Biochhemicnt
`Pharmacology, Vol. 42. No. 4. pp. 889-897, 1991. CXKC2952/91 $3.00 + 0.00 Printed in Great Britain. 0 1991. Pergamon Press plc
`
`MECHANISM OF DEIODINATION OF ‘251-HUMAN
`GROWTH HORMONE IN VZVO
`
`RELEVANCE TO THE STUDY OF PROTEIN DISPOSITION
`
`Department of
`
`Drug Metabolism and Disposition, Lilly Research Laboratories, Eli Lilly & Company, Indianapolis, IN 46285, U.S.A.
`
`VICTOR J. WROBLEWSKI
`
`(Received 4 December 1990; accepted
`
`March
`
`1991)
`
`uioo
`Abstract-Examination of the disposition of proteins employing ‘251-labeled tracers can be complicated by the in
`in oiuo
`deiodination of the tracer. The purpose of this study was to characterize the mechanism by which ‘251-labeled proteins are deiodinated
`
`using “‘I-human growth hormone (hGH) as a model compound. Intravenous (i.v.) administration of ‘2SI-hGH resulted in a biphasic plasma kinetic pattern, with the majority of radioactivity removed from the plasma during the first 15 min. The level of circulating radioactivity at 2 hr was similar to that 15 min after administration. Radioactivity was eliminated from the animals almost exclusively in the urine. The chemical form of radioactivity in the plasma and urine was analyzed by HPLC, and precipitation of radioactivity with silver nitrate or trichloroacetic acid. Fifteen minutes after administration of ‘251-hGH, 30% of the circulating radioactivity was present in the form of iodide (1251-). By 2 hr, the majority of radioactivity in the plasma was in the form of ‘251m. The radioactivity in the urine was present exclusively in the form of ‘251-. In uiuo deiodination of ‘251-hGH was reflected by the accumulation of radioactivity in the thyroid glands. There was no evidence for the presence of lz51- peptide intermediates in the plasma or urine of treated animals. In
`metabolic product. However, in the presence of dithiothreitol and NADPH as cofactors, the predominant metabolic product formed by thyroid gland homogenates was ‘251~, The deiodination of ‘Z51-hGH by thyroid gland homogenates was inhibited by the serine protease inhibitor phenylmethylsulfonyl fluoride (PMSF), indicating that proteolysis of “‘1-hGH was required for de- iodination tb occur: This was supported by the observation that ‘251-labeled proteolytic- fragments of ‘251-hGH. but not ‘*51-hGH, were deiodinated bv liver or kidnev homoeenates in the oresence of these cofactors.’ Deiodination by thyroid gland homogenates was inhibited bi the sulfhydryl-group blocking reagent, iodoacetate, in aconcentration-dependent manner. Thecharacteristicsof the
`
`‘251-hGH was degraded to ‘251- peptide intermediates by thyroid gland but not liver or kidney homogenates. In the absence of cofactors, 1251- was not observed as an
`
`in vitro
`
`uitro,
`
`in vitro
`
`deiodination reaction suggest that a form of thyronine .5’-monodeiodinase may be involved in the in uiuo deiodination of ‘251-hGH and possibly other 1251-proteins. These data suggest that the disposition of proteins may be determined more accurately with 3H-, 14C- or 35S-labeled molecules which better represent the charac- teristics of the native protein. Recombinant technology has allowed the large-scale production of highly pure protein molecules which had previously been available in small quantities with questionable purity. Naturally occurring pro- teins such as human insulin, growth hormone and tissue plasminogen activator are now being modified in ways which can provide more desirable kinetic and pharmacologic profiles [l-3]. The development of endogenous and modified proteins as phar- maceutical agents has led to uncertainties regarding the appropriate procedures to study their pharma- cokinetics and disposition. The plasma kinetics of exogenously administered proteins have been commonly examined by radio- immunoassay, employing polyclonal or monospecific antibodies to the protein of interest [4-71. Tissue distribution, degradation and excretion studies involving proteins have typically employed lz51- labeled molecules as tracers. Previous studies have examined the disposition of an array of proteins and peptides utilizing 1251-labeled material in combination with qualitative measures of protein degradation such as solubility of label in trichloroacetic acid or immunoprecipitation [8-l 11. These studies have provided basic information on the disposition of proteins, but suffer from the assumption that the fate of the labeled molecule accurately reflects that of the unlabeled material. Labeling with iodine involves addition of an atom which is not part of the structure of the natural molecule and provides a potential point of metabolism not present in the unlabeled molecule. In
`
`uiuo,
`
`1251-proteins appear to undergo deiodination which may result from destabilization of the iodine by the adjacent hydroxyl group on the phenyl ring or the ability of iodinated tyrosyl residues to act as substrates for deiodinases of thyroid hormones [ll-141. The mechanism by which deiodination occurs and how deiodination impacts upon the interpretation of studies employing iodinated tracers has not been directly addressed. The purpose of this study was to characterize the fate of 1251-human growth hormone (1251-hGH) in an attem t to more clearly define the mechanism by which !? ’ 51-labeled proteins are deiodinated in vivo.
`
`Materials
`
`METHODS
`
`1251-Biosynthetic human growth hormone (hGH*) BP 42:4-L 889
`
`Page 1 of 9
`
`YEDA EXHIBIT NO. 2079
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`18
`

`
`Assays for degradation products of “‘I-hGH
`Precipitation with trichloroacetic acid
`
`(TCA).
`
`Precipitation with silver nitrate.
`
`890
`
`V.
`
`studies
`Plasma kinetics and tissue accumulation.
`
`J. WROBLEWSKI was prepared by the lactoperoxidase technique [15] and purified by gel filtration chromatography. The specific activity of the lz51-hGH used in these studies ranged from 40 to 55 pCi/pg and was diluted with unlabeled biosynthetic hGH (Lot 222-HG6, Eli Lilly & Company) in 0.9% saline prior to dosing. Unlabeled 3-iodo-L-tyrosine and dithiothreitol was purchased from the Sigma Chemical Co. (St. Louis, MO). In vivo
`Male Fischer 344 rats (Harlan, Indianapolis, IN), 210- 225 g, were administered 1251-biosynthetic human growth hormone (166 pg/kg, 12 @i/animal; or 20 pg/kg, 8.1 &i/animal) intravenously by tail vein. Blood was obtained by cardiac puncture at 1, 5, 15, 30, 60 and 120 min after injection, and plasma was prepared by centrifugation at 3000 rpm for 15 min at 4”. Thyroid glands and samples of liver were obtained 2, 15, 30, 60 and 120 min after injection. Levels of rz51-radioactivity were measured in a gamma counter. The data are expressed as counts per minute per gram tissue or percent of administered dose per gram tissue or milliliter plasma. The chemical nature of the labeled material in plasma was analyzed by size-exclusion HPLC (SE-HPLC) and reverse-phase HPLC (RP-HPLC).
`
`Routes of elimination.
`
`Male Fischer 344 rats, 210- 225 g, were administered 1251-hGH (60 pg/kg body wt, 8-10 pCi/animal) intravenously by tail vein. The animals were placed in metabolism cages with free access to food and water. Urine and feces were collected over a 24-hr time period and levels of radioactivity were quantified in a gamma counter. Labeled excretion products in the urine were characterized by SE-HPLC and RP-HPLC.
`Samples of plasma and urine (lO- 50 pL) were injected directly onto a Zorbax GF 250 column (9.2 x 250 mm). The column was eluted with 0.025 M ammonium bicarbonate buffer (pH 6.5) at 2mL/min. Column effluent was collected and radioactivity measured in a gamma counter, or radioactivity profiles were determined with the use of an on-line Ramona 5-LS (Raytrest, U.S.A.) radiochemical detector.
`
`Chromatographic analysis
`
`SE-HPLC.
`
`RP-HPLC.
`
`Samples of plasma or urine (10-20 pL) were injected without prior preparation onto a Brownlee Aquapore RP-300 column (4.6 x 250 mm) equipped with a 30 mm guard column of the same packing material. The column was eluted with either of two gradients: (1) 15A%/85%B to 80%A/20%B in 20 min, 80%A/20%B to 95%A/5%B in 5 min; (2) O%A/lOO%B to 55%A/45%B in 30 min at 1 mL/ min. Solvent A = acetonitrile/O.l% trifluoroacetic * Abbreviations: hGH, human growth hormone; RP- HPLC, reverse-phase HPLC; SE-HPLC, size-exclusion HPLC; 5’-MD, thyronine 5’-monodeiodinase; TFA, trifluoroacetic acid; TCA, trichloroacetic acid; rT3, reverse triiodothyronine; DTT, dithiothreitol; i.v., intravenous; 3- MIT, 3-monoiodo-L-tyrosine; NaI, sodium iodide; and PMSF, phenylmethylsulfonyl fluoride. acid (TFA); solvent B = water/O.l% TFA. Radio- activity profiles were determined with the use of an on-line Ramona 5-LS radiochemical detector, or column effluent was collected and radioactivity measured in a gamma counter.
`
`Urine (25-100 FL) and plasma (50 pL) samples from animals treated with “‘1-hGH were precipitated with ice-cold 15% TCA (final concentration). The precipitated proteins were pelleted in an Eppendorf microfuge at 16,000g for 10min. The supernatant was removed and the pellet was subsequently washed two more times with 15% TCA. The radioactivity in the pellets was determined in a gamma counter, and was assumed to represent undegraded and/or large fragments of “‘1-hGH. Similar experiments were run with control rat urine, plasma and buffer spiked with 12’I-hGH or lz51-NaI.
`Urine (50-100 pL) and plasma (100 ,uL) samples from animals treated with lz51-hGH were precipitated with 0.33% silver nitrate (final concentration). The samples were allowed to sit at room temperature for 10 min and complexed material was pelleted at 16,000g for 10 min. Radioactivity in the pellet was counted and considered to be inorganic rrsII. Similar experiments were also performed with control urine and plasma spiked with “‘I-NaI or “‘I-hGH, and buffer containing 3-monoiodo-L-tyrosine (1 mg/mL, HPLC assay as described below). In vitro
`Male Fischer 344 rats (21& 225 g) were killed, and homogenates of liver, kidney, and thyroid glands were made to 5% (w/v) in 50 mM sodium phosphate, 1.15% KC1 ( pH 7.2) with a hand- held glass/glass homogenizer at 4”. Homogenates were sedimented at 2000 g for 2 min to pellet large debris and the supernatants were used in the assay of deiodination. Protein was determined by the method of Bradford [16] with bovine serum albumin (BSA) as standard.
`Deiodination of “‘I-hGH and ‘251-hGH fragments was measured using a RP-HPLC method (see below). Conversion to free 1251- was expressed as the increase in the percentage of total radioactivity eluting at 5 min under the conditions used. The peak at 5 min cochromatographed with 1’51-NaI and was precipitated completely with silver nitrate, indicating that it was inorganic iodide.
`After precipitation with formic acid, the reaction supernatants were applied onto an Applied Biosystems Aquapore RP-300 (7.0 x 250 mm) column equipped with a 30 mm guard column of the same packing material. The column was eluted with either of two gradients: (1) lO%A/90%B for 5 min, lO%A/90%B to 30%A/ 70%B in 3min, 30%A/70%B for 5min, 30%A/ 70%B to lOO%A in 10 min, lOO%A for 2 min; (2) 5%A/95%B for 5 min, 5%A/95%B to 45%A/ 55%B in lOmin, 45%A/55%B to 95%A/5%B in 10 min, 95%A for 5 min at 1.5 mL/min. Solvent A =
`
`metabolism
`Tissue preparation.
`
`Deiodination
`Assay.
`
`RP-HPLC
`
`analysis.
`
`Page 2 of 9
`
`YEDA EXHIBIT NO. 2079
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`

`
`A
`
`0
`
`using 12sI-hGH as substrate
`
`Deiodination of ‘251-hGH 891 0 2 15 30 60 120 Time atter iv administration (min) Fig. 1. Accumulation of ‘251-radioactivity in thyroid glands of rats after i.v. administration of ‘251-hGH. Rats were dosed with ‘2SI-hGH (20 pg/kg, 8.1 &i/animal) and radioactivity was measured in thyroid glands, plasma, and liver samples. Results are expressed as cpm/g or percent dose/g tissue or mL plasma. Data are means 2 SD (N = 3). acetonitrile/O.l% TFA; solvent B = water/O. 1% TFA. Radioactivity profiles were obtained with the use of an on-line Ramona 5-LS radiochemical detector, or column effluent was collected and radioactivity measured in a gamma counter. Studies
`I z i 2 - 2mln z : - Mmln -v -v -- TC
`Liver, kidney and thyroid homo- genates (3.75 mg/mL) were incubated at 37” with 1251-hGH fragments (100-150 ng equivalents/ml) in a volume of 2OpL. Reactions were carried out for 90 min in the presence of 5 mM DTT and 1 mM NADPH, and terminated by addition of formic acid 100 60 0 t
`3 6 Retentlon (minutes) L-m 9 Fig. 2. SE-HPLC profiles of radioactivity in rat plasma and urine after i.v. administration of ‘251-hGH. (A) Plasma (2 and 60 min after i.v. dosing) and (B) urine samples were fractionated on a Zorbax GF 250 column as described in Methods. Plasma samples are from the experiment described in Fig. 1; the profile represents a typical chromatogram. The retention of standard iz51-hGH and of iz51-NaI are indicated. to 15%. Proteins were pelleted at 16,000 g for 1 min, and supernatants were analyzed by RP-HPLC.
`There was a biphasic decline of ‘*‘I-radioactivity in plasma following i.v. administration of 1251-hGH to rats (Fig. 1). The majority of radioactivity was cleared from the plasma during the first 15 min post- administration. After 15 min, an extended phase with a Ti/* of >2 hr was observed. The levels of radioactivity in the plasma at the 15- and 120-min time points represent approximately 13 and 8% of the radioactivity present at 1 min, respectively. Intravenous administration of 1251-hGH resulted in a time-dependent increase in 1251-radioactivity in the thyroid glands (Fig. 1). Two hours after administration, the level of radioactivity in the thyroid glands was 200-fold higher than the level after 2 min. In contrast to the thyroid, radioactivity in the plasma and liver decreased with time (Fig. 1,
`
`Several experiments were performed to elucidate the mechanism of ‘251-hGH deiodination. All incubations were carried out in 50 mM sodium phosphate/l.lS% KC1 (pH 7.2), over a period of 2 hr at 37”, using a homogenate protein concentration of 3.75 m /mL, f and a 1251-hGH concentration of 165 ng mL. Dithiothreitol (D’IT, 5 mM) and NADPH (1 mM) were added as cofactors as needed. When necessary, thryoid homogenates were also incubated with ‘251-hGH in the presence of the serine protease inhibitor, phenylmethylsulfonyl fluoride (PMSF), followed by an additional 60-min incubation with 5 mM dithiothreitol (DTT) and 1 mM NADPH.
`Thyroid homo- genates were incubated with 1251-hGH for 2 hr as above. After addition of formic acid to 15%, precipitated protein was pelleted at 16,000g for 3 min. The soluble fraction was dried in a Speed-vat (Savant), washed twice with distilled water, and redried in a Speed-vat. Radioactivity corresponding to 1251-hGH fragments (assessed by RP-HPLC, Fig. 7A) was suspended in phosphate buffer and used in subsequent assays.
`
`Studies using ‘251-hGH fragments as substrate
`Preparation of
`‘251-fragments.
`
`RESULTS
`
`studies
`Plasma kinetics and tissue accumulation.
`
`Incubations.
`
`Page 3 of 9
`
`YEDA EXHIBIT NO. 2079
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`In vivo
`

`
`892
`
`Routes of e~~~~nat~~n.
`
`V. J. WROBLEWSKI II hGH Peak EI Iodide Peak Time after Iv administration (mln) Fig. 3. SE-HPLC characterization of “‘I-radioactivity in rat plasma. Rats were dosed i.v. with 1251- hGH and plasma samples were fractionated on a Zorbax GF 250 column. The results indicate the percent of radioactivity at each time point corresponding with the retention of standard ‘*jI-hGH or ‘251-NaI. Data are means f SD (N = 4). Total radioactivity applied to the column ranged from 1000 to 7000 cpm. inset). Only 13% of the radioactivity present in the liver at 2min was present 2 hr after the i.v. administration. Two hours after iv. administration, the level of 1251-radioactivity/g tissue was 1150-fold greater in the thyroid gland than in liver. The profiles of radioactivity in the plasma 2 and 60 min after i.v. administration were determined by SE-HPLC (Fig. 2A). Two minutes after dosing, the majority of the radioactivity in plasma was associated with material having a retention time corresponding to parent 1251-hGH (4min). At 60min, however, the radioactivity was associated almost exclusively with a lower molecular weight component which had a retention time the same as *Z51-NaI. The radio- activity in this peak was precipitated completely by silver nitrate. The proportion of total radioactivity in plasma corresponding to *251-hGH or 1251-deter- mined by SE-HPLC indicated a time-dependent conversion of 1251-radioactivity to inorganic 1251- (Fig. 3). RP-HPLC analysis of radioactivity in plasma 60min after i.v. administration also indicated that the predominant form of radioactivity at this time was ir51- (Fig. 4A).
`The ‘*“I-radioactivity following i.v. administration of 1251-hGH was eliminated almost exclusively via the urine with >83% of the administered radioactivity recovered within 24hr (not shown). Size-exclusion (Fig. 2B) and reverse-phase HPLC (Fig. 4B) profiles of urine from the animals demonstrated the presence of a sin P; le peak having a retention time corresponding to 2SI-NaI. By RP-HPLC, the radioactivity in the urine cochromatographed with lz51-NaI but not with 3-monoido-L-tyrosine (not shown), providing further evidence that the radioactivity was inorganic in nature and not associated with tyrosine residues which could be liberated as a result of complete hydrolysis of lz51-hGH. Radioactivity in the urine of *251-hGH-treated rats was not precipitated by 15% TCA, resembling the results with biological samples spiked with ‘Z51-NaI. Similarly, the percentage of TCA precipitable radioactivity in plasma decreased with time after iv. administration of 1251-hGH (Fig. 5). 1251-Labeled material in urine was precipitated with silver nitrate, further indicating its inorganic nature (Fig. 6). Control urine spiked with 1251-NaI behaved similarly. The percentage of 1251-radioactivity in plasma which was precipitated by silver nitrate increased with time after i.v. administration of 12SI-hGH. ‘251-Labeled organic molecules ( 12sI-hGH and 3-monoiodo-L-tyrosine) were not precipitated by silver nitrate. In vitro
`In the absence of cofactors, 1251-hGH was converted by thyroid gland homo- genates to 12SI-labeled proteolytic products having retention times of 10 to 10.5 min and 16.75 to 17.5 min (Fig. 7A). These products were not formed by liver and kidney homogenates (not shown). A representative profile of the RP-HPLC assay for the conversion of 12s1-labeled peptides to rz51- in the presence of cofactors is shown in Fig. 7B. After a 2-hr incubation of iz51-hGH with liver, kidney or thyroid homogenates in the absence of cofactors the level of 1251- increased slightly from 2% in the control to 5.4 to 6.8% of the total radioactivity (Fig. 8). Addition of the cofactors, DTT and NADPH, to the incubation stimulated the formation of 12sI- by thyroid homogenates to 70%) but had no effect on rz51- formation by liver or kidney homogenates. The stimulatory effect of DTT and NADPH on deiodination in the thyroid homogenates was not apparent when reactions were carried out in the presence of the serine protease inhibitor, PMSF.
`
`assay of 12JI-hGI-I degradation, the increase in TCA solubility represented proteolysis of ‘251-hGH (Fig. 7A) as determined by RP-HPLC. In vitro
`
`studies
`TCA assay.
`vitro
`
`in
`
`deiodination.
`
`Page 4 of 9
`
`YEDA EXHIBIT NO. 2079
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`Under the conditions used in the
`

`
`0
`
`10
`
`Deiodination of 12’I-hGH 893 Although liver and kidney homogenates had poor proteolytic activity toward intact 1 ‘I-hGH and low levels of deiodinase activity when lz51-hGH was used as substrate, the deiodinase in these tissues was very active against 1251-labeled hGH fragments in the presence of DTT and NADPH (Fig. 9). Preincubation of thyroid homogenates with iodoacetate produced a concentration-dependent inhibition of the deiodination of 1251-labeled hGH ~agments (Fig. 10). The data suggest that the labeled fragments having a retention of 10 to 10.5 min were substrates for the deiodinase, while the more hydrophobic products which were present in smaller quantities appeared to be poorer substrates. DISCUSSION 0
`20 30 Retention (minutes) Fig. 4. RP-HPLC profile of radioactivity in plasma and urine of rats after i.v. administration of ‘2sI-hGH. (A) Plasma (60min after i-v. dosing) and (B) urine samples were fractionated on an Aquapore RP-300 column as described in Methods, using a two-step gradient of acetonitrile/O.l% TFA. Retention of standard ‘251-hGH and of ‘Z51-NaI are indicated. The figure shows a representative profile. Plasma and urine were from the same animal analyzed in Fig. 2. 100 - p 8 $ 50- ti z The information presented demonstrates that studies on the kinetics, metabolism and elimination of proteins or peptides which employ an ‘251-label should be interpreted cautiously. Whether the info~ation obtained from the in-depth analysis of a single 1251-labeled molecule (‘*%-hGH) pertains to a majority of 1251-proteins/peptides is uncertain. Differences in the molecular weight, tertiary structure, or metabolic stability of a 1251-labeled protein make it difficult to generalize about the in vivo disposition and deiodination of these molecules. However, data from previous studies with other lz51- labeled proteins [9-141, along with the results of the study described here suggest that deiodination reactions occur commonly and can obscure the interpretation of disposition studies. In the present study, conclusions regarding the catabolism of exogenously administered hGH could not be made due to the deiodination of the tracer employed. The kinetics of 1251-radioactivity after i.v. administration of lz51-hGH showed a biphasic pattern which was not comparable to information obtained afteri.v. administrationofunlabeled hGH torats[l7]. Since iodide has been shown to have a plasma half-life approaching 30 hr in the rat [ 181, this difference may be related to the kinetics of 12sE- which was the pre- dominant form of the radioactivity in the plasma \\ :::: ‘.‘. \‘\‘\ ‘\‘\ \‘\‘. I, ‘,>,> I \‘.‘. \‘\‘, \‘\‘\ \i’\ \‘.‘\ \‘C. \‘\I\ \‘\‘\ :::: \‘\‘\ e3 urine q
`
`(120 mln) H Plasma 12s I-Nal spike Cl Plasma 125 I-hGH spike Fig. 5. Trichloroacetic acid precipitation of ‘Z51-radioactivity. Urine and plasma (2, 30 and 120min post-dosing) from rats after iv. administration of ‘251-hGH (Fig. l), and control plasma spiked with ‘251-hGH or ‘2sI-NaI (SlOO-10,000 cpm) were precipitated with 15% TCA. The data show the percent of radioactivity in the samples that was precipitated, and represent means + SD (N = 3) or mean values from two individual experiments.
`
`mln)
`El Plasma (30 mln)
`q Plasma
`
`Page 5 of 9
`
`YEDA EXHIBIT NO. 2079
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`Plasma (2
`

`
`WROBLEWSKI
`
`n
`
`Ea Urine 0 Plasma (2 min) S Plasma (30 mln) Is Plasma (60 min)
`Urine/plasma ’ *’ I-Nal spike El Urlf;;plasma I-hGH spike
`
`q
`
`1
`
`3-MlT spike Fig. 6. Silver nitrate precipitation of rZSI-radioactivity. Urine and plasma (2, 30 and 120min post- dosing) from rats after i.v. administration of “‘1-hGH (Fig. l), control urine or plasma spiked with rz51-hGH or rz51-NaI (WOO-10,000 cpm), and water spiked with 3-monoiodo-L-tyrosine (3-MIT) were precipitated by the addition of silver nitrate to 0.33%. The data show the percent of radioactivity in the samples which was precipitated, and are means f SD (N = 3-5) or mean values from two individual experiments. ; f m- : v ,Frasmwm
`NADPH. Fig. 8. Deiodination of ‘*‘I-hGH by rat thyroid, liver and kidney homogenates. Liver, kidney and thyroid gland homogenates were incubated with lZ51-hGH in the presence or absence of DTT’ (5 mMf, and NADPH (1 mM). Thyroid homogenates were also incubated in the presence of PMSF (1 mM). Levels of 12sI- were determined by RP-HPLC and are expressed as percent of total radioactivity. Data are means t SD (N = 3-6) or mean values of two individual experiments. 15 min after the i.v. injection. The influence of the dose of hGH administered on the disposition of the 1251-labeled hGH is not clear from our studies. How- ever, saturation of hepatic uptake by unlabeled hGH could influence the disposition of labeled hGH as has been shown to occur with ‘231-insulin [ll]. The biphasic plasma kinetics of ‘251-radioactivity in the present study has also been observed after i.v. administration of ‘251-insuIin to rats [Ill. In that
`
`10 20 Retention (minutes) Fig. I. RP-HPLC separation of degradation products of ‘2sI-hGH formed in
`(A) Profile of i251-labeled products from a 120-min incubation with thyroid homogenate. (B) Profile from reaction performed in the presence of 5 mM DTTjl
`
`vitro.
`
`mM
`
`Page 6 of 9
`
`YEDA EXHIBIT NO. 2079
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`V. J.
`

`
`Deio~nation of i2%hGH 895 Thyroid Liver Kidney Fig. 9. Deiodination of llSI-hGH fragments by rat tissue homogenates. Liver, kidney and thyroid homogenates were incubated with ‘251-labeled hGH fragments in the presence of 5 mM DTT/l mM NADPH. The-percent conversion of labeled substrate to 1251- was determined by RP-HPLC. Data are means (liver, kidney; N = 2) or means 2 SD (thyroid, N = 5). study, the prolonged phase was also attributed to the presence of plasma 12sII which represented approxi- mately 50% of the total radioactivity 10 min after administration [ 111. The accumulation of radioiodine in the thyroid gland and excretion of non-protein associated radioactivity in the urine indicate that dei- odination was a major route of metabolism of 12sI- hGH. These findings correlate with the disposition of administerediodide [18], and have also been observed after the administration of 1251-monoclonal antibodies [13,14]. Although human urine has been shown tocontain intact growth hormone 1191, analysis of 1251-radioactivity in the plasma and urine from lz51- hGH-treated animals did not provide evidence for the formation of labeled hGH peptide intermediates. The inability to detect metabolic intermediates may result from the rapid clearance and subsequent deiodination of 1251-labeled fragments by deiodinase enzymes which are distributed ubiquitously in rat tissues [ll, 121. In this regard, studies on the disposition of 1251-(A14)-insulin have shown that the predominant form of radioactivity in plasma and liver 15 min after an iv. administration was iz51- [20]. Other inves- tigations have also examined the kinetics of 1251-hGH and hGH “variants” in animals and humans [8,21,22]. These studies have applied immunop- recipitation and solubihty of label in TCA assays to more accurately describe the kinetics of iz51-hGH and metabolites in plasma. This work has provided impor- tant information about the complexity of hGH kinetics. However, polyclonal antibodies do not clearly distinguish fragments closely related to hGH from hGH, nor does non-immunoprecipitable radio- activity or that soluble TCA necessarily reflect the presence of hGH fragments, as indicated in the present study. Thus, even with this more elaborate protocol, in vivo deiodination reactions can com- plicate significantly the analysis of the metabolism, disposition and kinetics of exogenously administered proteins. At neutral pH, 1251-hGH was converted to i2$ labeled peptides by thyroid gland but not liver or kidney homogenates as assessed by solubility in TCA and RP-HPLC. However, in the absence of cofactors there was no evidence of the deiodination observed after i.v. administration of ‘251-hGH. Previous studies have indicated the presence of an enzyme, iodothyronine 5’-monodeiodinase (5’-MD), which catalyzes the conversion of T4 (thyronine) to T3 [23]. Multiple “forms” of 5’-MD have been characterized [l, 24,251. These enzymes appear to be localized to the microsomal fraction of thyroid gland, liver, kidney and other tissues, and also catalyze the conversion of rT3 to 3,3’-diiodothyronine [12,26- 28]. The in vitro activity of 5’-MD shows an absolute % .Z .e ii $ El 125 I-Iodide 5 E 45
`
`n
`
`Fragment(s) 1 iG z [II Fragment(s) 2 B E 0.5 1 .o lodoacetate (mM) Fig. 10. Inhibition of deiodination of ‘%hGH fragments in thyroid homogenates. Thyroid homogenates were preincubated for 5 min with 0.5, 1, or 2 mM iodoacetate (IAA). “‘1-hGH fragments were then added and the reaction was incubated for 90 min in the presence of 500hM DIT/iOOpM NADPH. Percent of total radioactivity as rz51- or “‘I-fragment(s) 1 or 2 (see Fig. 7) was determined by RP- HPLC. Data are means ? SD of three (with IAA) or seven (control) experiments.
`
`Page 7 of 9
`
`YEDA EXHIBIT NO. 2079
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`

`
`WROBLEWSKI
`
`in viuo
`
`REFERENCES
`
`uitro
`
`requirement for thiol/dithiol cofactors and certain “forms” may require NADPH [28-301. In the absence of cofactors, lz51- was not present as a metabolic product of the reaction of 1251-hGH with thyroid, liver or kidney homogenates. However, addition of 5 mM DTT and 1 mM NADPH to the reaction stimulated the formation of 1251- in thyroid gland homogenates. Although liver and kidney have been reported to possess 5’-MD activity [ 12,25,28], these cofactors did not stimulate deiodination of lzsI-hGH in liver or kidney homogenates. In contrast to thyroid gland, these tissues did not proteolytically degrade “‘1-hGH under the in
`vitro data suggest that 1251-hGH- derived peptide fragments are suitable substrates for the deiodinase(s), while 1251-hGH is not a substrate. Since the natural substrate for 5’-MD, thyronine, has structural similarity to monoiodotyrosine, it is possible that the size and tertiary structure of 1251- hGH preclude access of lz51-residues to the active site of the deiodinase [31]. This seems reasonable since 1251-hGH fragments were substrates for deiodination in liver and kidney. In addition, the deiodination of 1251-hGH in thyroid gland homogenates was blocked by inhibition of hGH proteolysis. Inhibition of the deiodination of 1251- hGH fragments by iodoacetic acid implicates a sulfhydryl group in catalysis which is also consistent with the involvement of thyroid hormone deiodinases in this reaction [12]. The identity of the 1251- peptide fragment(s) which undergoes deiodination is unknown at this time. In a study of 1251-insulin, 1251- tyrosine was believed to be the predominant proteolytic end product prior to deiodination [ 111. These results are consistent with the mechanism proposed here that parent compound must first be proteolytically degraded in order for the in
`
`deiodination [32-341. Although these molecules may be useful for radioimaging studies, they are even more structurally different from the native protein than proteins iodinated by classical methods. This is a major concern for studies of protein disposition since both the method of labeling and the location of the iodine atom in the molecule can influence the kinetics and susceptibility of the molecule to degradation [ 10,351. In this regard, proteins labeled with 3H 14C or 35S more closely represent the biochemical and biological characteristics of the native molecule, and may be more appropriate for an accurate interpretation of disposition studies. Acknowledgements-1 would like to thank Ed Legan for labeling human growth hormone and Michael Masnyk for performing some of the
`
`in uivo
`
`conditions. However, kidney and liver homogenates did release “‘I- from 1251-hGH-derived peptide fragments. The pathway(s) for the degradation of hGH
`is still unclear. Whether lysosomal (acid) proteases are involved in hGH degradation
`is unclear from our studies, although the transitory uptake of radioactivity into the liver (Fig. 1) suggests that the lysosomal pathway for degradation cannot be excluded. The
`
`in uiuo
`
`in
`
`viuo
`
`deiodination of ‘*?-proteins to occur. In summary, evidence has been presented which indicates that after i.v. administration 1251-hGH undergoes an enzymatic deiodination. The charac- teristics of this enzymatic activity appeared similar to those of the enzyme(s) responsible for the deiodination of thyroid hormones. The ubiquitous nature of this enzymatic activity was consistent with the presence of 12sI- in the urine and plasma of treated animals. Recognizing this problem, several investigations have attempted to develop 1251- radioconjugates which are more stable to
`
`in vivo
`
`studies. 1. 2. 3. 4. Lucore CL, Fry ETA, Nachowiak DA and Sobel BE, Biochemical determinants of clearance of tissue-type plasminogen activator from the circulation. Circulation 77: 906-914, 1988. Canova-Davis E, Baldonado IP, Moore JA, Rudman CG, Bennett WF and Hancock WS. Prouerties of a cleaved two-chain form of recombinant h&an growth hormone.
`17: 14-19. 1689. - 6. 7. 5. Jolin T and Gonzalez C. Plasma clearance of hetero- geneous growth hormone components