Electron Microscopic Observation of Inclusion Bodies in Plasma Cells of Multiple Myeloma and Waldenstrom's Macroglobulinemia

Home , Bence Jones protein, Inclusion bodies

Tohoku J. exp. Med., 1975 , 117, 257-281

KOUSUKE OIKAWA

Department of Internal Medicine,* Tohoku University School of Medicine, Sendai

OIKAWA, K. Electron Microscopic Observation of Inclusion Bodies in Plasma Cells of Multiple Myeloma and Waldenstrom's Macroglobulinemia . Tohoku J. exp. Med., 1975, 117 (3), 257-281 Intracellular inclusion bodies in the plasma cells were sought by electron microscopy in 32 cases of multiple myeloma and 3 of Waldenstrom's macroglobulinemia. In some cases, round shaped intranuclear inclusions and intranuclear fibrillar bundles were observed. In other cases, Russell bodies and crystalline structures were found in the cisternae of rough surfaced endoplasmic reticulum (rER), and dense bodies, myelin-like structures, fibrillar formations, polysome lamellae complexes, crystalline structures, virus-like particles and phagocytosis were observed in the cytoplasm of plasma cells. The detailed ultrastructure of these inclusions was described, and their functional significance, origin and appearance rate were discussed. Finally, the presence of true viral particles in the plasma cells was ruled out. ------ultrastructure of inclusion bodies; multiple myeloma; Waldenstrom's macroglobulinemia

Since the fine structure of the plasma cell was first described by Braunsteiner et al. (1953), many workers have defined the ultrastructure of plasma cell organelles and studied their functions. Furthermore, a variety of intracellular inclusions have been described in the plasma cells of patients with multiple myeloma and Waldenstrom's macroglobulinemia by many authors as reviewed by Maldonado et al. (1966), Waldenstrom (1970), Tanaka and Goodman (1972), Azar and Potter (1973). Little is known, however, about functional significance and origins of these various inclusion bodies; and the studies concerning their appearance rate in plasma cells have not been completed. The purposes of the present study are to describe the ultrastructure of such inclusion bodies, to inquire into their functional significance and origin, to describe their appearance rate and to check if viral particles are truly present in the plasma cells.

Received for publication, August 5, 1975. * Director: Prof. K. Yoshinaga. Present address: The Department of Internal Medicine, Furukawa City Hospital, Furukawa. 257 258 K. Oikawa

MATERIALS AND METHODS The specimensof bone marrowwere obtainedby sternal or iliac puncture from untreated14 patientswith IgG myeloma, 7 with IgA myeloma,1 with IgD myeloma, 10 with BJ myelomaand 3 with Waidenstrom'smacroglobulinemia. They were fixed in 2.5% glutaraldehydewith subsequentfixation in 2% osmiumtetroxide, dehydrated in a graded seriesof acetonand embeddedin Epon mixture. Ultrathinsections were stained with uranyl acetateand lead citrate, and examinedwith a Hitachi7S or a JEM 100Ctype electronmicroscope. In each case,at least 200tumor cellswere observed in thin sections. Whenpartic- ular inclusionswere found, several hundreds of cellswere investigated.

RESULTS Frequency of appearance of various types of inclusions is shown in Table 1. Round shaped inclusions and fibrilar bundles were observed in the nucleus, Russell bodies and crystalline structures were observed in the cisternae of rough surfaced endoplasmic reticulum (rER), and dense bodies, myelin-like structures, fibrillar formations, polysome lamellae complexes, crystalline structures, virus-like particles and phagocytosiswere found in the cytoplasm of plasma cells. The detailed ultrastructure of these inclusions is described below. Round shaped inclusionsin the nucleus. Intranuclear round shaped inclusions were classifiedinto two types. One had low electron density and contained finely granular material. The other contained strongly osmiophilichomogeneous material similar to Russell body. The first type of inclusion of low electron density was observedin 3 patients, 1 with IgA myeloma (in 22% of plasma cells of Case 16)and 2 with macroglubulinemia(in 8% of plasma cells of Case 35 and less than 1% of plasma cells of Case 34) (Fig. 1). These inclusions were round to oval and their number was usually one, but occasionallytwo to five within a singlenucleus. The size of inclusion bodies varied from about 100 nm in diameter to a huge one of over a few ,um occupying most part of the nucleus. The inclusions were surrounded by a single membrane approximately 70 A thick. The second type of inclusion of high electron density was observed in 2 cases, 1 of IgG myeloma (in less than 1% of plasma cells of Case 5) and 1 of IgA myeloma (in 23% of plasma cells of Case 21). In these cases, one or several (up to ten or more) spherules of different sizes were seen within a single nucleus (Fig. 2). Sometimes, similar structures were noted both in perinuclear cisternae and in cisternae of rER in a single cell (Fig. 3). Occasionally,the inclusion contained the materials characteristic of cytoplasmicstructures such as endoplasmicreticulum and ribosomes(Fig. 4). Both of the above two types of inclusions, whether of low electron density or high density, had neither crystalline nor periodic structure within them. Fibrillar bundlesin the nucleus. Bundlesof fibrilsin the nucleus were observed in less than 1% of plasma cells from 3 patients (in Cases 21, 23, 26). Thesefibrils were always found in the interchromatin space and were arranged in a parallel fashion forming a bundle varying in length and about 200 nm in diameter (Fig. Inclusion Bodies in Plasma Cells of M ultiple Myeloma 259 5a and b). The fibrils had a diameter of a pproximately70 A. A single bundle consisted of dozens of fibers . There was no detectable connection between these structures and nuclear bodies ,nucleoli, nuclear envelope or intracytoplasmic fibrillar formation.

Russell body. We observed the Russell bodies in only 3 of 35 cases. In a case of 1gG myeloma(Case 8), the bodies were found in 18% of plasma cells, but in the other 2 cases, 1 of IgA myeloma (Case21) and 1 of BJ myeloma (Case 29), th e bodies were found only in less than 1% of plasma cells. Under the electron microscope, the bodies were found as round and homogeneousmaterial of high electron density within the cisternae of rER (Fig. 6). The sizesof the bodies ranged from about 100 nm to a few microns in diameter. No periodicity could be observed in these structures. Morphologically the bodies were similar to the above-mentioned intranuclear osmiophilic round shaped inclusions. intracisternal crystal. Crystalloid inclusions within the cisternae of rER were observed in 60% of plasma cells of a patient with IgG myeloma (Case 10). In a low magnification, inclusions appeared as many electron dense structures in the shape of needles, spindles or rods (Fig. 7). Under a higher magnification, a longitudinal section showed a marked linearity having periodicity of approximately 100 to 140 A (Fig. 8). In a cross section, geometricallyregular arrangement of tubules of approximately 140 A in diameter, having a central electron lucent core, was observed (Fig. 9). This observation suggested that the inclusions were composed of straight tubular structures measuring 140 A in diameter. Dense body. I have defined the term, dense body, of plasma cell as round to oval osmiophilicstructure within the cytoplasm surroundedby a smooth membrane in order to distinguish it from Russell bodies which are always present within the cisternae of rER. Although the number of dense bodies in plasma cells varied from one case to another, they were found in all 35 cases. In a certain case (Case 18), the bodies were observed in almost all plasma cells. The diameters of the bodies ranged from 0.4 to 1 ,um. Their interior seemed to be dense, faintly granular, and sometimes contained an eccentric dense portion and translucent portion (Fig. 10a), or laminated structures (Fig. 10b). Rarely, intermediate forms between dense bodies and myelin figures were observed (Fig. 10c). Dense bodies were observed near the Golgi area and between the lamellar cisternae; the bodies in the Golgiarea seemed to be lower in density and smallerin size than those in the peripheral cytoplasm (Fig. 11). These observations suggested that the dense bodies were produced in the Golgi area. Myelin-like structure. Myelin-like structures were observed in many plasma cells of approximately half the cases (16 of 35 cases). These structures were present in the cytoplasmic matrix, in mitochondria, in cisternae of rER, in perinuclear cisternae and outside the cells (Fig. 12). Sometimes the picture suggesting that they were being released from the cell was observed (Fig. 13a). 260 K. Oikawa

TABLE I. Types of

Figures in parentheses are percentage. .c., light chain; R.s., round l shaped; F.b., fibrillar bundle; R.b., Russell body; P.c., polysome lamellae complex; V.p., virus-like particle; Ph., phagocytosis. Inclusion Bodies in Plasma , Cells of Multiple Myeloma 261 inclusions observed

Cry., crystal; D.b., dense body; M.s., myelin-like structure; F.f., fibrillar formation; 262 K. Oikawa

Under higher magnifications (40,000 or higher), fine parallel striae consisting of light and dense lines of equal width of about 40 to 50 A, running usually in parallel

to the edges and forming the broad band, were detected in some places (Fig. 13b).

Many of these myelin-like structures were found in the cells considered to be in

necrosis, but they were also found in the cells that had no degenerative change in

other portions.

Fibrillar formation. Cytoplasmic fibrillar formations were observed in 21 of

35 patients. In general, the occurrence of these structures in each case was not very frequent, but occasionally they were noted in as many as 29% of plasma cells

(Case 14). Each fibrillar unit measured from about 50 to 70 A in width. The fibrils were arranged in fascicles and/or in a reticular network. The fibrils were distributed randomly in the cytoplasm with some tendency to concentrate near the

nuclear membrane and to surround the nucleus (Fig. 14). Sometimes the close association of these fibrils with a nuclear membrane and with mitochondria was

noted.

Polysome lamellae complex. Polysome lamellae complexes were recognized in

6 of 35 cases. In a case of IgG myeloma (Case 11), a case of IgA myeloma (Case 21) and 2 cases of BJ myeloma (Cases 23, 26), these inclusions were found in less

than 1% of plasma cells. However, in 2 cases of macroglobulinemia, they were found in 12% (Case 34) and 3% (Case 35) of plasma cells and/or lymphoid cells.

The complexes consisted of lamellar elements with ribosomes or polysomes. The lamellar element was usually of five to ten layers. The profiles of the complex

varied widely from one case to another; they were observed as ellipsoids (Fig. 15a),

parallel lines (Fig. 15b) or circles (Fig. 15c). The diameter of the circular profiles and the distance between the parallel lines were almost equal, measuring approxi-

mately 0.5 ,ƒÊm. Based on this fact, it seemed reasonable to assume that the

polysome lamellae complex is actually composed of a hollow cylinder with a diameter of 0.5ƒÊm. The length of the cylinder was estimated at about 5-7 ƒÊm. The interior of the cylinder seemed to be continuous to the surrounding cytoplasm

and filled with the cytoplasmic matrix including the endoplasmic reticulum and

mitochondria. Occasionally the complexes were seen in close topographical relationship with the rER (Fig. 16).

Intracytoplasmic crystal. Numerous intracytoplasmic crystalline structures

outside of rER were observed in as many as 77% of plasma cells of a patient with BJ myeloma (Case 23). Under a lower magnification, the inclusions appeared as

low to moderately high electron dense structures in the shape of stick, rhombus and polygon. They were usually bounded by a smooth membrane. No ultrastructural relationships between the crystalline structures and other cytoplasmic organelles such as Golgi areas or rER were found. However, they occasionally seemed to be closely related with the fatty granules (Fig. 17). In spite of higher magnification, no periodical structure was recognized within them. Sometimes invaginations of cytoplasm into the crystal were noted (Fig. 18). Inclusion Bodies in Plasma Cells of Multiple Myeloma 263

Virus-like particle. These inclusions were observed in 2 cases of BJ myeloma

(in 5% of plasma cells of Case 25 and less than 1% of plasma cells in Case 24) . R ound shaped inclusions in plasma cell were 0 .5 to 5 ƒÊm in diameter and contained numerous virus-like particles of 40 to 90 nm in diameter . Inclusions and particles were surrounded by a unit membrane which was about 90 A thick and could be resolved into a triple-layered structure . The particles were located along

periphery (Fig. 19), or scattered or closely packed in the inclusions (Fig . 20). The particles showed no internal structures which correspond to nucleoid of true viruses. Buddings cr releasing out of the particles from cell membranes , or mature viral particles such as B-particles and C-particles, were never found.

Phagocytosis. The phagocytic plasma cells were observed in an IgG myeloma

(Case 9) and in an IgA myeloma patient (Case 21). In Case 9, phagocytosis of

red cells and platelets by plasma cells was observed in 2% of plasma cells (Figs . 21,

22). In Case 21, phagocytized red cells were found in less than 1% of plasma cells . There were no morphologic differences between the plasma cells showing phago-

cytosis and those showing no phagocytosis .

DISCUSSION

Round shaped inclusions in the nucleus. Low electron dense inclusions in the

nucleus were reported by Gadrat et al. (1963), Ito et al. (1964) and Smetana et al. (1971). High electron dense inclusions have been reported by Bessis (1961),

Brittin et al. (1963), Maldonado et al. (1966) and Stavem et al. (1974). In addition,

coexistence of both types of inclusions in a single case have also been described by

Kuhn (1967). Furthermore, recently non-electron dense intranuclear inclusions

were reported in a BJ myeloma by Cohen and Lefer (1975). These intranuclear inclusions in plasma cells were studied with the immunofluorescence technique by

several authors (Brittin et al. 1963; Stavem et al. 1974; Cohen and Lefer 1975) and

it has been proved that these inclusions contain immunoglobulins secreted from

myeloma and macroglobulinemia cells. Various types of these intranuclear round shaped inclusions have essentially the same functional significance and represent

the storage of immunoglobulins concentrated in the nucleus.

As to the origin of the inclusions, there have been opposing opinions. One has

been that inclusions may be formed in the nucleus (Brittin et al. 1963; Maldonado et al. 1966; Stavem et al. 1974; Cohen and Lefer 1975), and the other has been

that inclusions are formed in the perinuclear cisternae (Bessis 1961; Gadrat et al.

1963; Ito et al. 1964; Smetana et al. 1971). In the present study, particularly in Case 21, we observed inclusions similar to those within the nucleus in the peri-

nuclear cisternae and in the cisternae of rER simultaneously in the same section.

Therefore, these inclusions possibly originated in the perinuclear cisternae, and were invaginated into the nucleus.

This study showed that these inclusions were rather common in IgA myeloma

and in macroglobulinemia, as supported by several authors (Waldenstrom 1970; 264 K. Oikawa

Tanaka and goodman 1972).

Fibrillar bundles in the nucleus. The natural occurrence of these structures has been reported in neurocytes in various animal species (Siegesmund et al. 1964), pancreatic islet ƒÀ-cells of mice (Boquist 1969), human lymphocytes (Stefani and Tonaki 1970), human glioma cells (Tani et al. 1971), cat neurons (Seite et al. 1971) and in human myeloma plasmacytes (Smetana et al. 1973). Kalnins et al. (1967) suggested that bundles of fibrils were "viral footprints" and served as an indicator of viral etiology in neoplastic cells. Tani et al. (1971), however, have suggested that the structures represent paracrystalline arrangement of nuclear material (possibly

RNA) which may eventually move from the nucleus to the cytoplasm or vice versa. On the other hand, cytochemical studies by Seite et al. (1971) showed that these structures were composed of proteins but neither of RNA nor of DNA. Because the number of cells containing intranuclear fibrillar bundles was very limited, we could not determine their nature or significance. Their true functional significance and origin remain unknown.

We found these structures in less than 1% of plasma cells in only 3 of 35 patients. Stefani and Tonaki (1970) found these structures in the blood lymphocytes from

15 of 30 normal individuals. It is conceivable that the intranuclear fibrillar bundles are much more widespread than they have been thought so far.

Russell body. Russell bodies were originally described by Russell (1890), who regarded them as the characteristic organism of cancer. Various cytochemical reactions have revealed that Russell bodies contain mucopolysaccharides but no

lipids, nucleic acid nor glycogen (Zlotnick et al. 1959; Bangle 1963). The immunofluorescence technique has shown that they contain gammaglobulins

(White 1954; Ortega and Mellors 1957). The Russell bodies have been observed in the electron microscope by Wellensiek (1957), Thiery (1958), Welsh (1960), Movat and Fernando (1962) and Maldonado et al. (1966). It has been revealed that

Russell bodies are made up of accumulated material contained within the cisternae

of rER. It is now accepted that these bodies are the result of condensation, coacervation of the synthesized immunoglobulin in the endoplasmic retculum. In

this respect, the Russell bodies may be analogous to the above-mentioned intranuclear round shaped inclusions.

In the present study, we could find Russell bodies only in 3 of 35 patients. So, the occurrence of these bodies in plasma cells is not so frequent as has been

described in textbooks.

Intracisternal crystal. Since about 35 years ago, needle-like inclusions in

plasma cells have been studied by light microscopy (Rangstrom 1951; Engle and Wallis 1957; Goldberg 1960). But it has not been clear whether these inclusions

exist in cisternae of rER or outside of them. The ultrastructures of crystalloid inclusions within the cisterane of rER have been described by Wellensiek (1957),

Bessis (1961), Movat and Fernando (1962) and Argani and Kipkie (1965) . These Inclusion Bodies in Plasma Cells of Multipl e Myeloma 265 crystals have the same cytochemical and i mmunological reactions as Russell bodies (B essis 1961, 1973). The inclusions presented in this study were composed of straight tubular structures measuring approximately 140 A in diameter. Similar structures have been reported in the lymphocytes of chronic lymph ocytosis (Nardo and Norton 1972). These inclusions were found in only 1 of 35 patients . However, they were f requently seen (as many as 60% of plasma cells) in this case.

Dense body. Sorenson (1964) and Maldonado et al. (1966) assumed that the dense bodies were secretory granules on the ground of their morphologic resem- blance to zymogen granules of the pancreatic acinar cell (Caro and Palade 1964). However, Shigematsu (1969) concluded that they were presumably the primary lysosome since they partially contain an electron dense mass. Suzuki et al. (1970), using a peroxidase-conjugated anti-immunoglobulin antibody method , demonstrated immunoglobulin activity in large cytoplasmic granules of myeloma cells, but failed to demonstrate it in the Golgi complex or in inclusion bodies in the Golgi area . We could find dense bodies in all 35 patients, though the number of them was variable from case to case. Maldonado et al. (1966) also found them practically in all cases. Therefore, the bodies may not be unusual inclusions, but the organelles proper to plasma cells. Ultrastructurally, dense bodies in the Golgi area seemed in the present study to be lower in density and smaller in size than those in the peripheral cytoplasm. So, most of the dense bodies are probably primary lysosomes produced from the Golgi area and may not be engaged in immunoglobulin secretion.

Myelin-like structure. Stoeckenius (Stoeckenius 1959, 1962; Stoeckenius et al. 1960) proposed a hypothetical scheme of the molecular structure of the myelin figure in a series of studies, and assumed that parallel dark lines of myelin figures probably represented the reaction product of OsO4 with double bonds in the fatty acid chains. Buckley (1962) studied myelin forms with the phase microscope, and characterized the effect of mechanical and enzymatic damage on cells of various types. Myelin figures are observed also in the various blood cells of the patients of lipidosis, such as Nieman-Pick disease and Fabry's disease (Lynn and Terry 1964; Tanaka et al. 1965). In these cases, however, the figures are huge and numerous and have more secure structures, and they are considered to be results of dysfunction in lipid metabolism. Azar (1968) described that after treatment with chemotherapeutic agents, lysosomes, phagosomes, dense bodies, and "myelin" formations were frequently observed in myeloma cells. In the present study, myelin figures were found in as many as 18 of 35 patients, and may not be unusual inclusions in plasma cells. Recently it is generally considered that the fixation in glutaraldehyde has a tendency to produce myelin figures. It is my opinion that overestimation of myelin figures in plasma cells as specific pathological indicators may be hazardous. 266 K. Oikawa

Fibrillar formation. Cytoplasmic fibrillar formations in a variety of normal, reactive, and neoplastic cells of blood forming organs have been reported by many authors (de Petris et al. 1962 ; Tanaka 1964 ; Zucker-Franklin and Franklin 1970; Beltran and Stuckey 1970; Ito and Hattori 1974). de Petris et al. (1962) and

Tanaka (1964) suggested that the function of the fibrils is to stabilize the cellular structure at rest. Ito and Hattori (1974) observed the fibrillar formation circum- scribing the nuclear constriction and suggested that they would play a role in

Rieder cell formation in certain types of leukemia. Zucker-Franklin and Franklin

(1970) reported two different types of fibrils in plasma cells of patients with multiple myeloma and amylodiosis. On the basis of immunofluorescent and histochemical technics, they suspected that the fibrils represented intracellular amyloid. Although Beltran and Stuckey (1970) opposed to her hypothesis on the basis of morphologic evidence, her hypothesis seems still very fascinating to us. Therefore, further investigations are required to confirm the relation between the cytoplasmic fibrils and intracellular amyloid.

Cytoplasmic fibrillar formations in the present study were morphologically quite similar to the above-mentioned intranuclear fibrillar bundles. However, no connection was found between these two structures.

Polysome lamellae complex. These complexes have been described under various names, such as granule-lamellae, ribosome-lamellae, and polysome lamellae complex. The complexes have been reported in plant cells (Bartels and

Weier 1967), in cells from the proximal renal tubules of a monkey (Bulger 1968), in adenoma cells from a human adrenal cortex (Hoshino 1969) and in "hairy cells" (Katayama et al. 1973). The ultrastructure of high resolution of polysome lamellae complex was investigated in detail by Hoshino (1969) . Our data of the ultrastructure of these complexes were almost identical with those of Hoshino, except that in the present study the length of the cylinder was 5-7 ƒÊm and several

(five to ten) layers were present. Similar structures was also reported in various cells treated with vinblastin sulfate (VLB) and vincristin sulfate (VCR) (Schochet et al. 1969; Bensch and Malawista 1969; Krishan 1970). Krishan (1970) observed prominent proteinous crystals and large complexes of ribosomes (which were arranged in clusters and helices) associated with fine granular, electron dense material in human leukemic lymphoblasts exposed to VLB. He considered that these ribosomes were involved in the synthesis of the electron dense material which was subsequently organized into prominent crystals.

As for the origin of these inclusions, transitional forms between polysome lamellae complex and microtubules, which had been described by Bensch and

Malawista (1969), were not observed in the present study. The complexes were seen in close topographical relationship with the rER. Therefore it is conceivable that the polysome lamellae complex in this study may have arisen from rER as

Hoshino (1969) has suggested.

These structures were found in 6 of 35 patients in the present study; they were observed in all types of paraproteinemia except for IgD myeloma. It should Inclusion Bodies in Plasma Cellsof Multiple Myeloma 267 be pointed out that in 2 of 3 cases of macroglobuli nemia, the complexes were f ound in more than 3%(12% and 3%) of plasma cells and/or lymphoid cells. The reason why polysome lamellae complexes are found frequently in macroglobulinemia is to be defined.

Intracytoplasmic crystal. Intracytoplasmic crystalline structures outside of rER in the plasma cells were described by several authors (Shigematsu 1969; Fujibayashi and Tanaka 1969; Ito et al. 1970; Suzuki et al. 1970). Most of these investigators regarded the crystalline inclusions as the coacervates of the immunoglobulin secreted f rom Golgi areas. Shigematsu (1969) found a crystalline structure in a part of some lysosomes. However , Suzuki et al. (1970) failed to demonstrate immunoglobulin in crystalline structures using a peroxidase-conjugated anti-immunoglobulin antibody method. In the present study , no ultrastructural relationship between the crystalline structures and other cytoplasmic organelles such as the Golgi area or rER were found except for a relation with probable fatty granules. Therefore, the crystals may be related to lysosomes, but their functional significance is not clear. These inclusions were found in only 1 of 35 patients , their occurrence being frequent (as many as 77% of plasma cells) in this patient . This finding is similar to that of intracisternal crystal .

Virus-like particle. The detailed description on Case 25, in which virus-like particles were found in 5% of plasma cells, was previously reported by us (Oikawa et al. 1973). Similar inclusions of virus-like particles in myeloma cells were first described in 3 cases of multiple myeloma by Sorenson (1961, 1965). He assumed that the particles were viruses since they were morphologically similar to A-particles of murine plasma cell tumor, and suggested the etiologic relationship of virus to multiple myeloma. However, very similar inclusions were reported in cases other than multiple myeloma. They were found in the tumor cells of a patient with lymphoreticulosarcoma (Vasquez et al. 1963), in leukemia cells of a patient with acute lymphocytic leukemia and in a patient with blastic crisis of chronic myelogenous leukemia (Tanaka et al. 1967), and also in lymphatic cells of a patient with acute lymphoblastic leukemia (Sun et al. 1972). Tanaka et al. (1967) showed by histochemical studies that the particles contained acid phosphatase and acid mucopolysaccharide, but no DNA nor RNA. In the present study, the particles superficially resembled A-particles of murine plasma cell tumor (Dalton et al. 1961), but probably they were not viruses, because they did not possess a double membrane and were not as uniform in size as A-particles. In addition, there was neither evidence for buddings or releasing out of the particles from cells, nor mature type of viruses such as B-particles or C-particles. They may be closely related to Golgi vesicles, because in these cases Golgi vesicles were well developed, inclusions existed near Golgi area, and the pictures which suggested transition from Golgi vesicles to the particles were occasionally recognized. Though they were found in 2 of BJ myeloma patients in the present study, 268 K. Oikawa they are not specific to BJ myeloma, since they were described in other types of myeloma by Sorenson (1961, 1965).

Phagocytosis. Bessis (1961) stated that plasma cells were not capable of phagocytosis. Since then it has been believed in general that plasma cells have no phagocytic function. However , Butterworth et al. (1953) demonstrated erythrophagocytosis of plasma cell from a patient with plasma cell leukemia. In addition, Abramson et al. (1970) reported 2 patients with myeloma in whom they observed phagocytosis of erythrocytes, leukocytes and platelets . Fisher and Zawadzki (1970) observed plasma cells containing an engulfed erythrocyte by electron microscopy in several patients with myeloma. In the present study we also found erythrophagocytosis and thrombophagocytosis of plasma cells with the electron microscope in patients with myeloma. It may be certain that some of the plasma cells may have phagocytic function. The possibility that the plasma cell might have phagocytic function was expected previously by several authors (Goodman and Hall 1966; Lerner and Parker 1968). These authors observed hemo- siderin inclusions in plasma cells from patients with various diseases. Recently, most of plasma cells and myeloma cells are considered to be derived from lymphocytes, either by direct transformation or by conversion of the lymphocyte to a blast cell that differentiates into a plasma cell (Cohn 1971; Feld- mann and Basten 1972). However, the possibility that some of the plasma cells may be derived from a reticulum cell can not be completely ruled out, since the phagocytic activity is never found in lymphocytes, but is pronounced in the reticulum cell. Two cases which suggest close relationship between plasma cell and reticulum cell were reported by Okano et al. (1966) and Gach et al. (1971). Under certain circumstances plasma cells may retain or acquire the phagocytic properties of reticulum cell line. The conclusions are as follows: 1. Intranuclear round shaped inclusions are considered to be immunoglobulins which have originated in the perinuclear cisternae . They are more frequently observed in IgA myeloma and macroglobulinemia than other types of myeloma. 2. Intranuclear fibrillar bundles are considered to be much more widespread structures than they have been thought so far. Their origin and functional significance are unknown. 3. Russell bodies are not so frequently seen as they have been described so far. 4. Intracisternal crystals showed composition of very fine straight tubules arranged in parallel fashion. 5. Most of the dense bodies may be primary lysosomes produced by the Golgi area. 6. Myelin-like structures in the plasma cells may be secondary lysosomes yielded physiologically or artefacts produced during fixation. 7. Cytoplasmic fibrillar formations were found in many cases. But the exact significance of these fibrils remains uncertain. Inclusion Bodies in Plasma Cells of Multiple Myeloma 269

8. Polysome lamellae complexes may be precursors of crystals of protein and arise from rER. These inclusions are found frequently in Waldenst rom's macroglobulinemia.

9. Origin and functional significance of intracytoplasmic crystals are unknown . However, they have a close relation probably to fat granules . 10. Virus-like particles which were regarded as viral particles by Sorenson may not be true viruses.

11. The detection of phagocytic plasma cells suggests the presence of some clones of plasma cells derived from reticulum cell line . 12. The structures which could be regarded morphologically as true viruses were never found in plasma cells .

Acknowledgment

I am deeply indebted to Prof. K. Yoshinaga, Dr. S. Onodera, Dr. A. B. Miura and Dr. T. Murata for their advices throughout this study. Mr. S. Kato was of invaluable assistance in the operation of the electron microscope.

References

1) Abramson, N., von Kapff, C. & Ginsburg, A.D. (1970) The phagocytic plasma cells. New Engl. J. Med., 283, 248-250. 2) Argani, I. & Kipkie, G.F. (1965) The cellular origin of macroglobulins. A study of the protein-secreting cells in Waldenstrom's disease. Lab. Invest., 14, 720-728. 3) Azar. H.A. (1968) Diffuse (nonmyelomatous) plasmacytosis with dysproteinemia. Amer. J. din. Path., 50, 302-312. 4) Azar, H.A. & Potter M. (1973) Multiple Myeloma and Related Disorders. Harper & Row, New York, vol. 1, pp. 86-152. 5) Bangle, R. (1963) A morphologic and histochemical study of cytoplasmic Russell bodies in cancer cells. Amer. J. Path., 43, 437-448. 6) Bartels, P.G. & Weier, T.E. (1967) Particle arrangements in proplastids of Triticum vulgare L. seedlings. J. Cell Biol., 33, 243-253.

7) Beltran, G. & Stuckey, W.J. (1970) Nuclear lobulation and cytoplasmic fibrils in leukemic plasma cells. Amer. J. din. Path., 58, 159-164. 8) Bensch, K.G. & Malawista, S.E. (1969) Microtubular crystals in mammalian cells. J. Cell Biol., 40, 95-107. 9) Bassis, M.C. (1961) Ultrastructure of lymphoid and plasma cells in relation to

globulin and antibody formation. Lab. Invest., 10, 1040-1067. 10) Bessis, M. (1973) Living Blood Cells and Their Ultrastructure. Springer-Verlag, Berlin, pp. 519-541, 632-642. 11) Boquist, L. (1969) Intranuclear rods in pancreatic islet ƒÀ-cells. J. Cell Biol., 43, 377-381. 12) Braunsteiner, H., Fellinger, K. & Pakesch, F. (1953) Demonstration of a cyto-

plasmic structure in plasma cells. Blood, 8, 916-922. 13) Brittin, G.M., Tanaka, Y. & Brecher, G. (1963) Intranuclear inclusions in multiple myeloma and macroglobulinemia. Blood, 21, 335-351. 14) Buckley, I.K. (1962) Delayed secondary damage and leucocyte chemotaxis follow- ing focal aseptic heat injury in vivo. Exp. mol. Path., 2, 402-417.

15) Bulger, R.E. (1968) Granule-lamella complex in monkey renal proximal tubular cells. J. Ultrastruct. Res., 24, 150-156. 16) Butterworth, C.E., Frommeyer, W.B. & Riser, W.H. (1953) Erythrophagocytosis in a case of plasma cell leukemia. Blood, 8, 519-523. 270 K. Oikawa

17) Caro L.G. & Palade, G.E. (1964) Protein synthesis, storage, and discharge in the pancreatic exocrine cell. An autoradiographic study. J. Cell Biol., 20, 473-495. 18) Cohen, H.J. & Lefer, L.G. (1975) Intranuclear inclusions in Bence Jones lambda plasma cell myeloma. Blood, 45, 131-139. 19) Cohn, M. (1971) The take-home lesson. Ann. N.Y. Acad. Sci., 190, 529-584. 20) Dalton, A.J., Potter, M. & Merwin, R.M. (1961) Some ultrastructural characteristics of a series of primary and transplanted plasma cell tumor of the mouse. J. nat. Cancer Inst,, 26, 1221-1267. 21) de Petris, S., Karlsbad, G. & Pernis, B. (1962) Filamentous structures in the cyto- plasma of normal mononuclear phagocytes. J. Ultrastruct. Res., 7, 39-55. 22) Engle, R.L. & Wallis, L.A. (1957)Multiple myeloma and the adult Fanconi syndrome. 1. Report of a case with crystal-like deposits in the tumor cells and in the epithelial cells of the kidney. Amer. J. Med., 22, 5-12. 23) Feldmann, M. & Basten, A. (1972) Cell interactions in the immune response in vitro. III. Specific collaboration across a cell impermeable membrane. J. exp. Med., 136, 49-67. 24) Fisher, E.R. & Zawadzki, Z.A. (1970) Ultrastructural features of plasma cells in patients with paraproteinemias. Amer. J. din. Path., 54, 779-789. 25) Fujibayashi, T. & Tanaka, K. (1969) Intracytoplasmic inclusion bodies in the myeloma kidney. J. Electronmicrosc., 18, 190-192. 26) Gach, J., Simar, L. & Salmon, J. (1971) Multiple myeloma without M-type proteinemia. Report of a case with immunologic and ultrastructure studies. Amer. J. Med., 50, 835-844. 27) Gadrat, J., Monnier, J., Izard, J., Bourse, R. & de Boissegon, J.F. (1963) Etude en microscopie electronique d'inclusions intranucleaires dans une plasmocytose medullaire atypique avec dyeglobulinemie. Bull. Cancer, 50, 251-261. 28) Goldberg, A.F. (1960) An usual lymphomatous disease associated with intracytoplasmic crystals in lymphoplasmacytoid cells. Blood, 16, 1693-1707. 29) Goodman, J.R. & Hall, S.G. (1966) Plasma cells containing iron. An electron micrographic study. Blood, 28, 83-93. 30) Hoshino, M. (1969) "Polysome-lamellae complex" in the adenoma cells of the human adrenal cortex. J. Ultrastruct. Res., 27, 205-215. 31) Ito, K., Ito, N. & Katsunuma, H. (1964) A case of myeloma demonstrating an interesting nuclear structure. (Review on nuclear inclusion body in myeloma). Acta haemat. jap., 27, 87-94. 32) Ito, S. & Hattori, A. (1974) Study on fibrillar formation surrounding the nuclear bridge in some type of leukaemic cells. Scand. J. Haemat., 12, 321-328. 33) Ito, S., Goshima, K., Niinomi, M., Horikoshi, N., Nomura, S., Sugiura, K., Yamazaki, K., Hirabayashi, N. & Nishi, Y. (1970) Electron microscopic studies of the crystalline inclusions in the myeloma cells and kidney of K-Bence Jones protein type myeloma. Acta haemat. jap., 33, 598-617. 34) Kalnins, V.I., Stich, H.F., Gregory, G. & Yohn, D.S. (1967) Localization of tumor antigens in adenovirus-12-induced tumor cells and adenovirus-12-infected human and hamstar cells by ferritin-labelled antibodies. Cancer Res., 27, 1874-1886. 35) Katayama, I., Nagy, G.K. & Balogh, K. (1973) Light microscopic identification of the ribosome-lamellae complex in "hairy cells" of leukemic reticuloendotheliosis. Cancer, 32, 843-846. 36) Krishan, A. (1970) Ribosome-granular material complexes in human leukemic lymphoblasts exposed to vinblastine sulfate. J. Ultrastruct. Res., 31, 272-281. 37) Kuhn, C. (1967) Nuclear bodies and intranuclear globulin inclusions in Waldenstrom's macroglobulinemia. Lab. Invest., 17, 404-415. 38) Lerner, R.G. & Parker, J.W. (1968) Dysglobulinemia and iron in plasma cells. Ferrokinetics and electron microscopy. Arch. intern. Med., 121, 284-287. 39) Lynn, R. & Terry, R.D. (1964) Lipid histochemistry and electron microscopy in adult Nieman-Pick disease. Amer. J. Med., 37, 987-994. Inclusion Bodies in Plasma Cells of Multiple Myeloma 271

40) Maldonado, J.E ., Brown, A.L., Bayrd , E.D. & Pease, G.L. (1966) Cytoplasmic and intranuclear electron-dense bodies in the myeloma cell. Arch. Path., 81, 484-500.M 41) Thovat, H.Z. & Fernando , N.V.P. (1962) The fine structure of connective tissue. II. e plasma cell. Exp . mol. Pathol., 1, 535-553. 42) Nardo, J.M. & Norton , W.L. (1972) Chronic lymphocytosis with lymphocyte inclusions. Ann. intern. Med., 76, 265-268. 43) Oikawa, K., Murata , T., Ohtaki, M., Inaba, R., Mikami, M., Sato, I., Endo, K., Onodera, S., Suzuki, T., Niikawa, K., Suzuki, C. & Uchimi, M. (1973) A case of Bence Jones (a) type multiple myeloma with many virus-like particles in the myeloma cells. - Ultrastructural observation - Jap . J. Hemat., 14, 831-837. 44) Okano, H., Azar, H. & Osserman, E. (1966) Plasmacytic reticulum cell sarcoma. Case report with electron microscopic studies. Amer. J. din. Path., 46, 546-555. 45) Ortega, L.G. & Mellors, R.C. (1957) Cellular sites of formation of gamma globulin. J. exp. Med., 106, 627-640. 46) Rangstrom, S. (1951) Plasmacytomatosis with crystalline amyloid deposits in the tumor tissue. Acta path. microbiol. scand., 28, 366-372. 47) Russell, W. (1890) An address on a characteristic organism of cancer. Brit. med. J., 2, 1356-1360. 48) Schochet, S.S., Lampert, P.W. & Earle, K.M. (1969) Oligodendroglial changes induced by intrathecal vincristin sulfate. Exp. Neurol., 23, 113-119. 49) Seite, R., Escaig, J. & Couineau, S. (1971) Microfilaments et microtubulules nucleaires et organisation ultrastructurale des batonnets intranucleaires des neurons sympatiques. J. Ultrastruct. Res., 37, 449-478. 50) Shigematsu, T. (1969) The fine structure of various types of myeloma cells as revealed by electron microscopy. Arch. histol. jap., 30, 375-400. 51) Siegesmund, K.A., Dutta, C.R. & Fox, C.A. (1964) The ultrastructure of the intranuclear rodlet in certain nerve cells. J. Anat., 98, 93-97. 52) Smetana, K., Hermansky, F., Koblizkova, H. & Pospisil, V. (1971) A further note on the ultrastructure of myeloma plasmacytes. Neoplasma, 18, 3-13. 53) Smetana, K., Gyorkey, F., Gyorkey, P. & Busch, H. (1973) Ultrastructural studies on human myeloma plasmacytes. Cancer Res., 33, 2300-2309. 54) Sorenson, G.D. (1961) Electron microscopic observations of viral particles within myeloma cells of man. Exp. Cell Res., 25, 219-221. 55) Sorenson, G.D. (1964) Electron microscopic observations of bone marrow from patients with multiple myeloma. Lab. Invest., 13, 196-213. 56) Sorenson, G.D. (1965) Virus-like particles in myeloma cells of man. Proc. Soc. exp. Biol. Med. (N.Y.), 118, 250-252. 57) Stavem, P., Hovig, T., Froland, S. & Skrede, S. (1974) Immunoglobulin-containing intranuclear inclusions in plasma cells in a case of IgG myeloma. Scand. J. Haemat., 13, 266-275. 58) Stefani, S.S. & Tonaki H. (1970) Fibrillar bundles in the nucleus of blood lymphocytes from leukemic and nonleukemic patients. Blood, 35, 243-249. 59) Stoeckenius, W. (1959) An electron microscope study of myelin figures. J. biophys. biochem. Cytol., 5, 491-500. 60) Stoeckenius, W. (1962) Some electron microscopical observations on liquid-cyrstalline phases in lipid-water systems. J. Cell Biol., 12, 221-229. 61) Stoeckenius, W., Schulman, J.H. & Prince, L.M. (1960) The structure of myelin figures and microemulsions as observed with the electron microscope. Kolloid-Z., 169, 170-180. 62) Sun, C.N., Byrne, G.E. & Pinkerton, H. (1972) Acute lymphoblastic leukemia: Virus-like particles in lymphatic cell inclusions from bone marrow and peripheral blood. Exp. Pathol., 6, 72-79. 63) Suzuki, I., Takahashi, M. & Itoh, S. (1970) Ultrastructural study of human myeloma cells in relation to its function. J. clin. Path., 23, 339-350. 64) Tanaka, Y. (1964) Fibrillar structures in the cells of blood forming organs. J. nat. 272 . K. Oikawa

Cancer Inst., 33, 467-485. 65) Tanaka, Y., Bell, W.R. & Brindley, D. (1967) Pseudoviral inclusion bodies in acute leukemia. A report of two cases. J. nat. Cancer Inst., 38, 629-638. 66) Tanaka, Y., Frost, H. & Spaeth, G.L. (1965) Figures myeliniques dans les cellules spumeuses de la maladie de Fabry. Nouv. Rev. Fr. Hemat., 5, 425-430. 67) Tanaka, Y. & Goodman, J.R. (1972) Electron Microscopy of Human Blood Cells. Harper & Row, New York, pp. 273-332. 68) Tani, E., Ametani, T., Ishijima, Y. Higashi, H. & Fujihara, E. (1971) Intranuclear paracrystalline fibrillar arrays in human glioma cells. Cancer Res., 31, 1210-1217. 69) Thiery, J.P. (1958) Etude sur le plasmacyte en contraste de phase et en microscopie electronique. III. Plasmocytes a corps de Russell et a cristaux. Rev. hemat., 13, 61-78. 70) Vasquez, C., Pavlovsky, A. & Benhard, W. (1963) Lesions nuclearies et inclusions cytoplasmiques particulieres dans deux cas de lymphoreticulosarcomes humains. C.R. Acad. Sci., 256, 2261-2268. 71) Waldenstrom, J. (1970) Diagnosis and Treatment of Multiple Myeloma. Grune & Stratton, New York, pp. 15-28. 72) Wellensiek, H.J. (1957) Zur submikroskopischen Morphologie von Plasmazellen mit Russelschen Korperchen und Eiweibkristallen. Beitr. path. Anat., 118, 173-202. 73) Welsh, R.A. (1960) Electron microscopic localization of Russell bodies in the human plasma cell. Blood, 16, 1307-1312. 74) White, R.G. (1954) Observations on the formation and nature of Russell bodies. Brit. J. exp. Path., 35, 365-376. 75) Zlotnick, A., Gerichter, C.B. & Nir, I. (1959) Experimental production of "grape cells" and their relation to the serum gammaglobulin and seromucoids. Blood, 14, 564-570. 76) Zucker-Franklin, D. & Franklin, E.C. (1970) Intracellular localization of human amyloid by fluorescence and electron microscopy. Amer. J. Path., 59, 23-41.

Fig. 1. Intranuclea r round shaped inclusion of low electron density surrounded by a single membrane (Case 35). •~ 9,800

Fig. 2. Intranuclear round shaped inclusions of high electron density. Several spherules of different sizes are seen within a single nucleus and separated by clear space (Case 21). 10,000 •~

Fig. 3. Dense spherules localized in the nucleus, perinuclear cisternae and cisternae of rER (Case 21). •~ 32,000

Fig. 4. Dense spherules and materials characteristic of cytoplasmic structures found in the nucleus (Case 21). x 19,000

Fig. 5. a: A fibrillar bundle in the interchromatin space of the nucleus (Case 23). •~ 18,000 b: High magnification of Fig. 5a. The bundle consists of dozens of fibers (Case 23). 44,000 •~ Inclusion Bodies in Plasma Cella of Multiple Myeloma 273 274 K . Oikawa

Fig. 6. Russell bodies appearing as round and homogeneous material of high electron density within the cisternae of rER (Case 8). •~ 8,300

Fig. 7. Intracisternal crystals showing the shape of needles , spindles and rods. Low magnification (Case 10). •~ 4,600

Fig. 8. Longitudinal section of the intracisternal crystal showing marked linearity with

periodicity of approximately 100 to 140 A (Case 10). •~ 41,000

Fig. 9. Cross section of the intercisternal crystal showing geometrically regular arrange-

ment of tubules of approximately 140 A in diameter (Case 10) . •~ 114,000 Inclusion Bodies in Plasma Cells of Multiple Myeloma 275 276 K. Oikawa

Fig. 10. a: Dense body containing an eccentric translucent portion (Case 18). •~ 18,000 b: Laminated structure in the dense body (Case 18). •~ 57,000 c: The structure showing an intermediate form between dense body and myelin figure

(Case 18). x 26,000

Fig. 11. Dense bodies in the Golgi area. They are lower in density and smaller in size than those in the peripheral cytoplasm (Case 21). •~ 36,000

Fig. 12. Myelin-like structures in cytoplasm and outside of plasma cell (Case 24). •~ 20,000

Fig. 13. a: Myelin-like structure being released from the cell (Case 24). •~ 26,000 b: Myelin-like structure. It shows fine parallel striae consisting of light and dense lines of equal width of about 40 to 50 A (Case 24). •~ 225,000 Inclusion Bodies in Plasma Cells of Multiple Myeloma 277 278 K. Oikawa

Fig. 14. Cytoplasmic fibrillar formation surrounding the nucleus (Case 14). x 40,000

Fig. 15. a: Polysome lamellae complex appearing ellipsoid in oblique section (Case 34). x 23,000 b: Polysome lamellae complex appearing as parallel lines in longitudinal section (Case 34). x 50,000 c: Polysome Iamellae complex appearing as a circle on cross section (Case 34) . x 43,000

Fig. 16. Polysome lamellae complex in close topographical relationship with the rER (Case 34). x 60,000

Fig. 17. Intracytoplasmic crystals outside of rER with the shape of rhombus and polygon. Crystals seem to have close relations with fatty granules (Case 23). x 18,000

Fig. 18. Intracytoplasmic crystals outside of rER showing no periodical structure within them. Invagination of cytoplasm into the crystal is noted (Case 23). x 50,000 Inclusion Bodies in Plasma Cells of Multiple Myeloma 279 280 K . Oikawa

Fig. 19. Virus-like particles of 40 to 90 nm in diameter seen along the periphery of an inclusion body. A triple-layered membrane is noted (arrow) . The particles show no internal structures which correspond to nucleoid of true viruses (Case 9) . X 71,000

Fig. 20. Virus-like particles closely packed in an inclusion body (Case 9). x 40,000

Fig. 21. A red cell engulfed in a plasma cell (Case 9). x 8,600

Fig. 22. A platelet engulfed in a plasma cell (Case 9). X 9,200 Inclusion Bodies in Plasma Cells of Multiple Myeloma 281