`Vol. 52, :-.lo. :l, July 1972
`I'rinted in U.S.A.
`
`Immunoglobulin-Producing Tumors
`and Myeloma Proteins of Mice
`
`MICHAEL POTTER
`
`Laboratory of Ceil Biology, National Cancer Institute, National I nstitutes of H ealth,
`B ethesda, Maryland
`
`I ntroduction . _ . _ .. __ . . _ . _ . __ . _ ............. .. . . . .... . .... . . . __ . _____ .... .
`Abbreviations . __ . . . . ..... . ..... . ..... . ... _ . . .. . _____ . _ . .. . . . . . ....... _ .
`Plasmacyton1as .... .. ............... .. .... . . _ . .... _ .
`_. _ . ........ . . .
`I nduction of plasm acytomas .. . .. . . . .. . _ . .. .. .. . _. _ _ _ _ _ _ _ _ ... . . . .. . . . . . _.
`F actors influencing plasmacytoma induction and incidence_ . . . .............. .
`I mmunoglobulins in p lasmacytoma development . ... . . _. _. _ ..... . . _ ... .. .. . .
`V iruses: intracisternal viruslike particles, viral antigens _ . .. _ .. _ .. . __ .. _ . _ . .. . .
`Karyological changes .. . .. . ... . __ .... ........ ... . . _ .. .. ....... - - . . . ... . .
`Spontaneous ileocecal plasmacytomas . __ . .. .. _ .. .. .. . .. _ ... . . . . ... - .. . ... _ .
`Comments on plasmacytoma development .......... . . . .. . . .... __ .. . .. _ ... .
`Tumors of Immature Cells in I mmunocyte Series ..... _ ..... _ . . _
`... .. .... . .
`N onsecreting tumors . . . . . _ . . . . . .. _ . . _ . _ . _ . . .. _ _ _
`_ . . . . . _ . _ -
`Secreting tumors _ .. _ . _ . _ . . . . . . . . . . . . . . .
`____ . . _ ... . .
`Comments ... . _ _ . _ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`_ ... . ..... .
`Propagation and Growth of Plasm acytomas . . . . . . _ . . . . .
`. . .
`_ ... . . . ... .
`Transplantation .. ..... . ... . ........ . .. . . _ ...... __ .. _ . .
`_ .. . ... . . .
`Colony formation in spleen .. .. . .. ... _ . . . . .
`. . . . . . . . .
`Preservation of plasmacytomas by freezing _ .. . . . .... _ . _
`.. . . . . . . . . .. _ .
`Tissue culture.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`.
`Commonly used transplantable plasm acytomas ...... _ .. . . . . . . . . . . _ . .. ... . . .
`Chemotherapy .. _ . . ........ _ . . . . . . . . .
`_ . . . . _ . _
`_ . __ .. . .. .
`Pa thological and Pathoph ysiological Changes Associated with Plasmacytornas _ _ . . .
`Bone lesions ....... . .. .. _ .. . .. _ ..... . .. _ . _ .
`Renal lesions_ . . ... ..... _ .. . . .... . . . . . .. . ... . .. .. . . . . ... . . . . _ . . . .. .. . .. .
`Bence Jones proteinuria . ..... .. . . . .. ...... . .
`.\myloidosis . ___ . __ .. _ ...... ........... . -
`Effects on indut:ed antibody formation, IgG catabolism
`. .\ntigens Associated with Plasma Cells and Plas ma Cell Tumors.
`Diffcrentiatiou anligeus _
`...... ...... . ... . ... . . . . . . .. . . . . .
`I mmunoglobulin _ .. .. _ .
`. ... . . . . . .. ... . . ... . . . .. ........ . .. . . . . . . .
`Tumor-specific antigens . _ . ... . ....... _ . . . _ .. _. _ ...... __ .
`. .. . ... .
`Comments
`. . . . . . . . . . . . . . . . . . . . . . . . . . . .
`. - . .. . . .
`·Myeloma Proteins: Structure .. . . . . . .
`. _ . _ . _
`_ . _ . . . _ . .
`. .. . .... . .
`Two-chain IgA .......... __ . . . . .
`. . . . .
`. . . . . . . . . .. ... . . . _ . .. . ... . . . . .. . .
`IgC . .............. .. - -. -- · · · · · · · · · · · · · - · · · - - · · · · · · · · - · - · · · · -· · · · · · · · ·
`l mmunoglobulin genes .. . . . ... . . . . .. . . . . .. _
`c,,..... ......
`. . .. . ... -
`Vr,CV1., V,) .
`__ ...... .. . . . .. . .. . . ..... .
`CH . ... - - . - . - .
`. .. . . .. .. .. . - . .. -· - -
`V u . -
`Cornrncnt.~ .. _ . . _
`
`632
`634
`636
`636
`640
`644
`64G
`648
`649
`649
`650
`651
`651
`659
`661
`661
`662
`663
`663
`664-
`665
`666
`666
`667
`667
`667
`668
`668
`668
`670
`670
`671
`671
`674
`675
`67fi
`677
`678
`681
`684
`684
`
`63 1
`
`1 of 89
`
`BI Exhibit 1110
`
`
`
`632
`
`MICHAEL POTTER
`
`Volume 52
`
`_. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 684
`Antigen-binding mydoma proteins . .
`Nitrophenyl. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 686
`:1-Acetyluracil, D NA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689
`Phosphorylcholine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 690
`ti 1 --> 6-linked galactoscs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 692
`,3-D-N -acetylglucosa minc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 693
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 693
`t 3 d extrans . . . . . . . . .
`a 1 -
`693
`,V-acetyl-o-mannosamine
`694
`Other antigens .. ... . .. .
`694
`. .... ..... .. . . . . .... .... .. .. ... . .. ... . . . .
`Comments...... ... ... ... .
`695
`I mmunuglubulin Biosynthc.5is . ... . . .. . .. . . . . . . . . . . . . .. ..... . . .. . . . . . .. . . .. .
`. .. .. . . . _ .... .
`695
`G eneral descriptio n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`698
`I mmunoglobulin mRN :\ studies: light and heavy polysomes .. . .. . . . . .. . .. . .. .
`Biologic studies on origin of d efective imm unoglobulin-producing cells . . .. ... . . .
`70 1
`Cell-free synthesis of immunoglobulin hybrid ization studies .... . . .. .. .. . . . .. . .
`703
`. .. . .... ......... . . .... .. . . .. .. .
`704
`. .
`Nucleic-acid hybridization studies. .
`705
`. . . ... .. . . . . . . . . .. . ... . . ... . . ... . . ... .
`Transfer RN A . . . . .
`70fi
`Biosynthesis of carbohydrate side chains ... ..... ... . _ . .. .. ... .... . .. .. .
`707
`Con11ncnts ... . . .. . . . .. . ..... ...... . . .. .... . . .. . .. . . . . ... ... .... .. . . . .. .
`708
`Su1nn'lary .. . . . . . . . ... .. .. . .. .. . .. . _ . . ... . . .. . . . . . . .. . . . .. .. . . . ... . . ... . .
`
`I. 1:-\TRODUCTION
`
`The tumors in mice thus far associated with immunoglobulin production are
`the plasmacytomas and lymphomas. These represent neoplastic derivatives of
`various cell types within the immunoeyte system of cells; the immunoglobulin
`produced may be secreted or bound to the cell surface or both (Table I).
`The most widely studied immunoglobulin-producing tumors in mice are the
`plasmacytomas. These tumors appear to arise from cells specialized in such a way
`that they can produce only a single molecular type of immunoglobulin. It is gen(cid:173)
`arally accepted that normal plasma cells are similarly specialized and that the tumors
`by their great proliferative potential are amplifications of ind ividual cell types
`within the immune system. T he plasmacytomas thus arc an extremely useful means
`for studying the individual components in the vastly heterogeneous immunocyte
`system of cells.
`l''tasmacytornas a-re D1 fil\)ne>l'.~l)D.'d\ \)t\~\n. ~\'l \'u<e t ·21.15.\'5. ~{ \h<: h~m~i<:.~<:.'1\!.'5.
`immunoglobulin that each turnor produces. Nearly every plasmacytoma cell of
`independent origin is restricted to producing only a single species of immunoglobu(cid:173)
`lin molecule, i.e., molecules that all contain the same types of light and heavy
`polypeptide chains. l t cannot be stated categorically that each tumor or normal
`immunocyte produces only a single type of immunoglobulin molecule because
`there arc now a few exceptions among both normal ( 184, 201 ) and tumor (20 1,
`207, 29 1) immunocytcs where at least two different classes of immunoglobulins are
`produced. These special cases are discussed in the sections on tumors of immature
`plasma cells and imrnunoglobulin synthesis. While the greatest majority of plas(cid:173)
`mac:ytornas are indeed monoclonal when appraised by immunoglobulin produc(cid:173)
`tion, they nonetheless themselves originate 'vvithin normal clones. The abnormal
`proliferative property probably develops only within a few cells of a clone.
`
`2 of 89
`
`BI Exhibit 1110
`
`
`
`July 1972
`
`MOUSE TUMORS AND MYELOMA PROTEINS
`
`633
`
`TABLE 1. lmmunoglobulin-producing tumors in mice
`Normal lmmunocyte from I
`Immunoglobulin Production
`which Tumor Originates - - - - - - - - - - - - - - - -
`!
`i __ + ______ I
`Secretory
`Cell surface
`+*
`?
`
`R lymphocyte
`T lymphocyte
`I mmunoblast
`
`Plasma cell
`Lymphoplasmacyte
`
`+§
`?
`
`Pathological Designation of Tumor
`
`- - - - - · - - · - - -
`Lymphocytic neoplasm
`Lymphocyte neoplasmt
`Plasma cell leukemia, reticu-
`lum cell neoplasm type B t
`Plasmacytoma (myeloma)
`Lymphoma
`
`+
`+
`+
`:R lymphocyte = bone marrow-derived lymphocyte; T lymphocyte = thymus-derived
`lymphocyte
`* Imrnunoglobulin on the surface of T lymphocytes is very difficult to demonstrate
`(83, 185).
`t Neoplasms of T lymphocytes (commonly called leukemias in the mouse)
`arc the most common lymphorcticular neoplasms in the mouse. Receptor immunoglobulin
`t Reticulum cell neoplasm type B in the
`has not yet been demonstrated on these tumors.
`Dunn classification (38, 61) is a pleomorphic tumor containing mixed cell types. Some of the
`ce1ls may be large dendritic macrophages that are associated with reticulum fibers. It has
`not been shown that dendritic macrophages themselves produce immunoglobnlin. The
`source of immunoglobulin associated with these tumors presents problems that are dis(cid:173)
`§ Some plasmacytomas have been shown to have immunoglobulin
`cussed in the text.
`on their surface (263, 275).
`
`The plasmacytomas represent only one group of immunoglobulin-producing
`tumors. Lymphomatous neoplasms in mice (reticulum cell sarcomas, plasma cell
`leukemias) are other palhologic forms that have been associated with immuno(cid:173)
`globulin production. These tumors appear to be derived from precursors of plasma
`cells; unlike the plasmacytomas, imrnunoglobulin production is not consistently
`associated with tumors of these morphologic types. There is potentially a third class
`of tumors associated with irnrnunoglobulins; these are tumors of cells that have only
`immunoglobulin receptors on their surface and do not possess the potential for
`secreting immunoglobulin. Very little is currently known about such tumors
`although a few lymphoid tumors have recently been described in man (201) and
`mouse (263) that have immunoglobulin receptors on their surface.
`It is a remarkable and fortunate experimental fact thal plasrna cell tumors
`can be induced in umlimited numbers in the highly inbred BALB/c strain of mice
`by relatively simple procedures such as the intrapcritoneal injection of mineral oil.
`Although the plasmacytomas are relatively easily produced, the process by
`which they evolve is very complex. A number of factors have been described;
`foremost among these is the unique genetic susceptibility of the inbred BALB/c
`strain of mice. The genetic basis of susceptibility has not been worked out and
`remains one of the intrig-i.\ing problems in this field. A second essential factor in
`plasmacytoma formation is the abnormal peritoneal environment-the anatomic
`site of plasmacytoma formation. An abnormal peritoneal environment can be
`created by implantation of large solid plastic materials or injection of mineral oils.
`There has been also speculation about the possible role of viruses in plasma cell
`
`3 of 89
`
`BI Exhibit 1110
`
`
`
`634
`
`MICHAEL POTTER
`
`Volume 52
`
`tumor formation based on the finding of intracisternal type-A particles in virtually
`CYery plasma cell tumor so far examined by electron microscopy.
`Plasma cell tumors have been most useful in providing a source for large
`quantities of homogeneous immunoglobulin, since most of the tumors arc relatively
`easy to transplant in syngeneic hosts. The transplants obtain massive size,
`approaching one-third of the body weight, and large quantities of immunoglnbulin
`can be i8olaLed from the serum, ascites, or urine of these mice. Tumor transplant
`lines arc usually quite stable and maintain the continuous production of the char(cid:173)
`acteristic immunoglobulin through many generations (215). The oldest immuno(cid:173)
`globulin-producing tumor (X5563) has been in nearly continuous passage since
`195 7 (216) and still produces the same immunoglobulin.
`~fany of the first uses of homogeneous immunoglobulins concerned classifying
`\·arious forms of immunoglobulins by chemical and serological methods. The dis(cid:173)
`covery of antigen-binding activity of a few mouse myeloma proteins has stimulated
`many sLuclies on the immunochemistry of "homogeneous antibody." M yeloma
`proteins with antigen-binding activity are "antibody-like" and probably resemble
`an individual species of immunoglobulin that might be found in a population of
`molecules that bind the same antigen (antibody).
`The study of immunoglobulin synthesis has been greatly facilitated by the
`availability of murine plasma cell tumors. Much of the early work dealt with the
`assembly of the immunoglobulin molecule, whereas recent investigations, though
`far from resolved, have been more concerned with the problem of the immuno(cid:173)
`globulin messenger R>JA. In an exciting series of investigations Scharff and his
`colleagues have developed selective in vitro cloning procedures based on specific
`immunoglobulin production and have discovered a high rate of development of
`cell types with defective programs of immunoglobulin synthesis.
`
`A. Abbreviations
`
`1) Plasmacytomas. Plasmacytomas arc designated by a prefix (which usually
`contains PC to indicate plasma cell tumor), an accession number, and occasionally
`a letter that indicates a transplant subline. Prefixes may indicate the agent used to
`induce the tumor or the name of the investigator in whose laboratory the tumor
`was induced; for example, MOPC 104£ = mineral oil-induced plasmacytoma 104,
`transplant sub line E. Common prefixes:
`Adj
`Adjuvant
`MO
`Mineral oil
`TE
`Tetramcthylprntadccane
`HO
`7n hexyloctadecane
`Salmonella associated
`SA
`M
`Merwin
`D
`Dunn
`Sanford
`S
`Mcintire
`Mc
`
`4 of 89
`
`BI Exhibit 1110
`
`
`
`July J.972
`
`\IO USE TUJ\JOH.S .\ND 1\fYELOM:\ PROTEINS
`
`635
`
`Yancey
`Y
`Guldslein
`G
`.\/loriwaki
`:YI S
`~.f. Cohn and associates, Salk Institute, h ave 2 series of plasmac.y tomas desig(cid:173)
`nated the S (Salk) and J series; they do not use PC in the prefix.
`The first transplantable plasmaeytorna was design ated by the experimental
`accession number X5563 and has retained the name.
`Scharff has designated the non producing line of MPC 11 as NP2.
`2) l mmunoglobulins. lmmunoglobulin (Ig) rnolcc:ulcs are usually classified ac(cid:173)
`cording to the heavy-chain subunit. Two nomenclatures exist for the mouse, one
`originated by Fahey et al. (72, 73) and the other by Potter and Lieberman (219,
`220).
`
`Class
`
`Potter
`lgM
`l g A
`IgF
`IgG
`I gH
`l gJ606
`
`Fahey
`lgM
`lg A
`lg-y I
`l g')'2a
`lg-y2b
`IgG3
`
`Heavy Chain
`Poller ct a!.
`µ
`a
`(}
`
`'Y
`1/
`
`Fahey ct al.
`µ
`a
`'YI
`-y2a
`')12b
`
`3) Imrnurwglohulin subunits
`Proteolytic fragments: Fan = fragment with antibody activity, Fe = crystal(cid:173)
`lizabk fragment. The proteolytic fragments were originally isolated from
`antibod y, corresponding fragments can be derived as well from nnmuno(cid:173)
`globulins th at have no known antigen binding activity.
`Polypeptide chains : II = heavy chain, L = light chain .
`Polypeptide chain segments : V = varia ble, C = constant. Each chain has a V
`, C11 , and V L , Cr.. Greek letter designa tions
`;rnd C segmen t designated V H
`for chain dasscs can be used in place of H or L.
`.J) Immunoglobulingenes. The complete designation for an immunoglobulin gene
`is l g followed by the symbol for the irnrnunoglobulin polypeptide segment followed
`by the class symbol: e.g. l g C u A. CH genes arc A, .F, G, H, M, and J 606 . CL genes
`arc K (kappa), L 1 and L2 fo r the two lamhda types.
`Allotypic antigenic determinants assigned to polypeptide segments arc desig(cid:173)
`nate<l by a superscript arabic numeral: e.g. lg CH A 12 • ia. 14•
`Su-ains of mice that arc congenic. for irn111unoglobulin genes arc designated by
`symbols that include the paren t strain, strain source of new immunoglobulin
`genc(s), number of introgrcssivc backcrosses (BC), and nu mber of homozygous
`brother-sister matings (F); e.g. BALB/c .C5 7BL/Ka l g Cu BC 20 F2 •
`5) Lighl-chain dasses: K = ka ppa, /..l = la mbda 1, /,.~ = lambda 2.
`6) Minernt oils: Bayol F, Primo] D (Hurnblc Oil, R ahway, XJ.), Drakeol 6VR
`(Pennsylvan ia Oil).
`7) .i\--1 component. l Iomogeneous immunoglobulin found in high concentration
`in scrum or other body Huid; rnay be of tumor or normal cell origin.
`
`5 of 89
`
`BI Exhibit 1110
`
`
`
`636
`
`MICHAEL POTTER
`
`Volume 52
`
`8) Antigens
`MuLV
`G
`G 1x
`
`= murinc leukemia virus
`= Gross
`= Gross antigen controlled in part by · a gen e in the 9th
`linkage group of the mouse
`Gross cell surface an ti gen
`GCSA
`= group-specific antigen of MuL V
`gs
`= plasma cell alloantigen
`PC.1
`MSPCA = mouse plasma cell tumor a n tigen
`theta antigen
`=
`{)
`9) Proteins, protein hormones
`= bovine scrum albumin
`BSA
`MUP = major urinary protein complex in the mouse
`ACTH = adrenocorticotropic hormone
`= follicle-stimulating hormone
`FSH
`= luteinizing hormone
`LH
`thyroid-stimulating hormone
`TSH
`=
`I 0) Chemicals
`DNP = dinitrophenyl
`TNP = trinitrophenyl
`SDS = sodium dodecylsulfate
`DOC = deoxycholatc
`Ac U = 5-acetyluracil
`11) Amino acids
`C
`H
`I
`M
`s
`v
`A
`G
`L
`p
`·r
`PC A
`12) Cell types
`B cell = bone marrow-derived lymphocyte
`T cell = thymus-derived lymphocyte
`
`cysteine (Cys)
`histidine (His)
`isolcucine (Ile)
`methionine (Met)
`serine (Ser)
`valine (Val)
`alanine (Ala)
`glycine (Gly)
`leucinc (Leu)
`proline (Pro)
`threonine (T hr)
`pyrrolidone carboxylic acid
`
`phenyla lanine (Ph e)
`F
`K
`lysine (Lys)
`arginine (Arg)
`R
`tyrosine (T yr)
`Y
`Vv
`tryptophan (Trp)
`aspartic acid (Asp)
`D
`N
`asparagine (Asn)
`asp or asn
`B
`glu tamic acid (Glu)
`E
`Q
`glu tamine (Gin)
`Z Glu or Glx
`
`I. PLASMACYTOMAS
`
`A. Induction of Plasmacytomas
`
`Merwin and Algire ( 165) were the first to induce plasmacy tomas in BALB/ c
`mice. They were stu dyin g the long-term survival of allogeneic tissu e (C3H mam-
`
`6 of 89
`
`BI Exhibit 1110
`
`
`
`July 1972
`
`MOuSE TUMORS A'.'l'D MYELOtvfA PROTElNS
`
`637
`
`mary tumor tissue) inside .Millipore diffusion chambers that had been implanted
`intraperitoneally in BALB/c mice. After 6 months the BALB/c mice developed
`hemorrhagic ascites that was caused by plasmacytoma or fibrosarcorna that had
`developed in the subperitoneal connective tissues. Menvin and Redmon (166)
`demonstrated later that empty chambers, Millipore discs (17.5 or 21 mm diam), or
`rough-edged plexiglass (Lucite) borings (1 mm diam) also induced plasmacytomas.
`i\1ore tumors were induced with the larger (2 I mm) discs and borings. Few or no
`tumors were induced with smooth Lucite rings, Lucite fragments, or small discs.
`The large discs and borings caused considerable irritation and chronic inflamma(cid:173)
`tion. :Merwin and Redmon suggested the irritative inflammatory reaction was an
`important factor in plasma cell tumor development. The fibrous chronic reactive
`tissue developed over most of the contiguous peritoneal surfaces. Fibrosarcomas
`arose in the capsules covering the discs or chambers while plasmacytornas de(cid:173)
`veloped clsev1rhere on the peritoneal surfaces.
`Recently Anderson (5) has induced plasmacytornas with Lucite discs and
`noted, as did Ylerwin and Redmon, that the large discs often eroded the gut wall
`and in extreme cases the edge of the disc could be found in the gut lumen. These
`reactions apparently occurred without suppurative peritonitis, although many
`adhesions developed. Merwin and Redmon (166) also noted that inbred BALB/c
`mice appeared uniquely susceptible, since discs implanted in other strains induced
`only a few (if any) plasma cell tumors.
`After the observations of Merwin and Algire (165), other agents that induce
`plasma cell tumors in BALB/c mice were found. The first was a staphylococcal(cid:173)
`adjuvant mixture (225) that contained one part heat-killed Staphylococcus and one
`part incomplete Freund's adjuvant [8.5 parts Bayol F and 1.5 parts Arlacel A
`(mannide monoolcate)]. This mixture was used by Lieberman euil. (138) to induce
`ascites in mice in order to produce quantities of antibody (138, 140). The staphylo(cid:173)
`coccal-adjuvant mixture also induced extensive peritoneal adhesions and a chronic
`inflammatory response. Lieberman et al. ( 140) subsequently induced plasma(cid:173)
`cytomas in BALB/c mice with this adjuvant material.
`In a search for the active component in the adjuvant, Potter and Boyce (213)
`found that the mineral oil alone induced plasmacytornas. A few tumors were in(cid:173)
`duced with a single injection of 0.5 ml of the mineral oil Bayol F; three 0.5-ml
`injections of mineral oil spaced 2 months apart induced plasmacytomas in 40-60 3
`of females so injected (206, 213). A number of light and heavy pure white mineral
`oils were active, e.g. Bayol F, Drakcol 6VR, Primal D, and other USP grade oils
`sold commercially (.3, 206). Mineral oils contain large numbers of straight-chain,
`branched-chain, and ring-structured saturated hydrocarbons. Only a few in(cid:173)
`dividual components are available in pure form, but among those tested with the
`same regimens used for mineral oils all have been as active or more so than mineral
`oil alone; this includes pristane (2, 6, 10 1 14-tctrarncthylpcntadecane), phytane
`(2,6,10,14-Letramethylhcxadccanc), and 7n hexyloctadecane (5, 6). Usually the
`latent period for pristanc is shorter than with mineral oil (Fig. 1 ). In preliminary
`experiments n-hcxadecarn: was also tried but found to be toxic and could not be
`evaluated properly. Carcinogenic polycyclic hydrocarbons in trace quantities are
`apparently not the active components, as they are not present in the oils nor in
`
`7 of 89
`
`BI Exhibit 1110
`
`
`
`f
`I
`
`6t)~ I
`
`---- Pris tone
`- - Roy o l F
`
`--'
`J w
`
`()
`
`<i
`:;,,
`'1i 20 ..
`--' "
`
`')
`
`I
`b G
`
`l;>O
`
`F
`F
`
`/ J
`
`I
`I
`I
`
`F
`
`f
`I
`I
`I
`I
`J (/
`1/
`fl , ,
`(/
`Jr
`//
`f I 1/
`
`63B
`
`MICHAEL POTIER
`
`Volume 52
`
`/ !5 i 9 I~
`
`~
`/!.
`18 0
`
`240
`
`l
`300
`
`i
`360
`
`420
`
`48 0
`
`54 ()
`
`~/)()
`
`FIG. l. Incidence of plasma cell tumors in BALB/c mice treated with three o.~-ml intra(cid:173)
`peritoneal injections of pristane (broken line) or Bayol F (solid line) from four different experi(cid:173)
`ments; injections were given bimonthly. These experiments were selected because the mice were
`carefully followed for tumor incidence. Fractious indicate actual number of plasma cell lumors
`over total number of mice injected. Time of appearance of plasma cell tumors was plotted from
`the day the mice received the lst injection (30-90 days of age). Pristane-injected mice developed
`plasma cell tumors sooner and more precipitously than did mineral oil-treated mice. Plasma cell
`tumors did not usually begin to appear until 30-GO days after the 3rd injection of oil. The greatest
`number develop within the I st year, but others continue to form during the 2nd year.
`
`the pure alkancs (206, 233) that are active inducers. It may be concluded that
`many different alkanes induce plasma cell tumors.
`Several characteristics of the mineral oils and branched-chain alkancs are
`relevant. \Vilncr et al. (302) synthesized 7n hexyloctadecane and several other
`straig-ht- and branched-chain a lkancs in a search for a more effective chemically
`defined immunologic adjuvant. They found straight-chain alkancs C10-C1s were
`highly irritating to the skin of guinea pigs, but that this adverse property could be
`overcome by using branched-chain compounds of roughly similar molecular
`weight. They cornpared a va riety of branched-chain compounds, induding both
`pristane and 7n hexyloctadecanc, to Drakeol 6VR as adjuvants. They found that
`pristanc and 7n hexyloctadecane, for example, were superior or equal to Drakeol
`6VR as an immunological adjuvant by sustaining high titers of antibody over long
`periods of time. (The test oils and Arlacel A were mixed with antigens to make
`water-in-oil emulsions; for plasma cell induction the unmixed oil or hydrocarbon
`is injected directly.)
`Pristane was first isolated in 1917 by Tsujirnoto (sec ref 18) from basking
`shark livers. Subsequent studies have shown its origin is probably biogcnic as 1-3 %
`of the body fat of several species of marine copcpods is pristanc (27). Pristane has
`also been found in a variety of mineral oils. Avigan ct al. (18) isolated 2-9 µg
`pristane/ g of human scrum or liver and 52 µg/ g in skin; phytane >vas also observed
`but in lovvcr concentrations. It was suggested that pristane m ay enter the organism
`
`8 of 89
`
`BI Exhibit 1110
`
`
`
`July 1972
`
`MOUSE TUMORS AND MYE!.OMJ\ PROTEINS
`
`639
`
`nc. 2. Peritoneal oil granu(cid:173)
`loma in a primary plasrnacytoma
`(hcmatoxylin and cosin stain).
`Upper: photomicrograph show(cid:173)
`ing
`extensive mesenteric
`oil
`g ranuloma.
`l mcnscly
`stained
`plasmacytoma tissue is irregularly
`distributed on some of peritoneal
`[10 X J L-Ower: high(cid:173)
`surfaces.
`power photomicrograph of a
`single area from same section
`showing plasmacytoma tissue on
`peritoneal surface of oil granu(cid:173)
`loma. fl65X]
`
`via skin absorption. O'Neill et al. ( 187) have also identified pristane in human skin,
`and Boitnolt el al. (28) have repeatedly isolated mineral oil frorn human Lissues.
`These compounds, however, have not been implicated in human plasrna eeJ turnor
`formation.
`
`CH3
`CH:1
`CHa
`CH3
`I
`I
`I
`I
`CHaCH2CH2CH2CH2CH2CH2CH2CH2CH2CII2CH2CH2CH2CHa
`
`2, 6, l 0, 14-tetramethylpentadeeane (pristane)
`
`Plasma cell tumors arise in the peritoneal oil granulorna and appear to require
`this environment for development as they are not observed elsewhere nor have they
`been induced by subcutaneous injections of mineral oil (223) (Fig. 2). Intraperi(cid:173)
`toneal mineral oil evokes the formation of an oil granuloma. on peritoneal surfaces,
`abdominal wall, d iaphragm, mesentery, peripancreatic regions, and retroperi(cid:173)
`toneally (223). This tissue begins to develop within a few days after oil injection.
`The main bulk of the oil granulorna consists of oil-laden histiocytes that adhere to
`
`9 of 89
`
`BI Exhibit 1110
`
`
`
`640
`
`MICHAEL POTIER
`
`Volume 52
`
`each other, attach to peritoneal surfaces, and become vascularized from the mcsen(cid:173)
`teric vessels beneath. The tissue varies in the amount of reactive components; some
`oil granulomas contain many focal collections of lymphoid cells and occasional
`small focal areas of granulocytic proliferation containing primitive and mature cells
`in the series. Often the oil granuloma contains only a few scattered lymphoid cells
`between the histiocytes. Plasma cells with a normal morphology (a small eccentric,
`rather pyknotic, nucleus) are often observed lying between histiocytes or long blood
`vessels. Further, they are often found admixed with lymphoid cells in areas con(cid:173)
`taining many lymphocytes. Focal proliferations of plasma cells appear to arise in
`similar locations. Dividing plasma cells usually are hyperchromatic and their
`nuclei arc larger and less uniform than normal. Plasma cells in tumors are usually
`characterized by their relatively large size and hyperchromicity with hematoxylin
`and eosin stain. Many of these cells have multilobed nuclei or double nuclei. A
`perinuclear clear zone (hof) is a characteristic feature. Incipient histologic stages of
`plasma cell tumor development have been reported to develop in mice 4 months
`after the first injection of oil; others have been found in mice autopsied at various
`times after they had received a complete series of mineral oil injections (223). That
`plasma cell tumors may actually develop in the peritoneal fluid has not been ruled
`out, although direct evidence for this has not yet been obtained. Ascitic origin is
`suggested by histologic study of primary tumors where the plasmacytomatic process
`is seen to be confined almost exclusively to peritoneal surfaces (Fig. 2), suggesting
`primary origin in the fluid and subsequent multiple seedings of tumor cells on
`various loci. The vast majority if not all mice with large bizarre plasma cells
`(resembling those seen in neoplasrns) in the peritoneal fluid also have infiltration of
`the connective tissues. Invasion of the peritoneal connective tissue by plasmacytoma
`is associated with peritoneal bleeding, which in some cases develops precipitously
`and may kill the mouse. Interestingly, although plasma cell tumors apparently do
`not develop in other strains injected with mineral oil, the oil granuloma that de(cid:173)
`velops in these other strains is quite similar in morphology to that observed in
`BALB/ c mice.
`The only other tissues consistently containing oil granulorna are the superior
`mediastinal lymph nodes that drain the peritoneum. Plasmacytomas metastasize to
`these nodes but do not a ppcar to develop in them (223); i.e. isolated cells are found
`in the sinuses rather than the medullary cords. Further, plasma cell tumors have
`not yet been observed to develop in the bone marrow cavities or other lymph nodes.
`
`B. Factors Influencing Plasmacyloma Induction and Incidence
`
`J) General. A basic technical problem in studying factors influencing plasma
`cell tumor induction in BALB/ c mice is establishing the true incidence of tumors.
`Essentially incidence is defined as the number of plasma cell tumors obtained
`divided by the total number of mice alive after all the mineral oil treatments
`are given.
`Groups of 30-40 female BALB/c mice (average size of an experimental group)
`are difficult to follow over long periods, for often individual mice die rather sud-
`
`10 of 89
`
`BI Exhibit 1110
`
`
`
`July 1972
`
`lv10USE TUMORS AND MYELOMA PROTEINS
`
`641
`
`dcnly and unexpectedly of pncumonitis and other conditions during the course of
`an experiment. If this happens overnight or on a weekend, the mouse is autolysed
`before any diagnosis can be made from tissue sections. In order to establish
`an incidence, mice must be examined at least 3 times weekly for signs of illness
`(weight loss) and the development of ascites (bloody). All mice with ascites should
`be tapped and smears made to detect abnormal cells. Mice with a nonbloody
`aseites that contains no malignant cells can be returned to their cages and re(cid:173)
`examined in several days when deemed necessary. Although the diagnosis can be
`made from smears containing numerous characteristic cells, confirmation by tissue
`section is advisable in all cases where determination of incidence is critical. Main(cid:173)
`tenance of mice in closed colonies could potentially aid in reducing the number of
`deaths due to infections. In our laboratory, where mice arc brought in and out,
`\Ve routinely treat mice with aureomycin in the drinking water to control pneumo(cid:173)
`nitis when it appears (206). Since considerable time is involved in an incidence
`experiment it is important to carefully select a group of mice that have been suc(cid:173)
`cessfully acclimatized to the animal room and probably have developed a good
`natural immunity to the common flora in the room. Female mice arc used prefer(cid:173)
`entially because they can be housed in groups of eight; male mice under these
`c:onditions fight and kill each other.
`An incidence of 60 o/r .. ofplasmacytornas is a good yield of tumors in an average
`experiment (sec Fig. I). However, other factors, many not well understood (e.g.
`immunization, infections, stress, natural immunization, etc.), influence plasma cell
`tumor development and affect incidence and hence experiments must be carried
`out with adequate controls.
`2) Genetic. Plasma cell tumors have been induced in high incidence in only two
`inbred strains, BALB/c (166, 206) and NZB (294). Relatively few other strains
`have been studied: DBA/2, A/He, A/LN, C57BL/Hc, C57BLKa, C3H/ He, and
`S\VR (166). An occasional plasma cell tumor has been found after mineral oil
`injections or Millipore diffusion chamber implantation in members of these
`strains ( 166, 206, 307). The restriction of plasma cell tumor susceptibility to a
`few specific inbred strains of mice suggests that specific genes are involved in
`plasma cell tumor development. Xone of these have yet been identified. An in(cid:173)
`teresting experimental problem in the future will be to find how genetically de(cid:173)
`termined factors influence plasma cell tumor development.
`First-generation (F1) hybrids of BALB/c and other strains develop p