BASIC SCIENCES

Relative Reactivity of DTP A, Immunoreactive Antibody-DTPA Conjugates, and Nonimmunoreactive Antibody-DTPA Conjugates Toward Indium-111 Chang H. Paik, Jack J. Hong, Mercedes A. Ebbert, Susan C. Heald, Richard C. Reba, and William C. Eckelman*

Radiopharmaceutical/Medicinal Chemistry Section, The George Washington University Medical Center, Washington, DC

Anti-human serum albumin antibody {Ab) was reacted with cyclic DTPA dianhydride (cDTPAA) at various cDTPAA/Ab molar ratios between 1 and 40. Using a carrier In titration method for DTPA and DTPA-antibody conjugate (Ab-DTPA), we determined that the above reactions produced between 0.1 and 11 DTPA molecules per either immunoreactive antibody (sAb) or nonimmunoreactive antibody (nAb). The percentage of sAb remaining after the above reactions was between 88 and 62%. The reaction of no-carrier-added 111ln with the reaction mixture from cDTPAA/Ab molar ratios of 1 to 40 gave radiochemical yields <25% for the respective Ab-DTPA. The rest of the 111ln activity was associated with free DTPA. Our results indicate that Ab-DTPA containing > 1 DTPA molecule per Ab is more reactive than that containing <1 DTPA but is about as reactive as free DTPA. This allows us to label in the presence of free DTPA and consequently prevent colloid formation. The percentage of 111ln activity incorporated into sAb-DTPA from the reactions at these molar ratios was similar to that found from the uv analysis. This indicates that the reactivity of sAb-DTPA and nAB-DTPA from the same conjugation reaction is similar. As a result, we were able to conjugate about one DTPA molecule to the Ab without causing deactivation of the Ab and label it with 111ln in the presence of excess DTPA. We obtained a specific activity of 6 /JLC\ 111ln per /ug of Ab using research grade 111ln without further purification.

J NucI Med 26:482-487,1985

JL-/iethylcnetriaminepentaacetic acid (DTPA) can be conjugated to antibodies by way of three different ac-ylation methods using DTPA carboxycarbonic anhy­dride (1-4), cyclic DTPA dianhydride {5-8), and the O-acylisourea of DTPA (9). DTPA conjugated anti­bodies may be labeled with metallic radionuclides such as indium-111 ( n i ln) and technetium-99m (99mTc) for biodistribution studies or radioimmunodetection of tu­mors. We have studied two radiolabeling variables,

Received Jan. 3, 1984; revision accepted Feb. 4, 1985. For reprints contact. Chang H. Paik, PhD, 2300 1. Street NW,

Washington, DC 20037. * Present address: Department of Nuclear Medicine, Clinical Center, National Institutes of Health, Bethesda. MD 20205.

specific activity and radiochemical purity, in order to maximize the incorporation of U1ln. The theoretical specific activity of i n In-labeled antibody conjugates containing one DTPA molecule per Ab is 324 juCi//ig Ab (mol/wt = 150,000) assuming the formation of a 1:1 complex of n * In and DTPA. However, an experimen­tally obtained specific activity is, in general, about 300 times less than the theoretical specific activity primarily because of trace amounts of metallic ion contaminants and unfavorable kinetics of reactions where the con­centration of DTPA conjugated antibody is dilute. In general, DTPA antibody conjugates are separated from free DTPA using a Sephadex G-50 column and the di­luted DTPA antibody conjugate is concentrated before

482 Paik. Hong, Ebbert et al The Journal of Nuclear Medicine

by on December 30, 2014. For personal use only. jnm.snmjournals.org Downloaded from

labeling with l n I n . However, each additional experi­mental step increases the possibility of metallic ion contamination and consequently results in a lower spe­cific activity. In this respect, information on the reac­tivity of free DTPA and antibody conjugated DTPA (Ab-DTPA) toward l u I n should help us design an ex­periment to label Ab-DTPA at maximum specific ac­tivity in the presence of free DTPA. The second point addressed in this paper is the relative reactivity of non-immunoreactive antibody DTPA conjugate (nAb-DTPA) and immunoreactive antibody DTPA conjugate (sAb-DTPA) toward i n In . This information is espe­cially important for tumor-specific antibody DTPA conjugates which cannot be separated from nAb-DTPA before radiolabeling due to the lack of an appropriate affinity column.

MATERIALS AND METHODS

Conjugation of DTPA to Ab Affinity purified anti-human serum albumin (Ab) was

used for this experiment (4). Cyclic DTPA dianhydride was synthesized according to the method described previously (5,10). For the conjugation reactions at cDTPAA/Ab ratios of 1 and 5,2.1 and 10.5 mg each of cDTPAA were dissolved in 5 ml of dry dimethyl sulf­oxide. Five microliters of the above solution were im­mediately added to 100 fi\ of Ab at 5.8 X \0~5M (8.7 mg/ml) in 0. \M bicarbonate at pH 8.3. The solution was shaken gently and allowed to stand at 23°C for 1 hr. For the reactions at cDTPAA/Ab ratios of 20 and 40, 8.3 and 16.6 mg each of cDTPAA were dissolved in 1 ml of dry DMSO. The rest of the procedure was the same as for the ratios of 1 and 5.

Determination of the number of DTPA molecules per Ab

Twenty-five microliters of the conjugated antibody solution were diluted to 1 ml with 0.1 M bicarbonate buffer. The pH of the solution was reduced to 5.3 by addition of 0.4 ml of 0.1M citrate buffer. To this solution was added 0.2 ml of InCl3* (99.99% pure) containing a tracer amount ofl ] 1InCl3

t in 0.1M citrate buffer at pH 6. The final concentration of lnCl3 was about twice that of the total DTPA concentration to saturate free DTPA as well as the conjugated DTPA. The above solution was incubated for 30 min at 23°C. Unconsumed indium ions were then complexed with 10 fi\ of 0.1 M DTPA to pre­vent the formation of indium hydroxide upon the neu­tralization of the solution before affinity-gel filtration chromatography.

Nonimmunoreactive antibody DTPA conjugate (nAb-DTPA), free DTPA, and immunoreactive anti­body DTPA conjugate (sAb-DTPA) were separated by affinity-gel filtration chromatography as described (5). The percent immunoreactive and nonimmunoreactive

antibodies were calculated based on the uv peak inten­sity. The average number of indium atoms incorporated into each antibody fraction was calculated based on the 11 *In activity associated with the fraction, and the spe­cific activity of the labeling solution. Research grade [niIn]chloride was used for this study without purifi­cation.

Determination of relative reactivities at no-carrier-added i n In concentrations

Ten microliters of each reaction solution from cDTPAA/Ab ratios of 1, 5, 20, and 40 were mixed with 20 jul (~200 fid) of m InCl 3 in 0.03M citrate buffer. The pH of the final solution was 5.3. The solution was incubated for 30 min at 23°C. Twenty microcuries of the above Ab-DTPA-1 n In solution were diluted to 100 fil with 0.03Af citrate buffer at pH 5.3. About 1 jiCi of this solution was spotted on a silica gel thin layer chroma­tographic (TLC) plate* and developed with 2:2:1 10% ammonium formate in water:methanol:0.5A/ citric acid. The n ' In activity was detected11 and recorded on a strip chart paper. The above solvent system gave Rf values of Ab-DTPA-1'l In and DTPA-1 »» In at 0 and 0.6, re­spectively. The uncomplexed J11 In prior to the com-plexation reaction was shown at a Rf value of 0.8. The 1]1In complexation reactions were complete in 30 min as indicated by the absence of the peak at Rf at 0.8. The reaction solution containing 4 //Ci of DTPA-111 In, nAb-DTPA-1 n In , and sAb-DTPA-nlIn was then di­luted with 150 fig of Ab stock solution in 1 ml of 0.02M phosphate buffer-0.5Af NaCl solution at pH 7.6. This solution was then eluted through the affinity-gel filtra­tion column. The percent' n In activity incorporated to nAb-DTPA, sAb-DTPA, and free DTPA was deter­mined based on l n I n activity associated with each fraction.

RESULTS

Determination of DTPA molecules conjugated to Ab The number of DTPA molecules conjugated to Ab is

proportional to the concentration of cDTPAA and Ab. Thus, the reaction of 5.8 X 10~5M Ab with cDTPAA at cDTPAA/Ab molar ratios of 1,5,20, and 40 gave rise to 0.13, 1.4,6.4, and 10.7 DTPA molecules per sAb. The number of DTPA molecules conjugated per nAb for the same reaction conditions was 0.14, 1.7, 5.4, and 10.8, respectively. These reactions reduced the immunoreac-tivity to 88,87,73, and 61% from 91% for Ab before the reaction (Table 1). The percent immunoreactivity was determined based on the peak intensity of sAb and nAb at 280 nm.

Determination of the relative reactivity using m I n The relative reactivity of free DTPA, sAb-DTPA, and

nAb-DTPA fractions was determined by measuring

Volume 26 • Number 5 • May 1985 483

by on December 30, 2014. For personal use only. jnm.snmjournals.org Downloaded from

TABLE 1 DTPA Conjugation to Ab from Reaction at Various

cDTPAA to Ab Ratios

TABLE 3 Percent Labeling of Antibody DTPA Conjugate with No-

Carrier-Added 111ln

cDTPAA Ab

% sAb uv method

In atoms In atoms In atoms cDTPAA % Ab-DTPA-111ln

sAb

1 5

20 40

91.0 ± 3.0' — 88.2 ±3.3 0.13 ±0.03 87.3 ±3.1 1.4 ±0.3 73.0 ±5.8 6.4 ±1.4 61.7 ±4.9 10.7 ±2.4

nAb Total Ab Ab % Active Ab* Aff. Chromt TLC*

_ _ 91.0±3.0§ _ 0.14 ± 0.03 0.13 1 84.2 ± 3.4 5.5 ± 3.7 7.9 ± 4.2

1.7 ±0.2 1.5 5 77.0 ± 5.8 25.4 ± 8.9 37.9 ± 6.5 5.4 ± 2.4 6.1 20 64.2 ± 8.6 24.1 ± 6.3 46.3 ± 10.2

10.8 ± 1.6 10.7 40 45.7 ± 9.9 25.0 ± 9.8 39.3 ± 7.2

' Percent of active antibody in the stock anti-human serum al­bumin antibody solution. Each value represents average of three to six determinations and its standard deviation.

no-carrier-added ,MIn activity associated with each DTPA fraction separated by the Sepharose-HSA/ Sephadex G-50 combination column. The percentage of no-carrier-added ' ' ' In-labeled to the total Ab-DTPA was 6, 25, 24, and 25%, respectively. This suggests that the number of DTPA molecules available in Ab-DTPA for ] " In in the presence of free DTPA is 0.06, 1.2,4.8, and 10.0. Comparing these numbers with the number of DTPA molecules per Ab determined by the saturation experiment with lnCl3 containing a tracer amount of " 'In, one can conclude that the reactivities of saturable (or titrable) DTPA moieties of Ab-DTPA range from 46, 83, 78 to 92% of the reactivity of free DTPA (Table 2). The percentage of no-carrier-added ' "In associated with sAb-DTPA was 84, 77, 64, and 46% of total Ab-DTPA-'"In (Table 3). These percentages are similar to the percentages obtained from the uv analysis for sAb-DTPA and nAb-DTPA. This indicates that the reactivities of sAb-DTPA and nAb-DTPA toward no-carrier-added ' "In are about the same.

The percentages of no-carrier-added " ' In associated with Ab-DTPA obtained from a TLC analysis were 8,

* Based on 111ln activity associated with sAb from affinity chromatography.

t Percent of111in activity incorporated into total antibody DTPA conjugate when reacted with product mixture. Percent was calcu­lated by determining 111ln radioactivity associated with antibody fractions from affinity-gel filtration column. All residual 111ln was associated with free DTPA.

* Percent by TLC analysis. § Percent of active antibody in stock solution.

38,46, and 39%, respectively (Table 3). These numbers are larger than those obtained from the combination column chromatography. This might be caused by en­trapment of DTPA-" 'In by antibody when the protein is coagulated by the solvent system.

High specific activity labeling The product mixtures from the reactions of Ab with

cDTPAA at a cDTPAA/Ab ratio of 3 were labeled with " ! In ions at three different activities in order to find an optimum condition for the higher specific activity la­beling of antibody. The percent Ab-DTPA-"'In re­mained the same, 46 and 53%, respectively, when the product mixture from the reaction of ll .4 mg/ml Ab was labeled with > "In ions at the activity of 2 and 10 ^Ci per Mg of Ab. However, the labeling yield decreased drasti­cally to 27% when 20 fid per ^g of the Ab was reacted

TABLE 2 Relative Reactivities of DTPA and Titrable Ab-DTPA

Toward 111ln

Ab-DTPA* Relative cDTPAA Ab-DTPA cDTPAA Ab-DTPA-111lnT reactivity*

Ab Ab (%) (%) (%)

1 5

20 40

0.13 1.5 6.1

10.7

13 30 30 27

6 25 24 25

46 83 80 93

TABLE 4 Effect of Metallic Ion Impurities on Distribution of 111ln

Between Ab-DTPA and DTPA* 111ln Ab-DTPA- DTPA- Specific

[Abl cDTPAA activity 111ln 111ln activity mg/ml Ab fiCVug Ab (%) (%) lxC\/fig Ab

11.4 3 11.4 3 11.4 3 21.9 3 21.9 3 21.9 3

2.0 10.4 22.0

2.0 10.2 21.3

46.4 53.3 26.9 61.1 59.0 35.2

53.6 46.7 73.1 39.9 41.0 64.8

0.91 5.6 5.9 1.2 6.0 7.4

* Fractional yield of DTPA conjugation to Ab; obtained by dividing column two by column one.

1 No-carrier-added labeling yield. * Reactivity of titrable Ab-DTPA relative to DTPA.

* 111lnCl3 was research grade purchased from Medi-Physics, Inc. and used without purification.

t Specific activity of 111ln-labe!ed Ab-DTPA. Radiochemical yield was determined by TLC analysis.

484 Paik, Hong, Ebbert et al The Journal of Nuclear Medicine

by on December 30, 2014. For personal use only. jnm.snmjournals.org Downloaded from

o A -

2 -

- 100

- 80

- 40

(cDTPAA/Ab)

FIGURE 1 Relationships between cDTPAA/Ab molar ratio, number of DTPA mole­cules conjugated per Ab ( D - d ) and percent immunoreactivity of DTPA conjugated Ab (0 -0 : uv method, A-A: 11 inactivity)

(Table 4). The same trend was observed when the product mixture from the reaction of 21.9 mg/ml Ab with cDTPAA was reacted with m l n ions at the in­creasing activity. The highest specific activity obtained from the above conditions was about 6 juCi per jug of Ab.

DISCUSSION

We have reported factors influencing DTPA conju­gation to antibodies by way of three different acylation reactions and, therefore, the optimum conditions for conjugation reactions (4,5,9). We report in this paper important variables which influence the radiochemical yield. Ab-DTPA containing a large number of DTPA molecules per Ab is preferred for high specific activity labeling. On the other hand, there is an inverse rela­tionship between DTPA molecules conjugated per Ab and the antibody-binding activity. Other important questions to be answered are: What are the relative reactivities between free DTPA and Ab-DTPA, and between nAb-DTPA and sAb-DTPA toward l nIn? The first is important because free DTPA contaminating the Ab-DTPA solution might take a l l ' n In if it is much more reactive than Ab-DTPA or in higher concentration. The second question is important, especially for labeling tumor specific antibodies if appropriate affinity columns are not available to separate immunoreactive antibodies from nonimmunoreactive antibodies after the DTPA conjugation reactions.

In order to answer the above questions, Ab at 8.7 mg/ml was reacted with cDTPAA at various cDTPAA/Ab molar ratios. The results indicate that the number of DTPA molecules conjugated per sAb and

nAb from the same reaction are similar (Table 1). In­creasing the number of molecules conjugated results in increasing deactivation of the antibody. This agrees with our previous result (5) and results of Hnatowich et al. (6). This might result from an increased probability of modifying the sensitive regions for antigen binding as the number of DTPA conjugations increases. However, the inverse relationship of DTPA conjugation to immuno­reactivity of Ab is inconsistent with the hypothesis that nAb contains a larger number of DTPA molecules per antibody than sAb.

The DTPA conjugation yields, when determined by titration with carrier-added ' *' In, were 30, 30, and 27% for the reaction at cDTPAA/Ab ratios of 5, 20, and 40, respectively. These yields are equivalent to 1.5, 6.1, and 10.7 DTPA molecules conjugated per Ab (Table 2, Fig. 1). However, the conjugation yield of the reaction at the ratio of 1 was only 13%. Since we chose reaction condi­tions such that the concentration of the amino groups of lysine residues was larger than the concentration of cDTPAA (antibody contains about 90 lysine residues), it is unlikely that the conjugation yield is sensitive to the variation of the cDTPAA concentration. The difference might occur because the percentage of DTPA molecules exposed to indium ions in aqueous medium decreases drastically as the number of DTPA molecules conju­gated per Ab is less than one. Thus, the fraction of DTPA molecules available for the reaction with indium ions is much smaller. Our results are different from those of Hnatowich et al. (7). They reported a steep inverse re­lationship between the conjugation yield and the con­centration of cDTPAA at cDTPAA/Ab ratio of 1 to 10.

The complexation reaction of no-carrier-added M1In

Volume 26 • Number 5 • May 1985 485

by on December 30, 2014. For personal use only. jnm.snmjournals.org Downloaded from

FIGURE 2 Relationships between cDTPAA/Ab molar ratio, percent conjugation yield (A-A) and percent11 'In-labeling yiefd (0-0)

- 30

— I 20

(cDTPAA/Ab)

with the product mixtures from the reactions at cDTPAA/Ab ratios of 1, 5, 20, and 40 resulted in Ab-DTPA-1 »' In yields of 6,25,24, and 25%. The difference in the radiochemical yields reflects the difference in re­activity. The reactivity difference also seems to be caused by the difference in the fractional concentration available for n i I n ions. Comparing the yields from the carrier-added titration method and the no-carrier-added method, the reactivities of saturable (or titrable) DTPA moieties of Ab-DTPA range from 46, 83, 78 to 92% of the reactivity of free DTPA (Table 2, Fig. 2).

The proportions of sAb-DTPA and nAb-DTPA de­termined by uv analysis (Table 1, Fig. 1) are similar to those determined by no-carrier-added M ] In activity as­sociated with sAb-DTPA and nAb-DTPA (Table 3, Fig. 1). This suggesis that the reactivities of nAb-DTPA and sAb-DTPA from the same conjugation reactions are about the same for the complexation with ' ' ' In. One can conclude, therefore, that the contamination of free DTPA and nAb-DTPA would interfere with radiola-beling of sAb-DTPA in proportion to the relative con­centration when Ab-DTPA containing more than one DTPA molecule per Ab is labeled. The presence of free DTPA is important to prevent colloid formation.

We also studied the labeling conditions favorable to obtain a high specific activity ' ' ' In distribution between Ab-DTPA and free DTPA. When the product mixture from the reaction of Ab (11.4 mg/ml) with cDTPAA at cDTPAA/Ab molar ratio of 3 was reacted with 'Uln ions at the activity of 2 and lO^Ci per /ug of Ab, the la­beling yields of Ab-DTPA were about the same, 46 and 53%, respectively. However, the labeling yield decreased drastically to 27%. when 20 fxd of''' In per ^g of Ab was reacted (Table 4). The same trend was observed when the product mixture from the reaction of 21.9 mg/ml Ab with cDTPAA at the ratio of 3 was reacted with l l lIn ions at the increasing activity. This might occur because of metallic ion impurities which react preferentially with

Ab-DTPA and thereby reduce the concentration of Ab-DTPA, relative to free DTPA, available for H1In ions. The highest specific activity obtained from the above conditions was about 6 ^Ci per fig of Ab. Since this specific activity is sufficient for in vivo studies, we did not complicate the radiolabeling procedure with an additional purification step.

The reported data should simplify the experimental steps for radiolabeling. The removal of DTPA-1' 'In and nAb-DTPA-,,!In can be performed at the end of the experiment. The simplification of experimental steps is desirable especially when the mass of antibody available for radiolabeling is about 100 fig. It is important to ap­preciate that the molecular permeation chromatography to remove DTPA causes not only a loss of antibody but also a contamination of the Ab-DTPA solution with metallic ion impurities. A loss of antibody and contam­ination with even a minute amount of metallic impurities at this antibody and DTPA concentration can adversely affect the radiolabeling yield. Likewise, the presence of free DTPA during the radiolabeling procedure helps prevent colloid formation.

This radiolabeling approach should facilitate the production of radiochemical^ pure labeled antibodies.

FOOTNOTES

* Aldrich Chemical Co., Inc., Milwaukee, WI. f Medi-Physics, Inc., Richmond, CA. * MN Polygram, Brinkmann Instruments, Inc., Cantiague

Road, Westbury, NY 11590. 1 Packard Model 7220/21 Radiochromatogram Scanner.

ACKNOWLEDGMENTS

This work was supported in part by CA28462 awarded by the National Cancer Institute and by American Cancer Society Contract PDT-235.

4 8 6 Paik, Hong, Ebbert et al The Journal of Nuclear Medicine

by on December 30, 2014. For personal use only. jnm.snmjournals.org Downloaded from

REFERENCES

1. Krejcarek GE, Tucker KL: Covalent attachment of chelating groups to macromolecules. Biochem Biophys Res Commun 77:581-585,1977

2. Khaw BA, Fallon JT, Strauss HW, et al: Myocardial infarct imaging of antibodies to canine cardiac myosin with indium-lll-diethylenetriaminepentacetic acid. Science 209:295-297, 1980

3. Halpern SE, Hagan PL, Garver PR, et al: Comparison of In-111 anti-CEA monoclonal antibodies and endog-enously labeled Se-75 monoclonal antibodies in normal and tumor bearing mice. J Nucl Med 23: P8, 1982 (abstr)

4. Paik CH, Murphy PR, Eckelman WC, et al: Optimiza­tion of the DTPA mixed anhydride reaction with anti­bodies at low concentration. J Nucl Med 24:932-936, 1983

5. Paik CH, Ebbert MA, Murphy PR, et al: Factors in­fluencing DTPA conjugation to antibodies via cyclic DTPA anhydride. J Nucl Med 24:1158-1163, 1983

6. Hnatowich DT, Layne WW. Childs RL, et al: Radio­active labeling of antibody: A simple and efficient method. Science 220:613-615, 1983

7. Hnatowich DT, Childs RL, Lanteigne D, et al: The preparation of DTPA-coupled antibodies radiolabeled with metallic radionuclides: An improved method. J Immunol Methods 65:147-157, 1983

8. Wang TST, Srivastava SC, Fawwaz RA, et al: Prepa­ration of In-111 labeled monoclonal antibody to high molecular-weight melanoma associated antigen: A comparative study using the cyclic anhydride of DTPA vs the mixed anhydride of DTPA for antibody chelation. J Nucl Med 24: PI 25, 1983 (abstr)

9. Paik CH, Lassman CR, Murphy PR, et al: A carbodi-imide method for DTPA conjugation to antibodies. Hybridoma 2:248, 1938 (abstr)

10. Eckelman WC, Karesh SM, Reba RC: Fatty acid and long chain hydrocarbon derivatives containing a strong chelating agent. J Pharm Sci 64:704-706, 1975

Volume 26 • Number 5 • Mav 1985 487

by on December 30, 2014. For personal use only. jnm.snmjournals.org Downloaded from

1985;26:482-487.J Nucl Med.   EckelmanChang H. Paik, Jack J. Hong, Mercedes A. Ebbert, Susan C. Heald, Richard C. Reba and William C.  and Nonimmunoreactive Antibody-DTPA Conjugates Toward Indium-111Relative Reactivity of DTPA, Immunoreactive Antibody-DTPA Conjugates,

http://jnm.snmjournals.org/content/26/5/482This article and updated information are available at:

  http://jnm.snmjournals.org/site/subscriptions/online.xhtml

Information about subscriptions to JNM can be found at:  

http://jnm.snmjournals.org/site/misc/permission.xhtmlInformation about reproducing figures, tables, or other portions of this article can be found online at:

(Print ISSN: 0161-5505, Online ISSN: 2159-662X)1850 Samuel Morse Drive, Reston, VA 20190.SNMMI | Society of Nuclear Medicine and Molecular Imaging

is published monthly.The Journal of Nuclear Medicine

© Copyright 1985 SNMMI; all rights reserved.

by on December 30, 2014. For personal use only. jnm.snmjournals.org Downloaded from