ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS 78, 306-318 (1958) The Effect of Variation in Molecular Weight on the Antigenicity of Dextran in Man’ Elvin A. Kabat and Ada E. Bezer Departments of Microbiology and Neurology, College of Physicians and Surgeons, Columbia University; and the Neurological Institute, Presbyterian Hospital, New York, New York Received April 21, 1958 Earlier studies from this laboratory (1, 2), confirmed by Maurer (3), have shown that both native and clinical dextrans are antigenic in man and that injection of 1 mg. of material would give rise to precipitating antibodies and to wheal and erythema type skin sensitivity; similar findings were obtained with a native levan (4). Clinical dextrans for use as plasma volume expanders (5) are prepared by acid hydrolysis of the native dextrans, which have molecular weights in the tens of millions, and are fractionated by ethanol to give products of mean weight-average molecular weight of about 75,000 f 25,000. In the course of the evalua- tion of dextran as a plasma volume expander, a series of dextran frac- tions, of graded molecular weight ranging from 10,600 to 412,000, were prepared and characterized for the National Research Council by Com- mercial Solvents Corporation. The availability of these materials made it possible to study the relationship between molecular weight of the dextran and its capacity to induce antibody formation in man; several other clinical dextrans were also examined for antigenicity. The data indicate that products of mean molecular weight of about 90,000 or above were better in inducing antibody than were products of lower molecular weight. EXPERIMENTAL Dextrans National Research Council (NRC) fractions l-8 represent the series of carefully prepared dextran fractions made from partially hydrolyzed native dextran of the 1 These studies were supported by the National Science Foundation (G-1586) and the William J. Matheson Commission. 306
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 1
`
`

`

`ANTIGENICITY OF DEXTRAN IN MAN 307 NRRL B512 strain of Leuconostoc mesenteroides by Commerical Solvents Cor- poration and kindly made available through the Panel on Plasma and Subcommit- tee on Shock of the National Research Council. These fractions all contained less than 0.005~o nitrogen based on dry weight of dextran. Average molecular weights of each fraction were determined by light scattering and represent weight-average molecular weights. Since these materials are not monodisperse, an estimate of the degree of polydispersity of each fraction was obtained by methanol frac- tionation as follows: Methanol was added to precipitate about 510% of the material; this high fraction was removed, and more methanol was added to pre- cipitate the bulk of the remainder of the dextran, leaving about 510% of the total in solution. This main fraction was removed, and the low fraction was pre- cipitated by further addition of excess methanol. After washing the high fraction with dilute methanol of the same composition as that used for precipitation, the high and low fractions were dissolved, lyophilized, and weighed, and their molec- ular weight was determined by light scattering (6). Thus NRC Fr. 1 had a mean molecular weight of 10,600; the 10.5% precipitated by the lowest methanol con- centration, high fraction, had a molecular weight of 16,400; while the 8.7% re- quiring the most methanol had a molecular weight of 5900. Detailed data on these fractions are included in Table I and were supplied by Dr. Homer E. Stavely. Two other preparations of dextran APC Nos. 55 and 54 were prepared and characterized at the Northern Regional Utilization Laboratory, Peoria, Illinois, and kindly furnished by Dr. F. Senti. Dextran APC No. 55 was prepared by controlled enzymic synthesis (7, 8) with dextransucrase from the B512 strain of dextran; dextran APC No. 54 was prepared from the organism isolated by Dr. E. J. Hehre which synthesized dextran of low molecular weight (9) [NRRL B13.51 (lo)]. Dr. Senti also determined the molecular weighm of NRC Frs. 7 and 8 as 255,000 and 412,000, respectively. In addition, two clinical samples of British dextmn, Dextraven and Intradex, were also provided through Dr. Margaret H. Sloan of the National Research Council. The Intradex manufactured by the Glaxo Laboratories was furnished as a 6% solution batch No. 54/056 manufac- t.ured in May 1954. Dextraven was also a 6% solution from t,he Renger I,abora- tories I,td. as batch 2345 manufactured March 11, 1954. Antigenicity Studies The procedure followed was that described in Ref. (2). Healthy medical stu- dents were used. After an initial 50-ml. blood sample was taken and initial skin testing, each subject received two 0.5.ml. injections of a solution of 1 mg./ml. of the dextran subcutaneously a day apart. Three weeks after the last injection, a second blood sample was obtained and the subject was again skin tested. Sera were analyzed for antibody by the Heidelberger and MacPherson (11) microquan- titative precipitin technique [cf. (12)l using 3.0 ml. serum/tube, making addi- tions of dextran as described in Ref. (2). All sera were analyzed for sntidextran using the dextran with which the subject was immunized as well as with the native dextran prepared from the NRRL B512 strain; samples of native dex- tran N236 and 279 were kindly supplied by Commercial Solvents Corporation, and a sample NRRL B512 dextran was supplied by Dr. Allene Jeanes. Half- liter samples of blood were obtained from those individuals who showed a sub- stantial antibody response.
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 2
`
`

`

`
`
`
`
`
`ESHOm
`
`
`3553.5.
`
`
`H+Iw:«anm:m4§H+I3;2WEwea:
`IIm.mmaamedv:H+AHf:252a...m:
`+I3w2w.mmgm:
`II3Wmm4.mga:
`IIeac...33mm:
`IIbeaso.o«.9«S
`IInowefieo.om:
`IImy:edaso.om:aI35oas3mm:85.5#28mg«.283m”SOME
`IIEZ3meEIISnnoyo3o:89%w.mSac?m.m08.3mamomz
`II54No3‘wea:
`II$85823.o.oa:
`IIdodoasZa:
`IIHaj«a,noa:
`II3edasdo3:
`IIm;.3mg#o.ome
`mummy.mmwymyummyammmm.mmmmmmmmywyasin3.23
`
`
`
`
`
`uownmvmwwnkfiflwm3u>mudqknEduum
`
`u.n++Im;fin33m2.N+IMS2”.2a:§89%mao3«mm«.58m«A:.9592
`n.”IgNo2asEafl+I9%Wm9%.«Ao:89%5m8mR:my85mwgmomz
`
` $59:an
`
`3:Ex:.3:$31ewe
`
`
`
`Eabwunwmrcmbxuvcomuunin—Eficc»cum:“oompsw395m«85qu
`
`
`
`
`
`
`
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 3
`
`T
`A
`B
`L
`E
`
`I
`
`R
`e
`s
`p
`o
`n
`s
`e
`
`o
`f
`
`H
`u
`m
`a
`n
`s
`
`to
`I
`m
`m
`u
`n
`i
`z
`a
`t
`i
`o
`n
`
`w
`i
`t
`h
`
`V
`a
`r
`i
`o
`u
`s
`
`D
`e
`x
`t
`r
`a
`n
`
`F
`r
`a
`c
`t
`i
`o
`n
`s
`
`a
`n
`d
`
`w
`i
`t
`h
`
`C
`l
`i
`n
`i
`c
`a
`l
`
`D
`e
`x
`t
`r
`a
`n
`s
`a
`
`-
`
`U
`!
`
`-
`
`I
`
`c
`
`-
`
`-
`-
`
`-
`
`I
`
`-
`
`A
`n
`t
`i
`b
`o
`d
`y
`
`n
`i
`t
`r
`o
`g
`e
`n
`
`p
`r
`e
`c
`i
`p
`i
`t
`a
`b
`l
`e
`
`f
`r
`o
`m
`
`s
`e
`r
`u
`m
`
`b
`y
`
`d
`e
`x
`t
`r
`a
`n
`
`s
`e
`d
`
`f
`o
`r
`
`i
`m
`m
`u
`n
`i
`z
`a
`t
`i
`o
`r
`
`A
`n
`t
`i
`b
`o
`d
`y
`
`n
`i
`t
`r
`o
`g
`e
`n
`
`p
`r
`e
`c
`i
`p
`i
`t
`s
`b
`l
`e
`
`f
`r
`o
`m
`
`s
`e
`r
`u
`m
`
`b
`y
`
`n
`a
`t
`i
`v
`e
`
`l
`e
`x
`t
`r
`a
`n
`-
`B
`5
`1
`Z
`
`s
`t
`r
`a
`i
`n
`
`‘
`r
`e
`i
`m
`m
`u
`-
`
`P
`o
`s
`t
`i
`m
`m
`u
`
`n
`i
`z
`a
`t
`i
`o
`n
`
`n
`i
`z
`a
`t
`i
`o
`n
`
`_
`_
`-
`
`P
`g
`.
`/
`m
`l
`.
`
`p
`g
`.
`/
`m
`l
`.
`
`0.9
`1
`2
`.
`3
`
`2
`.
`1
`
`9
`.
`6
`
`2
`.
`2
`
`6
`.
`1
`
`N
`o
`t
`
`e
`n
`o
`u
`g
`h
`
`C
`u
`t
`a
`n
`e
`o
`u
`s
`
`r
`e
`a
`c
`t
`i
`o
`n
`
`t
`o
`
`d
`e
`x
`t
`r
`a
`n
`
`u
`s
`e
`d
`
`f
`o
`r
`
`i
`m
`m
`u
`n
`i
`z
`a
`t
`i
`o
`n
`
`M
`o
`l
`e
`c
`u
`l
`a
`r
`
`W
`e
`i
`g
`h
`t
`
`D
`e
`x
`t
`r
`a
`n
`
`s
`a
`m
`p
`l
`e
`
`S
`u
`b
`j
`e
`c
`t
`
`1
`6
`4
`
`1
`6
`5
`
`1
`6
`6
`
`1
`6
`7
`
`1
`6
`8
`
`1
`6
`9
`
`1
`7
`0
`
`1
`7
`1
`
`1
`7
`2
`
`1
`7
`3
`
`1
`7
`4
`
`1
`7
`5
`
`1
`7
`6
`
`1
`7
`7
`
`1
`7
`8
`
`1
`7
`9
`
`1
`8
`0
`
`1
`8
`1
`
`1
`8
`2
`
`1
`8
`3
`
`:
`i
`t
`
`1
`8
`6
`
`.
`-
`
`I
`
`-
`
`A
`v
`e
`r
`a
`g
`e
`
`H
`i
`g
`h
`
`L
`O
`W
`
`‘
`r
`e
`i
`m
`m
`u
`
`n
`i
`z
`a
`t
`i
`o
`n
`
`P
`g
`.
`/
`d
`
`0.9
`1
`.
`2
`
`A
`:
`:
`
`A
`:
`:
`
`0
`.
`1
`
`0
`.
`2
`
`0
`.
`4
`
`K
`i
`
`0
`:
`9
`
`1
`.
`2
`
`0
`.
`5
`
`0
`.
`0
`
`~
`~
`~
`
`1
`:
`5
`
`E
`i
`
`0
`:
`1
`
`0
`.
`0
`
`0
`.
`7
`
`o
`s
`t
`i
`m
`m
`u
`-
`
`n
`i
`z
`a
`t
`i
`o
`n
`
`P
`&
`/
`+
`=
`1
`.
`
`1
`0
`.
`3
`
`9
`.
`1
`
`4
`.
`8
`
`0
`.
`0
`
`1
`.
`1
`
`4
`.
`0
`
`0
`.
`7
`
`;
`:
`A
`
`1
`2
`.
`7
`
`3
`.
`1
`
`5
`.
`8
`
`2
`8
`.
`5
`
`A
`:
`:
`
`1
`:
`:
`:
`
`7
`.
`8
`
`0
`.
`2
`
`X
`:
`8
`
`$
`2
`
`‘
`r
`e
`i
`m
`m
`u
`-
`
`n
`i
`z
`a
`t
`i
`o
`n
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`r
`t
`b
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`‘
`o
`s
`t
`i
`m
`m
`u
`-
`
`n
`i
`z
`a
`t
`i
`o
`n
`
`%
`
`%
`
`7
`.
`4
`
`2
`3
`2
`,
`5
`0
`0
`
`4
`.
`8
`
`1
`2
`9
`,
`0
`0
`0
`
`8
`.
`2
`
`1
`6
`7
`,
`8
`0
`0
`
`5
`.
`8
`
`7
`8
`,
`0
`0
`0
`
`4
`.
`3
`
`1
`0
`7
`,
`5
`0
`0
`
`5
`.
`7
`
`6
`2
`,
`5
`0
`0
`
`1
`2
`.
`4
`
`5
`9
`,
`2
`0
`0
`
`1
`5
`.
`1
`
`3
`7
`,
`7
`0
`0
`
`.
`-
`
`N
`R
`C
`
`F
`r
`.
`
`6
`
`1
`9
`4
`,
`9
`0
`0
`
`1
`3
`5
`,
`0
`0
`0
`
`9
`1
`,
`7
`0
`0
`
`5
`1
`,
`3
`0
`0
`
`s
`e
`1
`
`0
`.
`6
`
`0
`.
`3
`
`0
`.
`7
`
`0
`.
`0
`
`E
`
`1
`:
`4
`
`n
`-
`
`5
`:
`;
`
`1
`.
`2
`
`1
`.
`7
`
`0
`.
`7
`
`1
`6
`.
`5
`
`3
`.
`5
`
`8
`.
`5
`
`3
`4
`.
`6
`
`2
`.
`3
`
`0
`.
`3
`
`0
`.
`1
`
`1
`4
`.
`9
`
`1
`1
`.
`8
`
`i
`i
`:
`;
`
`:
`i
`
`4
`:
`1
`
`E
`s
`
`N
`R
`C
`
`F
`r
`.
`
`5
`
`N
`R
`C
`
`F
`r
`.
`
`4
`
`N
`R
`C
`
`F
`r
`.
`
`3
`
`

`

`
`
`
`
`:23.m45..agaxmwOmZ.«omaompooflfi95«o@5509531:?aswimwho?03#236mm63muuwmnzw:339.35::55m.
`
`
`
`
`
`3395mASE/12.w:9%23{aJ6YE4830me233JED/Ea».m:3m.93no;Nm6339%hon...m5%naupxewfirs«53:5
`
`
`
`
`
`QED/34.w:fiwand.54J4can45334::283$32me.w:Wm.can.wdAedMERE?‘35:anwv33.3455.nosommiyou?
`
`
`
`
`qfimEmnomuodoaHe4838.2:use?was$33abfiamu:$3$53:aE.8an3$58532H.issuer:“3.3
`
`
`48358Eryn—Eva33:95“3:me
`
`
`
`
`
`
`
`£555.;nag—flaw:o3::if:”35%1.83:5xioagosdqnotafizabEnomwom€85.8289c3.9:».wmIOm4:o3am
`
`
`
`
`
`
`
`
`
`.figgmAH.HofiomQQéoSmSQSO353cmEGSEEOO:5:
`
`anAc8Scam<325%:S
`
`
`cood...EJEEEVUmmdm<
`
`a:I~«28:E8::3:,38,2359%
`
`mm:W3H,
`
`8H,
`
`
`
`
`
`mg33%“aim"81$,«.2?cookmmmmdhOWZ
`
`I
`
`
`fi+I
`
`+++I
`
`|
`
`
`ovoéfi omooofifioomoa NNOGMH
`
`
`«wmwww quwqewwmamh wadwwb
`ONOOCHODOv—«Om HNOCv—(v—Q
`oodmofi
`
`
`OODOr-IO Ov—‘OHOOOOOOON ODOOOO Hmr—OOQ
`_‘
`N
`
`
`cov—Iv—ILQCON
`wfiww~v waveweesvwww gameoo
`OOCOOO OCONOH
`ODDCv—‘C ODDOOODDOv—IOH
`
`
`3H
`o3
`2:ma2:
`a:
`In
`an
`EN
`3mEN
`an
`EN
`8m
`now
`mom
`
`
`
`
`
`
`wowW5mWcomfl
`
`5%,
`
`3Nmocm,W
`
`
`
`
`
`
`
`mmmoficzm283nm
`
`309
`
`Pharmacosmos AIS v. American Regent, Inc.
`Peti ioner Ex. 1050 - Page 4
`
`PGR2020-00009
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 4
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Ezfiwgclumqufifi
`
`H+IM}:a...an“
`
`u“Io.o9oman
`
`
`defined—2:8.EWWMMMmMIESxounozaumusfifimyoucum:to
`|lN4N5RN
`1Ifiw9N1mm
`IIm6m6man
`II.90Tomam
`
`|Im6DAHum+INd«.ocanEwanSmagma
`I|m5fie«mm
`II94m6«mm
`lIad«6«mm
`
`
`
`houni«5:Nu.O.A55th=95kahaflan—honMa»:00O
`
`
`
`
`
`
`wflhfiNw—wmImmfi“.%%u.m€mem.mmmfimufim.uwhwuflwwm.nflfiwmBodJamiumduu><
`uowuwskmscwfiwawhflflwnmfimflflfl“memmfimmfiwflEIBEHE
`
`
`
`
`fi+I95b43N33mAmmz
`H+I333«am$3:
`umINA5omeScum55x3.
`I|o.na.o9mmBEES
`
`I|5.0adman.5H3150nmmfinm
`
`
`
`I3"m6fieSn89$mg.cowdw«.mSodaE0330«mIOAAN
`
`|Im.wmgmam.xuw$2.335515
`a—mN—ENN~
`
`
`
`.N§\.u:figfiu:.N§\,M:.NE\.m1gN»
`
`Saw03535.550
`
`
`
`
`
`
`310
`
`Pharmacosmos AIS v. American Regent, Inc.
`Peti ioner Ex. 1050 - Page 5
`
`PGR2020-00009
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 5
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`=
`F
`+
`
`T
`f
`
`-
`
`=
`I
`=
`
`i
`
`=
`F
`+
`
`=
`F
`
`-
`
`I
`
`n
`o
`y
`q
`u
`
`-
`l
`l
`l
`l
`I
`l
`U
`!
`)
`S
`O
`,
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`=
`I
`=
`
`n
`o
`p
`e
`z
`y
`J
`
`n
`u
`r
`u
`l
`~
`a
`q
`
`P
`E
`Z
`
`F
`X
`Z
`
`2
`8
`2
`
`I
`G
`Z
`
`0
`8
`Z
`
`6
`Z
`Z
`
`8
`2
`2
`
`L
`Z
`Z
`
`9
`z
`z
`
`I
`j
`z
`z
`
`E
`E
`
`z
`z
`z
`
`T
`Z
`Z
`
`o
`z
`z
`
`6
`1
`Z
`
`8
`I
`Z
`
`L
`I
`Z
`
`0
`0
`9
`‘
`p
`Z
`
`8
`-
`L
`
`%
`
`0
`0
`8
`’
`9
`8
`
`P
`’
`S
`
`%
`
`u
`I
`s
`~
`u
`s
`%
`.
`I
`o
`
`(
`a
`.
`x
`q
`a
`H
`)
`
`I
`S
`%
`1
`8
`
`?
`X
`X
`N
`
`u
`x
`O
`l
`3
`
`u
`s
`.
`q
`x
`a
`p
`
`I
`~
`~
`!
`=
`!
`1
`3
`
`H
`-
`3
`6
`8
`
`

`

`ANTIGENICITY OF DEXTRAN IN MAI\; 311 Quantitative precipitin curves on these samples were determined with the various NRC fractions and with native dextran in the usual manner [cf. (2, 12)j. Precipitin curves were also obtained with a Commercial Solvents clinical dextran N1.5ON used in previous studies (2, 13, 14) which had an average molecular weight by viscosity methods of 60,000. Sera from individuals who showed substantial antibody responses were also examined to compare the relative capacities of isomaltotriose (15) and isomaltotetraose, isomaltopentaose, and isomaltohexaose (16) to inhibit precipitation of the antidextran by clinical dextran N15ON (13, 14) and compared in this respect with an antiserum, (No. 36Da), prepared by immunization with native dextran of the B512 strain which had been studied ear- lier (14). Isomaltotriose was kindly supplied by Dr. Allene Jeanes and the isomal- totetraose, -pentaose, and -hexaose by Drs. J. R. Turvey and W. 6. Whelan. RESULTS Table I presents the findings on immunizing healthy individuals with t:he various dextran fractions and the other dextran preparations. The antibody N precipitable per milliliter serum before and after immuniza- tion by the antigen used for immunization and by native dextran, the results of the skin tests, and the molecular weights of the dextrans are given. Increases in antibody nitrogen from the pre- to the postimmuniza- tion sample of two or more micrograms antibody nitrogen/ml. are con- sidered significant. It is evident that a number of the 18 individuals in- jected with NRC Frs. 6,5, and 4 developed appreciable quantities of pre- cipitable antibody N following injection of these fractions, and that those individuals showing 8 pg. AbN/ml. or more in the postimmunization series to the dextran used for immunization also had developed a positive wheal and erythema skin test to the dextran used for immunization at the time the postimmunization serum sample was drawn. The antibody nitrogen values obtained with native dextran are generally in good agree- ment with those found with the dextran used for immunization but tend t,o be a few micrograms N higher in agreement with previous findings (2) that native dextrans precipitate somewhat more antibody nitrogen than do clinical dextrans. In contrast, however, of the 29 individuals injected with dextran Frs. 3, 2, or 1, only one individual, No. 216, showed any development of wheal and erythema skin sensitivity or a rise of more than 2 pg. N/ml. to the dextran used for immunization; the response in this individual was unequivocal reaching 7.1 fig. AbN/ml. With native dextran, however, two individuals (Nos. 186 and 209) who had received Fr. 3 showed rises of 2.7 and 2.3 pg. antibody N/ml., respectively. With APC dextran Nos. 55 and 54, three of six and two of six individ- uals showed a significant antibody response, respectively, and a positive
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 6
`
`

`

`312 KABAT AND BEZER wheal and erythema reaction was obtained when the antibody level was 8.2 pg. N/ml. or higher to the dextran used for immunization. With clinical Intradex and Dextraven, two of six individuals injected with either material showed a significant precipitin response; the two who responded to Intradex each produced more than 8 fig. AbN/ml. and showed positive skin tests, while those who received Dextraven produced only 3.1 and 7 pg. AbN/ml. and failed to show positive skin tests; these individuals were both skin-tested a second time with identical results. Table II summarizes the findings with the NRC dextran fractions. With dextrans of average molecular weight of 91,700 or over, 12 of 18 individuals produced a significant precipitin response; while of those receiving dextrans with an average molecular weight of 51,300 or lower, only one of 29 individuals gave such a response when the antibody was determined with the dextran used for immunization. Chi square for this difference is 18, with one degree of freedom. When the antibody response was determined with native dextran, 12 of 17 individuals injected with the dextrans of molecular weight 91,700 or above produced antibody as contrasted with but 3 of 29 individuals injected with the lower molecular weight fractions; chi square for this difference is 15, with one degree of freedom. The probability of these results occurring by chance is well be- low 1%. The analysis in Table II groups all of the individuals, since the preim- munization antibody levels as tested with the fractions were below 2 pg. TABLE II Summary of Immunization Data with NRC De&an Fractions l-6 Antibody responsea tested with Dextran fraction Average molecular weight Dextran used for immunization Native dextraa 6 194,900 416 515 5 135,000 3/6 316 4 90,700 316 416 3 51,300 o/11 2/11 2B 35,000 l/12 l/12 1 10,600 O/6 O/6 0 Fraction of individuals showing a rise of 2.0 pg. AbN/ml. or more. Chi square one degree of freedom Fractions 6, 5, 4 vs. 3, 2, 1 18 15 P= <O.Ol <O.Ol
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 7
`
`

`

`UITIGENICITY OF DEXTRAN IN MA!i 313 TABLE III Analysis of Data to Consider Efect of Possible Anamnestic Response Antibody response” tested with: Initial antibody level Dextran fraction used for immunization Native dextran pg. N/ml. 1 0.8orabove ( Below0.8 1 0.8orabove ; BelowO.R-- Fractions 6, 5, 4 Fractions 3, 2, 1 n Fraction of individuals showing an antibody rise of 2.0 pg. AbN/ml. or more. AbN/ml., and makes no attempt to consider that some of them may have reacted anamnestically as is found to occur with dextrans (2). A referee thought that an anamnestic response of those individuals with 0.8 pg. AbN/ml. or more might introduce a bias in favor of groups 6, 5, and 4 as compared to groups 3,2, and 1 since the former group was more heavily weighted with individuals with antibody levels of 0.8 kg. N/ml. or above. Examination of the data in Table I on this basis is given in Table III. It is evident that there is some suggestion of an anamnestic response in that a higher proportion of individuals who had an initial level of 0.8 pg. AbN/ml. or above showed a significant rise in antibody as compared with those which did not. However, in all instances, Frs. 6, 5, and 4 showed a greater immunizing capacity than did Frs. 3, 2, and 1, both as measured by induction of a possible anamnestic type of re- sponse or by primary immunizing capacity. Figure 1 presents quantitative precipitin curves with the sera of three subjects (173, 176, 216) who had produced antibody 60 dextran NRC Frs. 5,4, and 2, respectively, as compared with subject 36 who had received native dextran from the same strain of microorganism. Pre- cipitin curves with a clinical dextran N150N are also included. The over-all general similarity of the precipitin curves with the various NRC dextran fractions and with native dextran is evident. With all four antisera, NRC Fr. 1 precipitated substantially less antibody N than did any of the other fractions. In all three instances studied, N150N was intermediate in precipitating power between NRC Frs. 1 and 2. Wit,h antiserum 216D2 to NRC Fr. 2 and with antiserum 30D4 t*o nativr dextran, t>he ot,her dext,ran F‘rs. 2-8 were all very similnr in precipitat,ing
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 8
`
`

`

`314 KABAT AND BEZER 1.5ml 36 D4 ANTI-NATIVE 6512 407 r 0 0 5 10 15 eo 0 5 10 15 20 25 30 ~401.0ml I76 D2 ANTI-NRC #4& 4 r r 0 5 10 15 20 0 5 10 15 20 25 30 0 5 IO 15 20 25 30 MICROGRAMS OEXTRAN ADDED FIG. 1. Quantitative precipitin curves of native Dextran and Dextran fractions with various human antidextran sera. power and generally precipitated, within about 2 pg. AbN, the same quantity of antibody N as did the native dextran. With antiserum 176D2 to NRC Fr. 4, Fr. 2 precipitated only 29 pg. AbN/ml. while Frs. 3- 8 precipitated from 33 to 36 pg. AbN/ml. at the maximum. With anti- serum 173,3 to Fr. 5, dextrans NRC Frs. 2 and 3 were somewhat lower in precipitating power, removing a maximum of 15 and 17 pg. AbN/2.0 ml, as compared with 21-25 pg. for Frs. 4-8, respectively. With this
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 9
`
`

`

`-4NTIGENICITY OF DEXTRAN IN MAN 315 I.5ml 36 Dq ANTI- NATIVE 8512 L.UU / I I , I I 02 06 IO 14 I8 22 2.6 3.0 0 2 0.6 10 14 I.8 22 0.6 MlCROMOLES OL100SACCHAR1DE ADDED FIG. 2. Oligosaccharide inhibition studies with various human antidextran sera and clinical dextran N150N. antiserum, native dextran and Fr. 8 precipitated an additional 3-5 pg. AbN/2.0 ml. Figure 2 summarizes the results of oligosaccharide inhibition studies with isomaltotriose, -tetraose, -pentaose, and -hexaose [cf. (14)] using clinical dextran N150N to precipitate the antibody under the conditions used in previous studies (13, 14). It is evident that the ratios of the inhibiting power for these four oligosaccharides to one another vary for the three antisera studied. Thus with antiserum 21SD3 to the lowest molecular weight fraction NRC Fr. 2, the penta- and hexasaccharides are only about 2-3 times more active in inhibiting than is the trisac- charide; with antiserum 176D2 to Fr. 4, and antiserum 36,,4 to native dextran, the hexasaccharide is about 16-18 times better than the tri- saccharide; these two sera differ, however, in that the tetrasaccharide is a much better inhibitor relative to the hexasaccharide in antiserum 36na than it is in antiserum 176,, 2 . DISCUSSION The findings indicate that the capacity of dextran to induce antibody formation in man under the conditions used drops off sharply when products with molecular weights of 51,300 or less are used as compared with products of 91,700 or above. The proportion of subjects lo/18 showing antibody response to products with molecular weights of 91,700 or above is not significantly different from that previously obtained for native dextrans B512, N236 and N279 (2) from which 9/18 individuals showed an antibody rise 2 pg. AbN/ml. or more. Thus, although limited
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 10
`
`

`

`316 KABAT AND BEZER numbers of individuals were necessarily used in these studies, it would appear that the hydrolytic procedure used to lower the molecular weight of the dextran does not affect the capacity of the products to induce an antibody response in humans until the molecular weight is reduced to less than 91,700. Products prepared from two samples of British dextran, Intradex or Dextraven, which have molecular weights of about 200,000 as well as samples APC No. 55 and APC No. 54 also showed a similar capacity to induce antibody formation. With respect to the capacity of the dextrans used to elicit a wheal and erythema response on intracutaneous injection of lo-20 pg. dextran, it appears that an antibody response of about 7 pg. AbN/ml. or above was required with the NRC fractions; this generally held true for the other dextrans, but with clinical Dextraven, one individual with 7 pg. AbN/ ml. had a negative skin test on two occasions, so that this might well represent the borderline region of skin sensitivity. It is of interest to compare these findings with those obtained in individuals immunized with the native dextrans from the same B512 strain (2) in which a skin reaction to native dextran could be elicited when the circulating pre- cipitin was as low as 3 pg. AbN/ml. These findings further substantiate earlier results (17) that clinical dextrans gave a much lower proportion of wheal and erythema reactions than did native dextrans when tested in a population which had never received an injection of dextran. This suggests that, on hydrolysis of native dextran, the capacity to induce a wheal and erythema reaction decreases before the capacity to induce antibody formation is appreciably affected. Moreover, the capacity to stimulate antibody formation is sharply reduced although the ability to precipitate antidextran is only relatively slightly affected. Since it is impossible to prepare completely homogeneous dextrans, the possibility that any single individual may be reacting to a small proportion of the 1 mg. dextran injected cannot be ruled out; thus subject No. 216 could conceivably have responded to a very small quantity of material of higher molecular weight. The quantitative precipitin curves in Fig. 1 show slight differences in the capacity of the various fractions and native dextran to precipitate antidextran produced to native dextran and to NRC Frs. 5, 4, and 2. However, these differences do not appear to correlate with molecular weight, since the antiserum 36 D4 to native dextran and 21SD3 from the in- dividual who responded to the fraction of lowest molecular weight (NRC- Fr. 2) gave very similar precipitin curves with the various fractions. That N150N is poorer in precipitating power than Frs. 2 and 3, although
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 11
`
`

`

`BNTIGENICITY OF DEXTRAN IN MAS 317 it has a higher average molecular weight, is probably due to its greater degree of polydispersity. To determine whether the differences in hetero- geneity in extent of the antibody-combining sites (14) could be respons- ible for the similarities and differences in the quantitative precipitin curves, the oligosaccharide inhibition studies in Fig. 2 were carried out. The data, however, show that despite their similarity in precipitin curves (Fig. 1) antiserum 36D4 and antiserum 216D3 were quite different. With antiserum 36D~ about 13 times as much isomaltotriose as isomaltohexnose on a molar basis was required for 50 % inhibition, while with antiserum 216D3 only about 2-3 times as much was required. In this respect anti- serum 216D3 resembled antiserum 30D4 previously reported (14). Anti- serum 176,2 was much more similar to antiserum 36D~ , although in the quantitative precipitin curves (Fig. 1) NRC Fr. 2 precipitated somewhat less antibody than did NRC Fr. 3 from the former antiserum, while with the latter antiserum they were equal in precipitating power. It thus does not appear possible at present to account for the slight differences in quantitative precipitin curves obtained from one serum to another; these differences could conceivably be due to slight differences in the solubility of dextran-antidextran specific precipitates formed with t’hr sera of different, individuals. ACKNOWLEDGMENT The authors are indebted to Dr. John W. Fertig for helpful dimwions on the statistical evaluation of the data. SUMMARY The capacity of a series of dextran fractions of graded molecular weight. t,o stimulate antibody formation in man upon injection of 1 mg. dextran was investigated. A significant drop in capacity to elicit antibody forma- tion was found with dextran fractions of average molecular weight of 51,300 or below. The precipitating antibodies to the dextran fractions were generally similar to those produced to native dextran as determined by quantitative precipitin and oligosaccharide inhibition studies. Wheal and erythema reactions to the dextran fractions were obtained when circulating precipitin levels were above 7-8 pg. antibody N/ml. REFERENCES 1. KABAT, E. A., AND BERG, D., Ann. N. Y. Acad. Sci. 66, 471 (1952). 2. KABAT, E. A., AND BERG, D., J. Immunol. 70, 514 (1953). 3. MAURER, P. H., Proc. Sot. Ezptl. Riol. Med. 83, 879 (1953).
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 12
`
`

`

`318 RABAT AND BEZER 4. ALLEN, P. Z., AND KABAT, E. A., J. Exptl. Med. 106, 383 (1957). 5. GRBNWALL, A., AND INGELMAN, B., Nord. Med. 21, 247 (1944); Acta Physiol. Stand. 7, 97 (1944). 6. WEISSBERQ, S. G., AND ISBELL, H. S., Natl. BUY. Standards (U. S.) Rept. No. 1130 (1951); 1713 (1952). 7. HELLMAN, N. N., TSUCHIYA, H. M., ROGOVIN, S. P., LAMBERTS, B. L., TOBIN, R., GLASS, C. A., STRINGER, C. S., JACKSON, R. W., AND SENTI, F. R., Ind. Eng. Chem. 47, 1593 (1955). 8. TSUCHIYA, H. M., HELLMAN, N. N., KOEPSELL, H. J., CORMAN, J., STRINQER, C. S., ROGOVIN, S. P., BOGARD, N. O., BRYANT, G., FEOER, V. H., HOFF- MAN, C. A., SENTI, F. R., AND JACKSON, R. W., J. Am. Chem. Sot. 77, 2412 (1955). 9. HEHRE, E. J., Bacterial. Proc. p. 23 (1952); J. Biol. Chem. 222, 739 (1956) (with a note by F. R. Senti and N. N. Hellman, p. 747). 10. JEANES, A., HAYNES, W. C., WILHAM, C. A., RANKIN, J. C., MELVIN, E. H., AUSTIN, M. J., CLUSKEY, J. E., FISHER, B. E., TSUCHIYA, H. M., AND RIST, C. E., J. Am. Chem. Sot. 76, 5041 (1954). 11. HEIDELBERGER, M., AND MACPKERSON, C. F. C., Science 97,405; 98,63 (1943). 12. KABAT, E. A., AND MAYER, M. M., “Experimental Immunochemistry.” Chas. C Thomas, Springfield, Ill., 1948. 13. KABAT, E. A., J. Am. Chem. Sot. 76, 3709 (1954). 14. KABAT, E. A., J. Immunol. 77,377 (1956); J. Cellular Comp. Physiol. 60 Suppl. No. 1, 79 (1957). 15. JEANES, A., WILHAM, R. W., TSUCHIYA, H. M., AND RIST, C. E., J. Am. Chem. Sot. 76, 5911 (1953). 16. TURVEY, J. R., AND WHELAN, W. J., Biochem. J. 67, 49 (1957). 17. KABAT, E. A., TURINO, G. M., TARROW, A., AND MAURER, P. H., J. Clin. In- vest. 36, 1160 (1957).
`
`PGR2020-00009
`Pharmacosmos A/S v. American Regent, Inc.
`Petitioner Ex. 1050 - Page 13
`
`