132
AN, GRATHWOHL, AND FRABLE
8. Kahn LB, Uys CJ, Dale J, Rutherford S: Carcinoma of the breast
with metaplasia to chondrosarcoma: a light and electron microscopic study. Histopathology 1978; 2:93-106
9. Kay S: Light and electron microscopic studies of a malignant cystosarcoma phyllodes featuring stromal cartilage and bone. Am
J Clin Pathol 1971;55:770-776
10. Llombart-Bosch A, Peydro A: Malignant mixed osteogenic tumors
of the breast—an ultrastructural study of two cases. Virchows
Arch A Pathol Anat Histol 1975; 366:1-14
11. McDivitt RW, Stewart FW, Berg J: Tumors of the breast. Atlas of
tumor pathology, Second Series Fascicle 2. Armed Forces Institute
of Pathology, 1967
12. Norris HJ, Taylor HB: Sarcomas and related mesencyhmal tumors
of the breast. Cancer 1968; 22:22-28
13. Oberman HA: Sarcoma of the breast. Cancer 1965; 18:1233-1243
A.J.C.P. • January 1984
14. Ozzello L: Ultrastructure of the human mammary gland. Pathol
Annu 1971;6:1-59
15. Reddick RL, Michelitch H, Levine AM, Triche TJ: Osteogenic
sarcoma—a study of the ultrastructure. Cancer 1980; 45:64-71
16. Smith BH, Taylor HB: The occurence of bone and cartilage in
mammary tumors. Am J Clin Pathol 1969; 51:610-618
17. Stewart FW: Tumor of the breast, Atlas of Tumor Pathology Section
IX Fascicle 34. Armed Forces Institute of Pathology, 1950
18. Sykes JA, Recher L, Jernstrom PH, Whitescarver J: Morphological
investigation of human breast cancer. J Natl Cancer Inst 1968;
40:195-223
19. Williams AH, Schwinn CP, Parker JW: The ultrastructure of osteosarcoma. Cancer 1976; 37:1293-1301
20. Willis RA: The borderland of Embryology and Pathology, second
edition. Washington, DC, Butterworths, 1962, pp 556-560
The Ultrastructural Identification of Auer Body Precursors in a
Case of Acute Promyelocytic Leukemia Using
High-angle Specimen Tilt
BRUCE R. DIXON, B.A., TAPEN M. MUKHERJEE, M.D., AND JAMES Q. K. HO, M.B.B.S. (F.R.A.C.P.)
Auer body (AB) precursors were identified in a case of poorly
differentiated acute promyelocyte leukemia (APL). They consist
of azurophilic granules containing membranous lamellae. In most
granules the lamellae were seen only after high-angle specimen
tilting. In small but more mature ABs, the periodic tubular
structure also was visualized best by specimen tilting. An intermediate granule having both lamellae and tubules is described
and discussed in relation to the fusion of azurophilic granules
to form ABs. The early diagnosis of APL, in the absence of
ABs and intravascular coagulation, is assisted by specimen tilting
to resolve the lamellae in the azurophilic granules. (Key words:
Acute promyelocyte leukemia; Azurophilic granules; Auer bodies; Ultrastructure; Specimen tilting) Am J Clin Pathol 1983;
80: 132-137.
THE PRESENCE of Auer bodies (ABs) in the cytoplasm
of leukocytes currently is considered a specific indicator
of acute nonlymphocytic leukemia. 3 ' 61416 They are seen
most often in acute promyelocytic leukemia (APL)14"16
and acute myeloblasts leukemia (AML), 1416 but they
also have been observed in some cases of acute monocytic
leukemia12 and in a rare case of acute myeloid leukemia
with blast crisis." It is well established, both morphologically and histochemically, that ABs are formed by the
fusion of azurophilic granules 1 " 3,5131416 and that the ma-
Received April 8, 1983; received revised manuscript and accepted for
publication May 30, 1983.
Address reprint requests to Mr. Dixon: Electron Microscope Unit,
Division of Tissue Pathology, Institute of Medical and Veterinary Science,
Frome Road, Adelaide, South Australia, 5000.
Electron Microscope Unit, Division of Tissue Pathology and
Division of Haematology, Institute of Medical and Veterinary
Science, Frome Road, Adelaide, South Australia 5000
ture ABs have differing ultrastructures in the different
types of leukemia.16
We report a case of APL, which, by light microscopy,
had only a few spindle- or splinter-shaped azurophilic
granules recognizable as ABs. Electron microscopic examination, using specimen tilting, allowed the identification of small submicroscopic ABs and azurophilic
granules, which were interpreted as AB precursor granules,
in many of the promyelocytes. The observations also provide a more detailed description of the transition from
azurophilic granule to mature AB than has been published
so far. 2310
Report of a Case
A 21-year-old woman was admitted with pancytopenia when she was
4'/2 months pregnant. The blood picture showed a hemoglobin of 6.9
g/dL, white cell count 0.4 X 109/L and platelets 23 X 109/L. Coagulation
studies for evidence of intravascular coagulation were normal. A bone
marrow aspirate showed abnormal large granulated cells, often with
bilobed and deeply indented nuclei, and moderately abundant cytoplasm
with granulation ranging from sparse, disklike granules to coarse granules
that tended to obscure the nucleus in a number of the cells.
Only a small number of cells were seen to contain Auer rods. The
bone marrow was hypercellular and packed with a monomorphic pop-
Vol. 81 -No. I
133
CASE REPORTS
fgr-
I
•^•OBfe
FIG. 1. A (upper). Large promyelocytes with granular cytoplasm. One cell shows a bilobed nucleus (X2,500).
B (lower). Promyelocytes showing mainly spheric or oval azurophilic granules. (X4,700).
ulation of cells, a diagnosis of acute promyelocyte leukemia (M3, FAB
classification) was made, and this was confirmed by Professor D. A. G.
Galton of the Royal Postgraduate Medical School, London.
Treatment was commenced with a combination of daunorubicin,
cytosine Arabinoside, and thioguanine. Heparin was given as a prophylaxis against the development of intravascular coagulation. The patient
had a stormy clinical course, which included staphylococcal septicemia
and severe interstitial pneumonitis. However, she responded to appropriate antibiotic treatment and a complete remission was achieved. The
patient subsequently was discharged to be readmitted later for consolidation treatment. Since that time she has remained in complete remission
and has delivered a normal, healthy baby.
134
DIXON, MUKHERJEE, AND HO
A.J.C.P. -January 1984
FIG. 2 {upper, left). Azurophilic granules showing an amorphous internal structure. A mitochondrion (M) is also present. Section untilted (X95,000).
FIG. 3 {upper, right). The same field as Figure 2 tilted 30 degrees to display the curved parallel lamellae (X95,000).
FIG. 4 {center, left). An azurophilic granule (A) in which the curved lamellae appear as one continuous spiral. Section tilted 5 degrees (X62,000).
135
CASE REPORTS
Vol. 81 -No. I
FIG. 5 (center, right). An azurophilic granule showing peripheral curved lamellae and a central aggregation of
randomly arranged tubules. Section tilted 15 degrees (X80,000).
FIG. 6 (lower, left). The same field as Figure 5 tilted 5 degrees. The tubules are invisible and the lamellae are indistinct (X80,000).
FIG. 7 (lower, right). A mature Auer body packed with tubules in differing orientations.
Those cut in cross-section show a hexagonal packing. Section untilted (X90.000).
Materials and Methods
Bone marrow aspirate was fixed immediately in 2.5%
glutaraldehyde in 0.05 M cacodylate buffer, postfixed in
2% osmium in the same buffer and then processed for
electron microscopy by conventional technics. Sections
were cut on LKB ultramicrotomes. Semithin sections were
stained with toluidine blue and ultrathin sections with
both uranyl acetate and lead citrate. The ultrastructural
analysis was performed with a JEOL 100CX TEMSCAN®
fitted with a side-entry goniometer stage.
Results
The majority of cells present in the specimen were
promyelocytes, characterized by the presence of azurophilic granules of generally spherical shape (Fig. 1). In
addition, there were spindle-shaped and oval granules,
and occasional small ABs. The granules had an amorphous appearance when viewed normally (Fig. 2), but
tilting the section revealed that many of the granules had
a lamellar substructure (Fig. 3). The lamellae consisted
of a trilaminar membrane 5-7-nm wide coiling within
the granule. Successive layers were separated by a uniform
space, the size of which varied from granule to granule
and was between 3 and 12 nm wide (Fig. 4). Occasional
granules showed a peripheral membrane coil and a central
collection of tubules (Fig. 5). The tubules were in a random
array and were 17 nm in diameter, with a 5 nm central
core. The tubules were seen only within ±5 degrees of
tilt of their clearest orientation (Fig. 6).
Transverse sections of more mature ABs showed the
tubules to be taking up a more hexagonal array, with
a periodicity of approximately 23 nm (Fig. 7). Longitudinal sections of ABs showed periodicities of 11.5 nm
and 19.5 nm (Fig. 8). Altering the tilt by 30 degrees relative
to the long axis of the AB resulted in the position of the
two periodicities being reversed (Fig. 9).
Discussion
The synthetic pathway of primary, or azurophilic,
granule formation during the promyelocytic phase of
neutrophil leukocyte maturation is well documented.3,7
By studying the myeloperoxidase content at each stage
of formation, Bainton and associates3 were able to show
abnormalities in leukemic cells relative to normal cells,
especially a reduction in peroxidase activity in the Golgi
cisternae. This low peroxidase activity in the Golgi has
also been found in AML.4 they suggest3 that the Golgi
•*§**•
FIG. 8 (left). A mature Auer body with the tubules cut longitudinally. Periodicities of 11.5 nm (A)
and 19.5 nm (B). Specimen tilted 15 degrees (X66.000).
FlG. 9 (right). The same field as Figure 8 tilted 30 degrees. The position of the two periodicities seen in Figure 8 has been reversed (X66.000).
136
DIXON, MUKHERJEE, AND HO
FIG. 10. Schematic diagram illustrating one possible mechanism for
Auer body formation. In (a) there is a peripheral coiled lamellae, the
internal edge of which splits off in strips which roll into tubules (b),
which then take up the stable hexagonal arrangement seen in mature
Auer bodies (c).
plays a role in the concentration of the granule contents
in the normal cells but in AML an alternate, as yet unknown, method of condensation occurs in the abnormal
granules. It may well be that these abnormal condensation
products give rise to the curved lamellae seen by ourselves,
and others 5,614 in the granules and that further progress
to the tubules as seen in more mature ABs.4'6'8''5-16 The
possibility that the lamellae are in fact longitudinally sectioned tubules is not tenable, in that the structure is seen
in one orientation only and superimposition of tubule
walls at several different angles of tilt was not observed.
Also, while the lamellar appearance could be observed
frequently, cross-sections of tubules were observed rarely
and then only in a central location.
In optimal orientation, the curved lamellae show a
coiled arrangement similar to the "helical tubular structure, with a space of about 10 nm between each turn of
the coil," which is a description of the internal structure
of a stage II melanosome.7 While the spherical shape of
the granules is a stable configuration, the maintenance
of a spindlelike or rodlike shape involves some internal
or external form of structural support, in this case supplied
internally by the regular array of tubules. Our finding of
A.J.C.P. • January 1984
granules with both a peripheral lamella coil and a few
irregularly spaced tubules suggests that this could be an
intermediate stage in the transformation. The thickness
of the wall of the tubules, approximately 6 nm, is similar
to the width of the lamellae, incicating a possible relationship. Peroxidase staining may have been helpful in
elucidating this point. The change from lamella to tubule
could occur by either the breakdown and reconstitution
of the lamella material or by the internal edge of the coil
splitting off in strips, each of which rolls up into a tubule.
This second possibility is shown diagramatically in Figure
10. It can be regarded as the leukemic extension of Bainton
and co-workers'3 Figure 3, which depicts the maturation
of normal granules. The tubules within an AB show several orientations, which can be interpreted as the aggregation, without alignment, of the tubules from several
fused granules. This would agree with our observation of
tubule formation in small granules before fusion. The
mechanism of tubule formation, by either reconstitution
or splitting of lamella material, is still unclear. The finding
of randomly arranged tubules within granules may thus
indicate the early stages of AB formation as well as indicating degeneration, as has been discussed by BretonGorius and Houssay.4
The periodicity of the hexagonally arranged tubules in
cross-section of mature ABs in our case was approximately
23 nm, which is close to the figure of 25 nm reported by
Breton-Gorius and Houssay.4 Our measured periodicities
in longitudinal section of 11.5 nm and 19.5 nm fit well
with those theoretically expected in a hexagonal array,
i.e., 23/2 nm and (23 sine 60) nm. The finding of a
periodicity of 16 nm by others 416 is as inexplicable as
their formula, (2 X 25)/3 nm, used to predict it.4 That
the tubules seen in longitudinal section are also in a hexagonal array is confirmed by their altered periodicity following a 30-degree change in tilt relative to the long axis
of the AB.
Although these results confirm, and possibly extend,
knowledge of the genesis of ABs, more importantly our
findings suggest the prospect of being able to confirm a
diagnosis of APL at a very early stage in its course. The
importance of recognizing and distinguishing APL from
other forms of leukemia has been reviewed by Jones and
Saleem.9 The use of specimen tilting to locate the whorls
or lamellae in azurophilic granules, when recognizable
ABs are absent or rare, may be a useful aid in confirming
this diagnosis when it is suspected from cytologic or
bleeding patterns.
Acknowledgments. The authors are grateful to Miss R. Williams and
Miss A. Galvin for skillful technical assistance, Miss S. Rath and Miss
L. Hall for printing the micrographs, and Miss D. Burbidge for typing
the manuscript.
CASE REPORTS
Vol. 81 - N o . I
References
1. Ackerman GA: Microscopic and histochemical studies on the Auer
bodies in leukemic cells. Blood 1950; 5:847-863
2. Akasaka K: An electron microscopic study of blood cells. Ultrastructure of acute myelogenous leukemic cells: with particular
reference to Auer bodies. Acta Haematol Jpn 1959; 22:899-915
3. Bainton DF, Freidlander LM, Shohet SB: Abnormalities in granule
formation in acute myelogenous leukemia. Blood 1977; 49:693704
4. Breton-Gorius J, Houssay MD: Auer bodies in acute promyelocyte
leukemia—demonstration of their fine structure and peroxidase
localization. Lab Invest 1973; 28:135-141
5. Freeman JA: Origin of Auer bodies. Blood 1966; 27:499-510
6. Fukushi K, Nakasato N, Narita N, Yoshida K: Electron Microscopic
study of the Auer body. Acta Pathol Jpn 1972; 22:509-515
7. Ghadially FN: Ultrastructural pathology of the cell and matrix.
Second edition. London, Butterworths 1982, p 602
8. Gralnick HR, Tan HK: Acute promyelocyte leukemia—a model
for understanding the role of the malignant cell in homeostasis.
Hum Pathol 1974; 5:661-673
137
9. Jones ME, Saleem A: Acute promyelocytic leukemia—a review of
the literature. Am J Med 1978; 65:673-677
10. Kondo K, Yoshitake J, Takemura K: The fine structure of Auer
bodies. J Electron Microsc 1966; 15:237-248
11. Krause JR, Kapadia SB: Auer rods in chronic myelogenous leukemia
presenting in blast crisis. Am J Clin Pathol 1978; 70:443-444
12. McDuffie NG: Crystalline patterns in Auer bodies and specific granules of human leukocytes. J Microscopie 1967; 6:321-330
13. Mintz U, Djaldetti M, Rozenszajn L, Pinkhas J, de Vries A: Giant
lysosome-like structures in promyelocytic leukemia. Ultrastructural and cytochemical observations. Biomedicine 1973; 19:426—
430
14. Nichols TM: An electron microscope investigation of Auer bodies
in myeloblasts leukemia. Can Med Assoc J 1970; 102:865-867
15. Tan HK, Wages B, Gralnick HR: Ultrastructural studies in acute
promyelocytic leukemia. Blood 1972; 39:628-636
16. Tulliez M, Brenton-Gorius J: Three types of Auer bodies in acute
leukemia—visualization of their protein by negative contrast
after peroxidase cytochemistry. Lab Invest 1979; 41:419-426