Crosslinking Studies in Gelatin Capsules Treated with Formaldehyde and in Capsules Exposed to Elevated Temperature and Humidity

of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Crosslinking Studies in Gelatin Capsules Treated withFormaldehyde and in Capsules Exposed to ElevatedTemperature and Humidity CLYDE M. OFNER III, YU-E ZHANG, VALERIE C. JOBECK, BILL J. BOWMAN Department of Pharmaceutical Sciences, Philadelphia College of Pharmacy, University of the Sciences in Philadelphia,600 South 43 rd Street, Philadelphia, Pennsylvania 19104  Received 11 August 1999; revised 30 June 2000; accepted 11 July 2000 ABSTRACT:  Incomplete  in vitro  capsule shell dissolution and subsequent drug releaseproblems have recently received attention. A modified USP dissolution method wasused to follow capsule shell dissolution, and a 2,4,6-trinitrobenzenesulfonic acid (TNBS)assay was used to follow loss of    -amino groups to study this shell dissolution problempostulated to be due to gelatin crosslinking. The dissolution problems were simulatedusing hard gelatin capsule (HGC) shells previously treated with formaldehyde tocrosslink the gelatin. These methods were also used to study the effect of uncrosslinkedHGC stored under stressed conditions (37 °C and 81% RH) with or without the presenceof soft gelatin capsule shells (SGC). A 120 ppm formaldehyde treatment reduced gelatinshell dissolution to 8% within 45 min in water at 37 °C. A 200 ppm treatment reducedgelatin   -amino groups to 83% of the srcinal uncrosslinked value. The results alsosupport earlier reports of non-amino group crosslinking by formaldehyde in gelatin.Under stressed conditions, HGC stored alone showed little change over 21 weeks.However, by 12 to 14 weeks, the HGC exposed to SGC showed a 23% decrease in shelldissolution and an 8% decrease in the number of    -amino groups. These effects on thestressed HGC are ascribed to a volatile agent from SGC shells, most likely formalde-hyde, that crosslinked nearby HGC shells. This report also includes a summary of theliterature on agents that reduce gelatin and capsule shell dissolution and the possiblemechanisms of this not-so-simple problem.  © 2001 Wiley-Liss, Inc. and the American Phar-maceutical Association J Pharm Sci 90: 79–88, 2001 Keywords:  gelatin crosslinking, gelatin dissolution, amino group assay, formalde-hyde crosslinking in proteins INTRODUCTION  After ingestion, gelatin capsule shells must dis-solve sufficiently and release their drug contentsto produce a therapeutic effect. Incomplete  invitro  capsule shell dissolution and subsequentdrug release problems have recently received at-tention. This problem was first reported in 1974for a hard gelatin capsule (HGC) product contain-ing chloramphenicol 1 and in 1977 for a soft gela-tin capsule (SGC) product containing digoxin. 2 These and later cases were associated with ad- verse storage conditions of elevated temperature,humidity, or prolonged storage. But in these twocases, as in almost all others, there was no re-duced bioavailability of the drug compared withthe readily dissolving product. 3–5 Two cases of reduced bioavailability are of his-torical interest. The first case occurred in 1985,with an HGC product containing phenytoin that Current address:  Sanofi Research, 9 Great Valley Park-way, P.O. Box 3026 Malvern, Pennsylvania 19355 Correspondence to:  C.M. Ofner III (Telephone: 215-596-8881; Fax: 215-895-1100; E-mail: Journal of Pharmaceutical Sciences, Vol. 90, 79–88 (2001) © 2001 Wiley-Liss, Inc. and the American Pharmaceutical Association  JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 1, JANUARY 2001  79  was stored at elevated temperature and humidityand was identified in a patient hospitalized forloss of seizure control. 6  An FDA drug recall of aSGC product in 1993, because of dissolution prob-lems and reduced bioavailability, 7 is the secondcase. The latter led to the formation of an FDA/ Industry Gelatin Capsule Working Group to ex-amine the effects of the  in vitro  dissolution prob-lem on drug bioavailability. The Working Groupsimulated this dissolution problem with formal-dehyde crosslinked HGC and SGC containing acetaminophen and evaluated their bioequiva-lence in humans. 7,8 It is now generally believedthat this  in vitro  dissolution problem has littlerisk of reduced drug bioavailability because gas-tric and intestinal enzymes that contribute tothe physiological dissolution process have notbeen routinely used in dissolution tests. Theseenzymes break down the swollen and poorlysoluble film of the capsule shell, sometimes calleda pellicle, that usually retards or prevents drug release under  in vitro  dissolution conditions.Carstensen and Rhodes, 9 as well as others, havesuggested introducing such enzymes into  in vitro dissolution testing. One result of these sugges-tions and of the Working Group investigation wasthe initiation of the regulatory process to allowso-called two-tier testing; that is, the use of a sec-ond USP dissolution test, which contains an en-zyme if the product failed the srcinal monographdissolution test because of this problem. 7,10  Although this  in vitro  dissolution problem isunlikely to produce biological consequences, it re-mains a theoretical safety issue and its causes area concern. There is a growing body of evidencethat this problem is primarily caused by gelatincrosslinking in the capsule from aldehyde impu-rities that form the pellicle barrier to drug re-lease. 9,11 Crosslinking gelatin leads to an intri-cate network of high molecular weight that pro-duces a swellable hydrogel but substantiallyreduces, or even prevents, dissolution of the gela-tin. 12–14  An earlier review of gelatin crosslinking and capsule dissolution, 11 as well as related stud-ies with a different focus have been reported. 15–17 Recently, approaches to minimize this crosslink-ing and dissolution problem using additives 18 andby accelerated detection testing  19 have also beenreported.This report has three goals. The first is to de-scribe and disseminate an  in vitro  evaluation of gelatin crosslinking in the formaldehyde-treatedcapsules used in the Working Group acetamino-phen biostudy. The second goal is to report theapplication of the methods just described toevaluate crosslinking and shell dissolution inplain, uncrosslinked capsules that have been“stressed” by exposure to elevated temperatureand humidity. The last goal is to collect the dis-parate literature on this issue for a discussionand comparison of these results for this not-so-simple dissolution problem. The ideal evaluationof crosslinking in gelatin would include determi-nation of the crosslinking extent, identification of the agent or cause and the specific reactionmechanism, as well as identification of the site,the molecular weight, and the number of suchcrosslinks. A macroscopic and molecular ap-proach was used in the current investigation toelucidate some of these aspects. The macroscopicapproach was to evaluate dissolution of the cap-sule shell alone, without the presence of drug orexcipients, using an assay for the dissolved gela-tin. This gelatin dissolution is used as an indica-tion of crosslinking  13 without regard to specificssuch as mechanism or site. The molecular ap-proach was to assay for the potential loss of aminogroups due to crosslinking because formaldehyde(and other aldehydes) reacts with amino groupson lysine and hydroxylysine amino acid residuesof gelatin during crosslinking. 20–22 Such specific-ity suggests that this formaldehyde crosslinking can be quantitatively determined. The formalde-hyde–gelatin crosslinking mechanism, however,is not entirely clear; non-amino group crosslink-ing also has been reported. 21  A comparison be-tween results of the two methods in this investi-gation could corroborate such non-amino groupcrosslinking. Another approach to enhance ourunderstanding of this dissolution problem is toapply these methods in an exploratory manner tostressed, plain, uncrosslinked capsules. Specifi-cally, the theoretical possibility of self-cross-linking occurring without an agent 11,12 can be ex-amined. Under the stressed conditions of thisstudy, a crosslinking agent was generated from aSGC shell that induced crosslinking in HGCshells. This stressed-induced crosslinking can bedirectly measured, which to the authors’ knowl-edge has not yet been reported. EXPERIMENTAL SECTION Materials Type B gelatin granules, with a bloom strength of 254 and an approximate moisture content of 11%(w/w) determined by loss on drying at 105 °C for 80  OFNER ET AL.  JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 1, JANUARY 2001  72 h, were supplied by Kind and Knox (SiouxCity, IW, sample #T7468, lot #1). Capsugel AG(Basel, Switzerland) supplied formaldehyde-treated HGC and controls that were prepared asdescribed later. Plain uncrosslinked HGC used inthe stressed conditions study were purchasedfrom Lilly. Untreated SGC were supplied by Ban-ner Pharmacaps Inc. (High Point, NC) and con-tained a poly(ethylene glycol) (PEG) fill liquid anda shell composed of gelatin, glycerin, sorbitol, andwater. The BCA (bicinchoninic acid) protein assaykit and 2,4,6-trinitrobenzenesulfonic acid (TNBS)was purchased from Pierce Chemical Company(Rockford, IL). Water was purified by reverse os-mosis. All other chemicals were at least ACS re-agent grade. Formaldehyde-treated HGC wereprepared by Capsugel, 23 by mixing formaldehyde(HCHO) with lactose at different levels. TheHCHO/lactosemixturewasfilledintoemptyHGCcapsules that were then exposed briefly to 40 °Cand 75% RH, then to 25 °C and 50% RH for 6weeks, at which time the HCHO/lactose mixturewas removed from the capsules. The treatmentlevels of HCHO in the mixtures were 0, 20, 30,120, 200, 400, and 1000 ppm. The 0, 20, and 120ppm levels were evaluated in the Working GroupBiostudy. The control passed the  in vitro  dissolu-tion tests in the absence and presence of enzyme(P/P). The 20 ppm capsule failed the conventionaldissolution tests but passed with the addition of enzyme (F/P). The 120 ppm capsule failed disso-lution tests in the absence and presence of en-zyme (F/F). The 20 ppm capsules were found bio-equivalent to the uncrosslinked controls, whereasthe 120 ppm capsules were not bioequivalent. Stressed Hard Gelatin Capsules  A saturated ammonium sulfate solution was usedin glass desiccates to produce 81%RH at 37 °C.The desiccates were stored in an incubator at 37°C. Gelatin granules and plain HGC, in separatepetri dishes, were placed in a desiccator desig-nated as “HGC stored alone”. Gelatin granules,plain HGC, and untreated SGC, in petri dishes,were placed in the desiccator designated as “HGCstored with SGC”. Prior to this exposure, the SGCwere cut in half and rinsed with small amounts of methanol to remove residue of the PEG fill liquid.HGC and granule samples were removed at de-sired times and evaluated by shell dissolution andamino group assay. Capsule Shell Dissolution  A modified USP paddle method was used toevaluate dissolution of the gelatin from the cap-sule shell. The Van-Kel apparatus (Chatham, NJ)used consists of a 200-mL kettle (20-cm height,4-cm inner diameter), an adapter ring to allow the“mini” kettle to fit into the water bath system forconventional kettles, 100 mL of purified water at37 °C, a 100 rpm speed for the minipaddle, andTeflon-coated coilers to weight the empty cap-sules to the bottom of the kettle. After removal of 1-mL samples, this volume was replaced withwarmed purified water. Dissolved gelatin was de-termined from predetermined calibration plots byBCA protein assay or ultraviolet (UV) absorbancemeasurements at 214 nm. In the resulting cap-sule dissolution profiles, each point is the mean of six samples and error bars represent standard de- viation (SD). Gelatin Amino Group Assay The uncrosslinked, or free,   -amino groups of gela-tin were assayed by a method reported previ-ously 24 This method uses TNBS as a UV chromo-phore that reacts with primary amino groups and,after acid hydrolysis and organic extraction, re-mains in the aqueous phase reacted with   -aminogroups on lysine residues in peptide fragments. Statistics The results were analyzed by a one-way analysisof variance (ANOVA). Multiple comparisons be-tween means were evaluated by a Neuman–Keulstest. 25 For the few cases of non-normal variance,the multiple comparisons between means wereevaluated using a Kruskal–Wallis nonparametric ANOVA test with log-transformed data. 25 RESULTS Evaluation of Formaldehyde-Treated Capsules Capsules crosslinked with various amounts of HCHO were evaluated using gelatin shell disso-lution as a general indicator of crosslinking andthe   -amino group assay as an indicator of specificfunctional group participation. The dissolved cap-sule profiles in Figure 1 show no difference be-tween the control, untreated capsules and thosetreated with the two lower amounts of 20 and 30ppm HCHO. Additional, but unsuccessful, experi- CROSSLINKING STUDIES IN GELATIN CAPSULES  81  JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 1, JANUARY 2001  ments were conducted to enhance differences be-tween these profiles using different paddlespeeds, an acid medium, capsule strips, fillers,and different weights of fillers (data not shown).The next levels of 120 and 200 ppm, however,produced substantial reductions leading to 8 and2%, respectively, of dissolved gelatin by 45 min,and only 15 and 4%, respectively, of dissolvedgelatin by 2 h. In addition, the variation among replicates in the higher two levels of crosslinking was less than that of the controls and two lowercrosslinking levels. The SD for the controlsranged from 4 to 11, that of the 20 ppm capsulesranged from 1 to 7, and that of the 120 and 200ppm capsules ranged from 0 to 2.The loss of    -amino groups in the HCHO-treated capsules is shown in Figure 2. The un-treated control capsules have 32.9 × 10 −5 mol/g of free, uncrosslinked   -amino groups. This valuecorresponds to 32.9   -amino groups on one gelatinmolecule of 1000 amino acid residues with anideal molecular weight of 100,000. The value is virtually identical to the 33.0 × 10 −5 mol/g previ-ously reported for Type B gelatin granules. 24 The variation is also the same with a relative SD(RSD) of 2.6%.There is no measurable difference between thecontrol capsules and those treated with 120 ppmHCOH, and by inference, the capsules treatedwith lower amounts. There is, however, a smallbutmeasurablelossof   -aminogroupsforthecap-sules treated with 200 ppm. The loss continueswith increasing HCHO treatments to a final value of 27.4 × 10 −5 mol/g for capsules treatedwith 1000 ppm. This is a 17% loss of    -aminogroups compared with untreated capsules. TheSD for these maximally treated capsules is 0.29. Evaluation of Stressed Hard Gelatin Capsules Figure 3 illustrates the conditions of HGC expo-sure to 37 °C and 81% RH in the presence of SGC. A separate desiccator experiment was conductedwithout SGC to evaluate changes induced in gela-tin by only the elevated temperature and humid-ity. Figure 4 shows the percent dissolved of cap-sule gelatin at 45 min as a function of weeks of exposure to the stressed conditions. The HGCsstored alone showed little change in their com-plete dissolution up to 21 weeks of exposure. TheHGC stored with SGC showed dissolution re- Figure 1.  Gelatin shell dissolution profiles of formal-dehyde treated hard gelatin capsules in 37 °C water(bars are SD,  n  6). Figure 2.  Loss of    -amino groups in formaldehydetreated hard gelatin capsules (bars are SD;  n  5; *,  p < 0.05). Figure 3.  Stressing of hard gelatin capsules andgelatin granules at 37 °C and 81% RH in the presenceof soft gelatin capsule shells. 82  OFNER ET AL.  JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 1, JANUARY 2001
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!