter, mak-
ing it transparent. The filter was trimmed to 38 mm
diameter (inside diameter of the filter funnel) and placed
cell-side-up on a 75 x 50 mm glass slide (Fisher #12-
550Cf in Eukitt mounting media (0. Kindler, West Ger-
many), and covered using a 45 x 50 mm coverslip (Fisher
#12-545H). No cells were observed on the portion of the
filter that had been removed.
Cell Recovery
To document cell recovery rates of the reported proto-
col, ten additional gynecologic samples were divided into
two equal aliquots following syringing. One specimen
aliquot was ethanol fixed and Papanicolaou stained
while remaining in the filtration apparatus. The other
specimen aliquot was ethanol fixed on a filter, removed
from the apparatus, and conventionally dip stained ac-
cording to the Papanicolaou technique. Following stain-
ing, both specimen aliquots were mounted on glass slides
and coverslipped. Cells in 50 random fields (as viewed
at 200~) were counted on each slide, to evaluate cell
loss.
RESULTS AND DISCUSSION
Samples that were Papanicolaou stained following A0
poststain fixation were compared with alcohol fixed Pa-
panicolaou-stained samples. Cells from both aliquots
were assessed for overall staining quality, cytoplasmic
and nuclear detail, and general ease of diagnosis. Nu.
clear morphology and cytoplasmic detail was very crisp,
and nuclear hyperchromasia was evident in abnormal
cells. Eosinophilic staining was evident in superficial
squamous cells, and cyanophilic staining was noted in
intermediate squamous and squamous metaplastic cells.
Polymorphonuclear leukocytes were characteristically
hyperchromatically stained, and Doderlein bacillis were
clearly discernible in several squamous cells. Nuclear
and cytoplasmic Papanicolaou staining was slightly re-
duced in the previously AO-stained glutaraldehyde-fixed
cells. Despite the possible exclusion of some binding of
Papanicolaou stains by the AO, glutaraldehyde buffer
poststain fixation protocol, sufficient stain was bound to
allow visualization of cytoplasmic and nuclear morphol-
ogy, and the stain permitted accurate cytologic evalua-
tion. Figure 1 illustrates the staining quality present in
normal and abnormal cells. These cells were previously
stained with A0 fixed in glutaraldehyde, and subse-
quently Papanicolaou stained.
Papanicolaou staining of cells with the filter main-
tained in the filtration apparatus greatly increased cell
recovery as compared to conventional dip staining. An
average 13-fold increase in number of cells was ob-
served. Cell loss appeared to be randomly distributed
over all cell types (leukocytes, normal epithelial cells,
and abnormal epithelial cells).
This protocol is applicable in situations where limited
numbers of cells are collected in large fluid volumes,
and where cell loss and specimen cross contamination
must be minimal. Evaluation of the postflow aliquots
from the X-Y-Z multidimensional slit-scan flow sytem is
only one application. This technique should prove useful
in other areas in which high sample recovery and ade-
quate Papanicolaou staining are required.
LITERATURE CITED
1. Cambier J L, Kay DB, Wheeless LL: A muitidimensional flow sys-
tem. J Histochem Cytochem 27:321, 1979.
2. Cambier MA, Wheeless LL, Patten SF: A new post-staining fixation
tcchnique for Acridine Orange. A&a Cytol (Baltimore) 21:477, 1977.
3. Cambier MA, Wheeless LL, Patten SF: A post-staining fixation
technique for Acridine Orange: Quantitative aspects. Anal Quant
Cytol 1:57, 1979.
4. Frost J L, Gill GW, Hankins AG, LaCorte FJ , Miller RA, Hollander
DH: Cytology filter preparations: Factors affecting their quality for
study of circulating cancer cells in the blood. Acta Cytol (Baltimore)
11:363, 1967.
5. Gelman Sciences: Diagnostic Cytology by Membrane Filter. Appli-
cation Bulletin 100R, 1982.
6. Lopez PA, Cambier MA, Wheeless LL: Syringing as a method of
cell dispersal. 11.Effects on abnormal cells. Anal Quant Cytol3:235,
1981.
7. Mead JS, Horan PK, Wheeless LL: Syringing as a mcthod of cell
dispersal. I. Effects on intermediate and superficial squamous cclls.
Acta Cytol (Baltimore) 22:86, 1978.
8. Wheeless LL, Patten SF, Berkan TK, Brooks CL, Gorman KM, Lesh
SR, Lopez PA, WoodJ CS Multidimensional slit-scan prescreening
system: Preliminary results of a single blind clinical study. Cyto-
metry 5:1, 1984.