[CANCER RESEARCH 41, 11 71-1179,
0008-5472/81
/0041-OOOOS02.00
March 1981 )
Use of Multiparameter
Studies and Scanning Electron Microscopy in the
Interpretation and Attempted Correlation of Surface Morphology with
Cell Type in 135 Cases of Human Leukemias1
Aaron Polliack, Mirón Prokocimer,
Haim Gamliel
Reuven Or, Aviva Korkesh, Rachel Leizerowitz,
Department of Hematology, Hadassah University Hospital, and Hebrew University-Hadassah
ABSTRACT
Multiparameter studies and scanning electron microscopy
(SEM) were performed on cells obtained from 135 cases of
leukemia in an attempt to clarify whether there was a reliable
correlation between surface morphology and cell type as de
fined by cytochemistry, membrane markers, and transmission
electron microscopy. These studies also attempted to deter
mine whether SEM could be used to distinguish lymphoid and
nonlymphoid leukemias, to recognize different types of lym
phoid leukemia, and to define the cell type involved in cases of
unclassified leukemia. The results of this study suggest that
there is a good correlation between surface morphology as
seen by SEM and cell type identified by multiparameter tech
niques. In most cases, nonlymphoid leukemic cells could be
distinguished from lymphoid leukemic cells on the basis of their
surface morphology. SEM did not appear to contribute to the
diagnosis of unclassified leukemia, but more cases of this
nature must be studied. Despite the fact that acute lymphoblastic leukemia cells frequently showed fewer microvilli than
did other lymphoid leukemias, overlap of surface features in
about one-third of the cases did not enable SEM to be used as
a reliable means of distinction. The above conclusions appear
to be supported by preliminary scanning immunoelectron mi
croscopic observations on leukemic cells. It is concluded that
SEM is a useful aid to other modes of microscopy in leukemia
but should not be used on its own to establish diagnosis.
Hannah Ben-Bassat, and
Medical School, Jerusalem, Israel
features of normal circulating lymphocytes, the results of other
SEM studies in leukemia have shown that it is frequently
possible to distinguish lymphoid from other nonlymphoid leu
kemias on the basis of cell surface architecture (11-13, 31,
32, 35, 40). These studies have succeeded in defining the
spectrum of surface morphology seen in a variety of leukemic
cells and demonstrated that certain cell types display a con
sistent surface morphology irrespective of their mode of prep
aration for SEM.
During the past 3 years, we have examined cells from an
additional 135 cases of leukemia which had been accurately
classified, prior to SEM examination, in an attempt to establish
whether this mode of microscopy can be used as a useful
adjunct to other methods of microscopy in the diagnosis of the
leukemic cell type. The present study attempted to answer the
following questions: (a) is there a reliable correlation between
the surface architecture of leukemic cells as seen by SEM and
the cell type as defined by other multiparameter studies? (b)
can SEM be used to distinguish lymphoid and nonlymphoid
leukemias? and (c) can it contribute to definition of the cell type
in AUL and subtypes of ALL?
The results of this study indicate that, in the vast majority of
leukemias studied by SEM, cell surface features served as a
reliable indicator of a given cell type. However, SEM could not
reliably distinguish subtypes of ALL, nor did it contribute to
defining the cell type involved in AUL.
MATERIALS
AND METHODS
INTRODUCTION
In recent years, the SEM2 has been used to study the surface
architecture of a variety of normal and leukemic cells (13, 29,
33, 35, 39, 43, 49). Despite controversial results relating to
the recognition of different lymphocyte types (1, 29, 33, 39,
45) and occasional overlap of surface features, it has become
apparent that it is often possible to distinguish lymphoid cells
from other leukocytes using the SEM (13, 33, 36). Controver
sial findings regarding lymphocyte surface morphology in the
past were in part due to the different preparatory techniques
used (1) and to comparisons of lymphocytes obtained from
different sources (48).
However, while controversy existed concerning the SEM
1 This paper is dedicated
Cells were isolated from the peripheral blood of 135 patients
in whom more than 60% of leukemic cells were present in the
peripheral blood. The following methods were used.
Light Microscopy Morphology and Cell Isolation
All peripheral blood and bone marrow smears were stained
with May-Griinwald-Giemsa.
In cases with high WBC, buffy
coat preparations were prepared from heparinized venous
blood; while in patients with lower counts, cells were concen
trated using the Ficoll-Hypaque gradient centrifugation tech
nique (9). Cells were washed with PBS (contents in stock
solution, 5000 ml: 8.6 g KH2PO4, 48.0 g Na2HPO„,25.0 g
NaCI, 4500 ce distilled H2O, pH 7.4), and the suspension was
adjusted to 4 to 106 x 106 cells/ml with PBS.
to the memory of Professor Gabi Izak, who died
suddenly on March 9, 1980.
2 The abbreviations used are: SEM, scanning electron microscopy; AUL. acute
Cytochemistry
unclassified (undifferentiated)
leukemia; ALL. acute lymphoblastic leukemia;
PAS, periodic acid-Schiff; TEM, transmission electron microscopy; CLL, chronic
lymphocytic leukemia; PLL. prolymphocytic leukemia; PBS, phosphate-buffered
saline.
Received August 25. 1980; accepted November 7. 1980.
The following reactions were determined and examined by
light microscopy: peroxidase, Sudan black, PAS (20), a-naphthyl-acetate esterase with and without fluoride inhibition, AS-D
MARCH 1981
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
1171
A. Polliack et al.
chloroacetate esterase (50), and acid phosphatase (5). Oil red
O (7), /S-glucuronidase (17), and muramidase (24) stains were
not performed in all instances. Except for some cases in which
platelet peroxidase (10) was done, ultrastructural cytochemis
try was not performed.
Biochemical Markers
The following tests were useful in aiding to establish specific
diagnoses in some cases: serum levels of lysozyme (34);
vitamin B12-binding proteins (42); lactic dehydrogenase (27);
and terminal deoxynucleotidyl transferase (21).
Membrane Markers
Classification of Leukemias Prior to SEM Examination
The following resetting tests were performed on buffy coat
suspensions or mononuclear cell suspensions isolated after
density gradient centrifugation: sheep erythrocytes (23); erythrocyte-antibody complement rosettes (22); and erythrocyteantibody (IgG)-ox rosettes (51 ). Surface immunoglobulins were
detected with polyvalent and monovalent antisera conjugated
with fluorescein (26) after cells were preincubated in medium
for at least 2 hr at 37°. Anti-T and anti-common type ALL
antisera (kindly provided by Dr. H. Rodt, Munich, West Ger
many, and Dr. M. Greaves, London, England) were also used
in selected cases.
Transmission Electron Microscopy
In many cases, concentrated cells were washed in PBS and
fixed in suspension with 2% phosphate-buffered
glutaraldehyde. Cells were then washed, postfixed in 2% osmium tetroxide, dehydrated, and embedded in low-viscosity epoxy resin
embedding medium (46). Ultrathin sections stained with uranyl
acetate and lead citrate were viewed with a Phillips electron
microscope. When performed, cytochemistry for platelet per
oxidase was done according to the technique of Breton-Gorius
era/. (10).
Scanning Electron Microscopy
Cells were fixed in suspension with 1% phosphate-buffered
glutaraldehyde for at least 1 hr (pH 7.4, 280 to 330 mosmol).
Thereafter, cells were collected onto poly-L-lysine-coated
coverslips, fixed overnight in 1% glutaraldehyde,
washed
briefly, dehydrated in a graded series of ethanol-Freon 113,
and critical-point dried using Freon 13. Coverslips were then
sputter coated with a thin layer of gold-palladium and stuck
onto aluminum stubs with double-sided sticky tape and silver
paint. Samples were stored under vacuum until examination in
a JEOL x 35 SEM equipped with a LAB-6 gun, at an acceler
ating voltage of 35 to 39 kV. At least 200 cells were counted
in each sample. These cases were classified according to the
dominant cell type present (i.e., more than 60%) and according
to the most frequent type of microprojection encountered (ruf
fles, ridge-like profiles, microvilli, blebs). In cases of lymphoid
leukemia, attempts were made to assess the number of micro
villi present. In some of our later cases, scanning immunoelectron microscopy was performed by a modification of the tech
nique described by Avrameas (3) and Kumon (28) using a
microlatex sphere which had been conjugated to a specific
antibody against a membrane receptor (anti-common type ALL,
anti-T-antigen, anti-surface ¡mmunoglobulin) using glutaralde
hyde. Further details of this technique are provided in more
detail in a separate publication.3
' H. Gamliel and A. Polliack. Positive identification
using scanning immunoelectron
1172
microscopy,
of human leukemic cells
submitted for publication.
The 135 cases were classified into the following categories
(summarized in Table 1).
Nonlymphoblastic
Leukemias
These were diagnosed when cells were peroxidase, ester
ase, Sudan black, or muramidase positive while PAS, paranuclear acid phosphatase, and /8-glucuronidase tests were neg
ative. Serum lysozyme and vitamin B12binders were sometimes
elevated, while lactic dehydrogenase levels were generally
normal or only modestly elevated (27). In general, B- and Tlymphocyte markers were not detected on these cells. In some
cases, TEM contributed when granules were seen and megakaryoblastic leukemia was diagnosed if cells contained platelet
peroxidase in the nuclear membrane and/or endoplasmic reticulum. All cases of chronic granulocytic leukemia examined
were Philadelphia chromosome positive.
Lymphoid Leukemias
ALL. Cells were negative for peroxidase, Sudan black, es
terases, and muramidase but positive, in some cases, for PAS,
paranuclear acid phosphatase, and /8-glucuronidase
and,
rarely, for oil red O. Serum lysozyme and vitamin B12-binding
proteins were usually within normal limits, while in many cases
lactic dehydrogenase levels were markedly elevated (27). fior T-type ALL surface markers were present in some cases.
CLL. CLL was diagnosed by classical clinical criteria and by
the presence of absolute lymphocytosis in the peripheral blood.
All cases were of the B-type and none were T-derived. Cyto
chemistry was usually negative for all stains.
Lymphoma in Leukemic Phase. All cases developed in
patients with non-Hodgkin's lymphoma who developed leuke
mia in the course of their disease. Of the 11 cases, 7 were
diffuse lymphomas and 4 were follicular in type. Seven were
classified as poorly differentiated lymphocytic (5 had B-cell
markers; 2 were T-cell lymphomas), one case was diffuse
"histiocytic"
B-type lymphoma, and 3 cases were mixed lym
phocytic and "histiocytic" B-type lymphomas.
Table 1
Classification
of 135 cases of leukemia using multiparameter
Type of leukemia
studies
No. of cases
Nonlymphoblastic leukemias
Acute myelo- or monocytic (M1-M4)"
Acute monoblastic (M5)a
46
34
Chronic granulocytic
Lymphocytic-lymphoblastic
ALL (L1 -L3)a
6
87
26
6
leukemia
50
11
2
135
CLL
Lymphoma in leukemic phase
Unclassified-"undifferentiated"
leukemias
Total
" FAB classification [Bennett et al. (6)].
CANCER
RESEARCH
VOL. 41
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
SEM and Leukemia
Acute Unclassified or Undifferentiated
Leukemia
When the above-mentioned tests were essentially inconclu
sive or less than 10% enzyme-positive blasts were found on
cytochemistry, cases were classified as AUL. In the 2 cases
encountered, TEM did not contribute; however, ultrastructural
cytochemistry was not performed, nor were terminal deoxynucleotidyl transferase levels available in these 2 patients.
RESULTS
Table 2
Surface markers and morphology
SEM features
typical for mature granulocytes
Mostly non-B non-T or T-derived ALL
T-derived ALL
T-derived ALL
B-derived ALL (Burkitt's
20
leuke
mia)
26
Table 3
SEM and cell origin of lymphoma in leukemic phase
Nonlymphoid Leukemias
Of the 46 proven cases, 44 patients had a majority of cells
displaying ridge-like profiles and ruffles (Figs. 1 to 4). Monoblasts and monocytes had prominent ruffles, while myeloblasts
and maturing myeloid cells had more raised transverse ridgelike profiles. The immature cells [e.g., M1 of the FAB classifi
cation (6)] had relatively smooth surfaces with underdeveloped
microprojections (Fig. 5) which were invariably ridge-like proc
esses. Cells bearing microvilli were seen, but these never
constituted the majority. The 2 remaining cases showed cells
with typical myeloid features, but many displayed large, irreg
ular blebs and pseudopodia of varying size (Fig. 6), which have
until now not been described in acute myeloblastic leukemia.
The latter 2 cases were shown to be predominantly megakaryoblastic by means of cytochemical and ultrastructural studies,
and cells contained the platelet peroxidase. Subsequently, we
have examined yet another case of this type with similar fea
tures under the SEM. All cases of chronic granulocytic leuke
mia had cells with raised well-developed ridge-like profiles,
No. of
cases
Cell origin
Smooth with relatively few micro
villi (Fig. 7)
Smooth with ridge-like profiles
(Fig. 9)
Moderately villous (Fig. 8)
Moderate to markedly villous (Fig.
10)
Total no. of cases
SEM Results and Attempted Correlation between SEM and
Other Multiparameter Studies
in 26 cases of ALL
of
originB-typeT-typeB-typeB-typeT-typeType
lymphomaLymphocytic-lymphoblasticLymphoblast
of
cases5131111Cell
featuresVillous
SEM
surfacesCells
lymphomaMixed
lymphocytic "histio
cytic"Diffuse
"histiocytic"Lymphoblastic
andcells
with villous surfaces
rufflesSmoother
with
ridge-likesurfaces with
profilesTotal
no. of casesNo.
lymphoma
Table 4
SEM features of B-type CLL
ofcases32143150%
ofcases642862100
Predominantly
villousMixed
cellsPredominantly
villous and smoother
smooth (with few micro
villi)Ridge-like
profilesTotal
no. of casesNo.
(Fig. 2).
Lymphoid Leukemias
ALL. In 23 of the 26 cases, cells were found which displayed
relatively smooth but sometimes undulating surfaces (Fig. 7)
with varying numbers of short microvilli (Fig. 8). In 20 cases,
cells with very few microvilli dominated. These cases were all
either T-ALL or non-B non-T-ALL. In another 2 cases of pre-TALL (with no erythrocyte-rosetting
cells or T-cell markers but
paranuclear acid phosphatase), moderate numbers of microvilli
were seen; while in 2 additional cases of T-ALL, spherical cells
showing undulating surfaces predominated, and a proportion
of cells had peripheral ridge-like profiles (Fig. 9). In the only 2
cases of B-ALL, both examples of Burkitt's lymphoma cell
leukemia, were cells with more villous surfaces evident (Fig.
10). Results are summarized in Table 2.
Lymphoma with Leukemic Phase. Five of the 11 cases
showed predominantly villous lymphocytes, and all these pa
tients had B-type disease (Fig. 11). Of the 2 patients with Ttype lymphoblastic lymphoma with leukemia, one had a majority
of moderately villous cells, while the other had a majority of
cells with irregular surfaces and ridge-like profiles (similar to
those seen in the 2 cases of pre-T-ALL described earlier). Of
the 4 remaining cases of lymphoma with leukemic phase in
patients with either mixed lymphocytic and "histiocytic"
lym
phoma (B-type) or diffuse "histiocytic"
lymphoma (B-type), a
majority of cells were of the villous type, but an admixture of
larger cells with well-developed ruffles (presumably "so-called
histiocytes"),
were also seen (Fig. 12). Results are summarized
in Table 3.
CLL (Figs. 13 to 16). All these cases were B-type, and 32 of
the 50 (64%) had more than 60% of cells with villous surfaces
(Fig. 13). A proportion of smoother cells with less microvilli and
others with raised ridge-like profiles were always seen. Four
teen cases (28%) had an even mixture of cells which were
either moderately villous or relatively smooth with few microvilli
(Fig. 14). Three cases (6%) had more than 50% of cells with
very few microvilli (Fig. 15), while one case (2%) showed a
majority of cells with irregular surfaces and some ridge-like
profiles (Fig. 16). Results are summarized in Table 4.
In summary, 83 of the 87 cases of lymphoid leukemia showed
a majority of cells which had surface features associated with
lymphocytes (i.e., microvilli). A smaller number of cases
showed cells which displayed ridge-like profiles, but these
were the dominant cell type in only 4 cases. The overwhelming
majority of null and T-type ALL cells had few microvilli, while
B-cell neoplasias showed a wider spectrum of surface mor
phology ranging from villous to smooth.
AUL. These 2 cases defied more accurate diagnosis, and
unfortunately ultrastructural cytochemistry and terminal deoxynucleotidyl transferase levels were unavailable. Both cases
were probably early myeloblastic [M1 by the FAB classification
(6)] because the cells lacked lymphoid markers and 5 to 8% of
them were Sudan black and/or peroxidase positive. However,
TEM was noncontributory. The SEM diagnosis may also have
MARCH 1981
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
1173
A. Polliack et al.
been compatible with M1-type leukemia because of the pres
ence of some short underdeveloped ridge-like profiles on a
proportion of the cells. However, the majority of cells were
smooth and lacked microprojections,
making final diagnosis
indefinite (Figs. 17 and 18).
Summary of Cases with Difficult Correlation
Of the 135 cases examined, only 8 provided difficulties in
interpretation and attempted correlation. Two cases of acute
myeloblastic leukemia had cells with irregular blebs, but these
cases were later shown to be megakaryoblastic. In 2 cases of
T-ALL, one patient with T-derived lymphoma with leukemic
phase, and one case of B-CLL, cells had undulating surfaces
with raised ridge-like profiles and would probably have been
misdiagnosed as nonlymphoid by SEM alone. In both cases of
AUL, SEM could not contribute to a more definitive diagnosis.
Thus, of the 8 cases, SEM may have been misleading in only
4 cases; in 2 others, it was inconclusive.
Scanning Immune-electron Microscopy
Preliminary results, to be published in detail in a separate
publication,3 tend to support the above-described SEM data.
Positive identification
of cells using immunolabeled latex
spheres confirmed the differences described in surface mor
phology of lymphoid and nonlymphoid leukemias. Represent
ative samples are illustrated in Figs. 19 to 24.
DISCUSSION
The results of the present study confirm the initial impres
sions obtained from a variety of earlier SEM reports on leu
kemic cells which had mostly not been classified by as many
multiparameter studies as in the present study. In over 90% of
the cases examined, there was a good correlation between the
surface morphology, as seen by SEM, and the light microscopic
diagnosis of the respective cell type. Only 8 cases provided
difficulties in correlation; retrospectively, SEM may have been
misleading in only 4 instances, all of which were lymphoid
leukemias. In these cases, the SEM features, considered on
their own, may have led to incorrect diagnoses because of the
presence of ridge-like profiles (features most frequently asso
ciated with nonlymphoid cells) on cells from 4 obvious cases of
lymphocytic leukemia. However, with more experience, it may
indeed be possible to define these cases as lymphoid in the
light of data indicating that a proportion of thymic lymphocytes
may have undulated "ridged" surface profiles (8, 44). In fact,
the latter features may be indicative of a subpopulation of
lymphocytes as suggested by other workers (25). This type of
surface morphology may differ slightly from that recorded for
myeloid cells, which display less undulation and more obviously
raised transverse ridges over the entire cell surface. This
subgroup of leukemias is currently under investigation; hope
fully, it will be possible to distinguish them from nonlymphoid
leukemias with greater ease in the future.
It is worthwhile to stress that none of the cases of proven
nonlymphoid leukemia provided difficulties in classification by
SEM. In the 2 cases which were problematic were cells with
unusual blebs and long pseudopodia; retrospectively, they may
1174
have showed what may be the characteristic features of the
megakaryocytic cell series. SEM findings were disappointingly
inconclusive in the 2 cases of AUL examined, and more cases
of this nature must be studied in the future.
Of the 87 cases of lymphoid leukemia, 83 had cells with the
typical surface features described for lymphocytes by SEM
(13, 16, 33, 41). As in earlier studies (12, 13, 32, 35, 37),
non-B non-T- and T-derived ALL cells characteristically showed
relatively smooth surfaces, while B-lymphoid cells had a
broader spectrum of surface morphology and a wider variation
in the number of microvilli. Recently, it has become evident that
CLL may be a more heterogeneous group of disorders than
originally thought; cases of PLL and CLL with "prolymphocytoid" transformation (14, 18) have until now been included in
most series of CLL studied by SEM. It is possible that the PLL
and transforming CLL cells have surface features different from
those of regular B-type CLL cells; further SEM studies of these
subgroups may provide interesting information on the hetero
geneity of CLL cell surface morphology as seen by SEM.
Earlier SEM studies have suggested that there may well be
major differences of cell surface morphology in CLL and ALL
(13, 16, 33, 41). However, the presence of appreciable num
bers of smoother lymphocytes in close to one-third of the cases
of CLL suggests that SEM cannot consistently distinguish Band T-cell leukemias. Accordingly, SEM should not be used as
a means to establish lymphocyte derivation. Scanning immunoelectron microscopy3 and a study of a larger number of
cultured ALL and lymphoma cell lines tend to confirm the above
conclusions.4 The frequent differences between ALL and CLL
cells noted by SEM may well reflect differences in maturation
(32), with less mature cells showing fewer microvilli than do
more mature cells. This is also true for precursor cells of the
myeloid cell series, which also show fewer microprojections
than do their more mature counterparts (2). On the other hand,
the differences recorded may also reflect variations in recir
culation and homing patterns in these 2 lymphoid populations,
as encountered in the murine system (48). Future SEM and
scanning immunoelectron microscopic studies of PLL, CLL
with "prolymphocytoid"
transformation, and thymic lympho
cyte subpopulations may well provide further important infor
mation in this respect.
Lastly, there is no doubt today that microvilli are extremely
labile structures, modulated by a variety of physiological and
microenvironmental
factors which may result in their rapid
redistribution on the cell surface. Microvilli redistribute in re
sponse to a wide variety of factors such as antigenic stimulation
(30), intercellular contact (29, 38), and spreading (15, 19).
They are also influenced by variations in preparatory tech
niques (17) and alterations in the osmolality of the buffer
vehicle used in the fixative (4). All of these considerations make
it difficult to rely upon the presence or absence of microvilli as
a consistent means of distinguishing lymphocyte subtypes.
Nevertheless, while all of the above is true, it is of interest to
note that other types of microprojections, such as ruffles and
ridge-like profiles, seem to be more stable and are consistently
encountered in nonlymphoid cells irrespective of their mode of
preparation for SEM.
" A. Polliack, H. Gamliel, H. Ben Bassat, A. Korkesh, R. Leizerowitz,
Minowada. Surface morphology and membrane phenotype
leukemia-lymphoma cells, submitted for publication.
CANCER
and J.
of cultured human
RESEARCH
VOL.
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
41
"'W
SEM and Leukemia
REFERENCES
1. Alexander, E. L., and Wetzel, B. Human lymphocytes: similarity of B and T
cells surface morphology. Science (Wash. D.C.), 188: 732-734, 1975.
2. Allen, T. D., and Dexter, T. M. Surface morphology and ultrastructure of
murine granulocytes and monocytes in long term liquid culture. Blood Cells,
2: 591-606, 1976.
3. Avrameas. S. Coupling of enzymes to proteins with glutaraldehyde. Immunochemistry. 6. 43-52. 1969.
4. Barber, T. A., and Burkholder, P. M. Relation of surface and internal
ultrastructure of thymus and bone marrow derived lymphocytes to specimen
preparatory technique. In: O. Johari, (ed.). Proceedings of the Eighth Annual
Scanning Electron Microscopy Symposium, pp. 369-378. Chicago: IITRI,
1975.
5. Barka. T., and Anderson, P. J. Histochemical methods of acid phosphatase
using hexazonium pararosanalin as coupler. J. Histochem. Cytochem.. 70:
741-747. 1962.
6. Bennett, J. M., Catovsky, D., Daniel, M. T., Galton. D. A. G.. Gralnick, H. R.,
and Sultan, C. Proposals for the classification of the acute leukaemias. Br.
J. Haematol., 33. 451-458, 1976.
7. Bennett, J. M., and Dutcher. T. F. The cytochemistry of acute leukemia.
Observations on glycogen and neutral fat in bone marrow aspirate. Blood.
33:341-347,
1969.
8. Bhalla. D. K., and Karnovsky. M. J. Surface morphology of mouse and rat
thymic lymphocytes: an in situ scanning electron microscopic study. Anat.
Ree., 191: 203-219, 1978.
9. Böyum, A. Separation of leukocytes from blood and bone marrow. Scand.
J. Lab. Invest. Suppl., 21: 97, 1968.
10. Breton-Gorius, J., Reyes, F., Vernant, J. P., Tulliez. M., and Dreyfus, B. The
blast crisis of chronic granulocytic leukaemia: Megakaryoblastic nature of
cells as revealed by the presence of platelet-peroxidase—a cytochemical
ultrastructural study. Br. J. Haematol., 39 295-303, 1978.
11. Brynes. R. K., Golomb. H. M., Desser, R. K., Recant, W., Reese. C.. and
Rowley, K. J. Acute monocytic leukemia. Cytologie, histologie, cytochemical
ultrastructural and cytogenetic observations. Am. J. Clin. Pathol., 65: 471481, 1976.
12. Coleman, M. S., Greenwood, M. F., Hutton, J. J., Bollum, S. J.. Lampkin, B.,
and Holland, P. Serial observations on terminal deoxynucleotidyl transferase
activity and lymphoblastic leukemia. Cancer Res., 36: 120-127. 1976.
13. Dantchev, D.. and Belpomme, D. Critical study of the mononuclear leukocyte
morphology based on scanning electron microscopy in normal subjects and
in patients with lymphoid or monocytoid proliferate
disorders. Comparison
with the T or B cell membrane phenotypes. Biomedicine, 27 202-222,
1977.
14. Enno. A., Catovsky, D., O'Brien, M., Cherchi, M., Kumaran, T. O.. and
Gallon, D. A. G. "Prolymphocytoid"
transformation of chronic lymphocytic
leukaemia. Br. J. Haematol., 41: 9-18, 1979.
15. Erickson, C. A., and Trinkaus. J. P. Microvilliand blebs as sources of reserve
surface membrane during cell spreading. Exp. Cell Res., 99 375-384,
1976.
16. Fischer, K., Cohnen, G., Szaniawski, W., and Brittinger. G. Scanning elec
tron microscopic study of leukemic human B lymphocytes. Acta Haematol.,
57: 247-256. 1977.
17. Flandrin. G., and Brouet, J. C. The Sézarycell: cytologie, cytochemical and
immunologie studies. Mayo Clin. Proc.. 49: 575-583, 1974.
18. Galton, D. A. G.. Goldman, J. M., Wiltshaw, E., Catovsky. D., Henry, K., and
Goldenberg, G. J. Prolymphocytic leukaemia. Br. J. Haematol., 27: 7-23,
1974.
19. Haemmerli, G., and Felix, H. Motility of acute leukemia cells: a study by timelapse cinematography and scanning electron microscopy. Blood Cells, 2.
415-426, 1976.
20. Hayhoe, F. G. T., Quagliono, D., and Doll. R. The Cytology and Cytochem
istry of Acute Leukemia, p. 105. London: Her Majesty s Stationery Office,
1964.
21. Hoffbrand. A. V., Ganeshaguru. K.. Janossy, G., Greaves, M. F., Catovsky,
D., and Woodruff, R. K. Diagnostic value of terminal transferase levels and
membrane phenotype in acute leukemia. Lancet, 2: 520-523, 1977.
22. Jondal, M. Characterization of the lymphoproliferative
response to three
different lectins and allogeneic lymphocytes by surface markers. Scand. J
Immunol.. 3: 749-756, 1974.
23. Jondal, M.. Holm, G., and Wigzell, H. Surface markers on Human B and T
lymphocytes. I. A large proportion of lymphocytes forming non-immune
rosettes with sheep red blood cells. J. Exp. Med., 736: 207-215, 1972.
24. Kageoka, T., Nakashima. K.. and Miwa. S. Simultaneous demonstration of
peroxidase and lysozyme activities in leukemic cells. Am. J. Clin. Pathol..
67:482-485,
1977.
25. Kelly, G. E., and Nockholds, C. E. Morphologic differences between subpopulations of human lymphocytes revealed by scanning electron micros
MARCH
copy. Scand. J. Haematol., 79. 172-184, 1977
26. Klein. E., Klein, G., Nadkarni, J. S., Nadkarni. J. J., Wigzell, H., and Clifford,
P. Surface IgM kappa specificity on a Burkitt lymphoma cell in vivo and
derived culture lines. Cancer Res., 28: 1300-1310,
1968.
27. Kornberg, A., and Polliack, A. Serum LDH levels in acute leukemia: marked
elevations in lymphoblastic leukemia. Blood, 56. 351-355, 1980.
28. Kumon, H. Morphologically recognizable markers for scanning immunoelectron microscopy. II. An indirect method using T4 and TMV. Virology, 74: 93103. 1976.
29. Lin, P. S., Cooper, A. C., and Wortis, H. H. Scanning electron microscopy
of human T cell and B cell rosettes. N. Engl. J. Med., 289: 548-551, 1973.
30. Loor, F., and Hagg, L-B. The modulation of microprojections on the lympho
cytic membrane and the redistribution of membrane-bound ligands, a cor
relation. Eur. J. Immunol., 5: 854-865, 1975.
31. Müller-Hermelink, U., and Müller-Hermelink, H. K. Scanning electron micro
scopic investigations of acute leukemia. Virchows Arch. Abt. B Zellpathol.,
23: 227-236. 1977.
32. Newell. D.. Payne, S. V.. Smith, J. L., and Roath, S. Surface morphology of
lymphocyte maturation. In: R. P. Becker and O. Johari, (ed.). Scanning
Electron Microscopy Symposium, Vol. 2, pp. 569-578. SEM Inc., II., USA
1978.
33. Newell, D., Roath, S., and Smith, J. L. The scanning electron microscopy of
normal human peripheral blood lymphocytes. Br. J. Haematol , 32 309316. 1976.
34. Osserman, E. F. Monocytic and myelomonocytic leukaemia with increased
serum and urine lysozyme as a late complication in plasma cell myeloma.
Br. Med. J.. 7: 327-330. 1971.
35. Polliack, A. Normal, Transformed and Leukemic Leukocytes A scanning
Electron Microscopy Atlas, p. 97. Berlin: Springer-Verlag, 1977.
36. Polliack, A. Scanning electron microscopy and the surface morphology of
human leukocytes: current status. Isr. J. Med. Sci., 15: 629-638, 1979
37. Polliack, A., Froimovici, M., Pozzoli, E., and Lambertenghi-Deliliers, G. Acute
lymphoblastic leukemia: a study of 25 cases by scanning electron micros
copy. Blut, 33: 383-395. 1976
38. Polliack, A., Fu, S. M., Douglas, S. D.. Bentwich, Z., Lampen, N , and De
Harven, E. Scanning electron microscopy of human lymphocyte-sheep
erythrocyte rosettes. J. Exp. Med., 740 146-148, 1974.
39. Polliack, A., Lampen, N., Clarkson, B. D.. De Harven, E., Bentwich, Z.,
Siegal. F. P., and Kunkel, H. G. Identification of human B and T lymphocytes
by scanning electron microscopy. J. Exp. Med., 738 607-624, 1973.
40. Polliack, A., McKenzie, S., Gee, T., De Harven. E., Lampen, N., and
Clarkson, E. D. A scanning electron microscopy study of 34 cases of acute
granulocytic, myelomonocytic. monoblastic and histiocytic leukemia. Am. J.
Med., 59: 308-315. 1975.
41. Polliack. A., Siegal. F. P.. Clarkson, B. D., Shu Mar, Fu, Winchester, R.,
Lampen, N., Arce, J., Siegal, M., and De Harven, E. A scanning electron
microscopy and immunological study of 84 cases of lymphocyte leukaemia
and related lympho-proliferative disorders. Scand. J. Haematol., 75: 359377, 1975.
42. Rachmilewitz, D., Polliack, A., and Hershko, C. Acute promyelocytic leukae
mia: a report of 5 cases with a comment on the diagnostic significance of
serum vitamin B,2 determination. Br. J. Haematol., 22: 87-92. 1972.
43. Renau-Piqueras, J. Morphological differences in human peripheral blood
lymphocytes. A freeze-etching and scanning electron microscopic study.
Acta Haematol., 59: 19-24, 1978.
44. Renau-Piqueras, J., and Miragall, F. Surface features of small thymocytes
of rats: a freeze fracture and scanning electron microscopy study. Biomedicine, 29: 232-238, 1978.
45. Reyes, F., Lejonc, J. L., Gourdin, M. R., Mannoni, P., and Dreyfus, B. The
surface morphology of human B lymphocytes as revealed by immunoelectron
microscopy. J. Exp. Med., 747: 392-410, 1975.
46. Spurr. A. R. A low-viscosity epoxy resin embedding medium for electron
microscopy. J. Ultrastruct. Res., 26 31-43. 1969.
47. Tsukada, M., Hanamura, K., and Eguchi, M. Scanning electron microscopic
study of peripheral blood lymphocytes, thymic cells and acute lymphoblastic
leukemic cells in children. Acta Haematol. Jpn., 39: 43-52. 1976.
48. Van Ewijk, W., Brons, N H. C., and Rozing. J. Scanning electron microscopy
of homing and recirculating lymphocyte populations. Cell. Immunol., 79:
245-261, 1975.
49. Wetzel, B., Erickson, B. W.. Jr., and Levis, W. R. The need for positive
identification of leukocytes examined by SEM. In: Proceedings of the Sixth
Annual Scanning Electron Microscopy Symposium, pp. 535-541. Chicago:
ITTRI, 1973.
50. Yam, L. T., Li, C. Y., and Crosby, W. H. Cytochemical identification of
monocytes and granulocytes. Am. J. Clin. Pathol.. 55: 283-286, 1971.
51. Yoshida, T., and Andersson, B. Evidence for a receptor recognizing antigen
complexed immunoglobulin on the surface of activated mouse thymus lym
phocytes. Scand. J. Immunol., /: 401-408, 1972.
1981
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
1175
A. Polliack et al.
Fig 1 Cells from a patient with acute myeloblastic leukemia showing typical ridge-like processes and ruffles, x 8,700.
Fig. 2. Myeloid cells from a patient with chronic granulocytic leukemia with transverse ridge-like profiles, x 7,300.
Fig. 3. Cells from a patient with monoblastic leukemia with prominent surface ruffles, x 6,600.
Fig. 4. Cells from a patient with Mi-type early myeloblastic leukemia showing relatively smooth surfaces with slightly raised transverse ridges, x 6,600.
Fig. 5. Close-up of one of these cells showing very few microprojections resembling early ridge-like profiles, x 11,000.
Fig. 6. Cell from a patient with megakaryoblastic
1176
leukemia showing surface blebs and elongated pseudopodia.
x 9,700.
CANCER
RESEARCH
VOL. 41
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
SEM and Leukemia
Fig. 7. Cells proma I
. surfaces and no microprojections. X 5,700.
Fig. 8. Cells from a patient with null-type ALL with small number of short microvilli. x 6,000.
Fig. 9. Cells from a patient with T-type ALL showing irregular surfaces with a few microprojections. Underdeveloped ridge-like profiles are seen at the cell
periphery. x 6,600.
Fig. 10. B-type ALL cells with varying numbers of short microvilli. x 7,200.
Fig. 11. Leukemic cells from a patient with B-type poorly differentiated lymphocytic lymphoma, with villous surfaces. x 7,600.
Fig. 12. Leukemic cells from a patient with B-type mixed lymphocytic-histiocytic lymphoma. Note upper cell with transverse ridges and ruffles and lower cell with
multiple microvilli. x 6,000.
MARCH 1981
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
1177
A. Polliack et al.
Fig 13. Two typical villous B-type CLL cells, x 9,500.
Fig. 14. B-type CLL cells with varying numbers of microvilli.
x 6.800.
Fig 15. Cells from a case of B-type CLL, in which most cells displayed relatively smooth surfaces, x 5,300.
Fig. 16. Cells from a case of B-type CLL in which most of the cells had irregular surfaces and some ridge-like profiles, x 6,400.
Fig. 17. Cells from a patient with AUL showing relatively smooth undulating surfaces with underdeveloped microprojections, resembling
6.500.
Fig. 18. Two cells from a patient with AUL showing smooth surfaces and some short stubby microvilli.
1178
ridge-like profiles,
x
x 7,800.
CANCER
RESEARCH
VOL. 41
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
@
@;@:-
,@
19
,
20
.
-ON
21
22
@,s8'.
,..e.@,&
Ti'
24
23
-.
.
(I
@1
41
Fig. 19. T-type ALL cells positively labeled with anti T-immunolatex. x 7,300.
Fig. 20. Smooth-surfaced ALL cells unlabeled with anti-B (surface immunoglobulin)-immunolatex. Note the well-preserved discocyte in the center. Small latex
particles are seen between the cells on the substrate. x 3,400.
Fig. 21 . Three B-type lymphoma cells with microvilli. Cells were incubated with anti-T-immunolatex. The latex conjugate is seen on substrate surrounding the
cells. x 3,300.
Fig. 22. B-type lymphoma cells positively labeled with anti-B (surface immunoglobulin)-immunolatex. x 3,300.
Fig. 23. Myeloblasts with transverse ridge-like profiles unlabeled with anti-T-immunolatex. The rounded cell with a few microvilli (lower right) is labeled with the
anti-T-immunolatex and is probably a T-lymphocyte. x 4,100.
Fig. 24. Rounded myeloblasts with relatively smooth surfaces and transverse ridge-like profiles which are unlabeled with anti-B (surface immunoglobulin)
Immunolatex. x 4,500.
1179
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.
Use of Multiparameter Studies and Scanning Electron
Microscopy in the Interpretation and Attempted Correlation of
Surface Morphology with Cell Type in 135 Cases of Human
Leukemias
Aaron Polliack, Miron Prokocimer, Reuven Or, et al.
Cancer Res 1981;41:1171-1179.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/41/3/1171
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1981 American Association for Cancer Research.