JOURNAL OF
`
`Pharaoy nd
`Pharmacology
`
`MARCH 1999
`
`ISSN 0022-3573
`
`F» J
`
`4,“_:r:::» ('1
`régfiiyz5:
`W‘
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 -1/11
`Royal Pharmaceutical Society of Great Britain
`1
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 1/11
`
`

`
`Journal of Pharmacy and Pharmacology
`
`Published by The Royal Pharmaceutical Society of Great Britain
`1 Lambeth High Street, London SE1 7JN Telephone 0171-735 9141
`FAX 0171-820 3917
`E-Mail jpp@rpsgb.org.uk http://dspacc.dial.pipex.com/jpp/’
`Editorial Office for Japan: Department of Phamiaceutics, Kanazawa University, Takara-machi 13-1, Kanazawa 920
`
`EDITOR DR J CHAMBERLAIN
`
`ASSISTANT EDITOR DR A L SUGDEN
`
`EDITORIAL ASSISTANT G M MCMAIION
`
`EDITORIAL BOARD
`
`K L AUDUS, Lawrence, Kansas
`L L AUGSBURGER, Baltimore, Maryland
`B W BARRY, Bradford
`E J BEJAR, San Diego, California
`E BEUBLER, Graz, Austria
`N BODOR, Gainesville, Florida
`D D BREIMER, Leiden, The Netherlands
`D J BURGESS, Storrs, Connecticutt
`Y W CH IEN, Piscataway, New Jersey
`D A COWAN, London
`S P DENYER (Chairman), Brighton
`F J EVANS, London
`J J FERRARA, New Orleans, Louisiana
`J L FORD, Liverpool
`D R FRIEND, Menlo Park, California
`D GANDERTON, London
`E P GOLDBERG, Gainesville, Florida
`D J GREENBLATT, Boston, Massachusetts
`M J GROVES, Chicago, Illinois
`R H GUY, Archamps, France
`
`ASSOCIATE EDITOR (JAPAN)
`
`PROFESSOR A TsU..i
`
`A HOLT, Edmonton, Alberta
`P G JENNER, London
`I W KELLAWAY, Cardiff
`W E LINDUP, Liverpool
`H I MAIBACH, San Francisco, California
`T MANDAL, New Orleans, Louisiana
`R J NAYLOR, Bradford
`S 0113, San Francisco, California
`J W PHILLIS, Detroit, Michigan
`K D RAINSI-‘ORD, Sheffield
`W R RAVIS, Auburn, Alabama
`J E RIVIERE, Raleigh, North Carolina
`J R ROBINSON, Madison, Wisconsin
`A ROLLAND, The Woodlands, Texas
`H H TAI, Lexington, Kentucky
`B TESTA, Lausanne, Switzerland
`E TOMLINSON, The Woodlands, Texas
`B WIDDOP, London
`D E WURSTER, Iowa City, Iowa
`
`1999 Journal of Pharmacy and Pharmacology.
`COPYRIGHT
`Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by the Journal of Plr..i1naCY
`and Pharmacology for libraries and other users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, provided that
`the base fee of $15 per article is paid directly to CCC, 21 Congress St, Salem, MA 01970, USA.
`CODE: 0022-3573/90.
`
`Annual joint subscriptions (1999) to Journal of Pharmacy and Pharmacology and Pharmacy and Pharmacology Communications: UK £450; 1‘3111‘°pe
`£470; America and Japan S790; rest of world £490. Single issues: UK £30; Europe and rest of world £35; America and Japan $56.
`All orders should be sent to: The Pharrnaceutical Press, PO Box 151, Wallingford, Oxon 0X10 SQU, UK. Telephone +44 (0)1491 829 272; IAX 1'44
`(0)1491 829 292; E—Mail rpsgb@cabi.org
`Claims for missing copies cannot be considered unless received within three months of publication.
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 - 2/11
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 2/11
`
`

`
`Contents
`
`VOLUME 51 0 NUMBER 3 0 MARCH 1999
`
`Guest Editorial
`F C STRONG
`
`The history of the double blind test and the
`placebo
`
`301-306
`
`S LIN S-Y WANG E-C CHEN Y W CHIEN
`Insulin lispro: in-vivo potency determination
`by intravenous administration in conscious
`rabbits
`
`Critical Review
`TB VREE A J AM VAN DERVEN
`
`Clinical consequences of the biphasic
`elimination kinetics for the diuretic effect of
`furosemide and its acyl glucuronide in
`humans
`
`Research Papers
`Pharmaceutics
`G SUNKARA C B NAVARRE U B KOMPELLA
`Influence of pH and temperature on kinetics
`of ceftiofur degradation in aqueous solutions
`
`313-318
`
`319-330
`
`Pharmacology
`D PUBILL A M CANUDAS M PALLAS F X SUREDA
`E ESCUBEDO A CAMINS J CAMARASA
`Assessment of the adrenergic effects of
`orphenadrine in rat vas deferens
`S PAL T SEN A K N CHAUDHURI
`
`Neuropsychopharmacologieal profile of the
`methanolic fraction of Bryophyllum
`pinnatum leaf extract
`H H PERTZ A M BROWN T L GAGER
`A J KAUMANN
`
`Simple O-acylated derivatives of lysergol and
`dihydrolysergol~I: synthesis and interaction
`with 5-HTZA, 5-HT2C and 5-HT”; receptors,
`and 0:1-adrenergic receptors
`
`257-261
`
`Biopharmaceutics
`H TOZAKI T FUJITA J KOMOIKE S—I KJM
`H TERASHIMA S MURANISHI S OKABE
`A YAMAMOTO
`
`263-269
`
`Colon-specific delivery of budesonide with
`azopolymer-coated pellets: therapeutic
`effects of budesonide with a novel dosage
`form against 2,4,6-ninitrobenzenesulphonic
`acid-induced colitis in rats
`
`L I HARRISON G L COLICE D DONNELL I SORIA
`R DOCKHORN
`
`Adrenal effects and phannacokinetics of
`CFC-free beclomethasone dipropionate:
`a 14-day dose-response study
`Y KOBAYASHI M MIYAMOTO K SUGIBAYASHI
`Y MORHVIOTO
`
`Drug permeation through the three layers of
`the human nail plate
`
`Biochemistry
`M KOBAYASHI H FUJISAKI M SUGAWAKA
`K ISEKI K MIYAZAKI
`
`The presence of an Na+/spermine antiporter
`in the rat renal brush-border membrane
`1 P WANG M F HSU S L RAUNG L C CHANG
`LT TSAO PL LIN C C CHEN
`
`Inhibition by magnolol of forn1ylmethionyl-
`leucyl-phenylalanine—induced respiratory
`burst in rat neutrophils
`M WATANABE T KAIHATSU M MIWA T MAEDA
`Ca2+/calmodulin-dependent protein kinase 1]
`inhibitors potentiate superoxide production
`in polymorphonuelear leukocytes
`
`Gastrointestinal Pharmacology
`K D RAINSFORD
`
`Inhibition by leukotriene inhibitors, and
`calcium and platelet-activating factor
`antagonists, of acute gastric and intestinal
`damage in arthritic rats and in
`cholinomimetic-treated mice
`C A HlRUMA—LTl\/IA J S GRACIOSO D S NUNES
`A R M SOUZA BRITO
`Effects of an essential oil from the bark of
`Crown cajucara Benth. on experimental
`gastric ulcer models in rats and mice
`
`341-346
`
`Drug Metabolism and
`Pharmacokinetics
`S NAITO M NISHIMURA H NOGAWA
`Pharmacokinetics of BOF-4272, a xanthine
`oxidase inhibitor, after single intravenous or
`oral administration to male mice and rats
`D W BOULTON U K WALLE T WALLE
`Fate of the flavonoid quercetin in human cell
`lines: chemical instability and metabolism
`H KATAYAMA M YASUHARA R HORI
`Effect of acute renal failure on the
`disposition of cefoperazone
`
`353-359
`
`361-366
`
`367-370
`
`Clinical Pharmacology
`M E MULLINS
`
`First degree atrioventricular block in
`alprazolam overdose reversed by flumazenil
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 - 3/11
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 3/11
`
`

`
`This material may be protected by Copyright law (Title 17 U.S. Code)
`
`J Pharm. Pharmacol. 1999, 51: 271-278
`Received August 25, l998
`Accepted November 2, 1998
`
`(Q 1999 J. Pharm. Pharmacol.
`
`Drug Permeation through the Three Layers of the
`Human Nail Plate
`
`YOICHI KOBAYASHI, MISAO MlYAMOTO*, KENJI SUGIBAYASHI
`AND YASUNORI MORIMOTO
`
`Faculty of Pharmaceutical Sciences, Josai University, ]-I Keyakidai, Sakado, Saitama 350-0295 and
`*Nissan Chemical C0,, Ltd, 3-7-] Kanda—Nishiki-cho, Chiyoda-ku, Tokyo 101-0054, Japan
`
`Abstract
`
`The in—Vitro permeation characteristics of a water soluble model drug, 5—fluorouracil, and a
`poorly water soluble model drug, flurbiprofen, were investigated through three layers of the
`human nail plate (namely, the dorsal, intermediate and ventral nail plates), using a modified
`side—by—side diffusion cell. The dorsal—filed nail plate, the Ventral—filed nail plate and the
`dorsal—and—ventral—filed nail plate were prepared to known thicknesses and then used with
`the full—thickness nail plate to investigate the permeation characteristics of each single
`layer.
`Most of the lipids in the human nail plate were found in the dorsal and ventral layers.
`The rank orders of the permeation fluxes for 5—fluorouracil and flurbiprofen were both:
`dorsal-and—ventral—filed nail plate > dorsal—filed nail plate > Ventral—filed nail plate > full-
`thickness nail plate. With respect to 5—fluorouracil permeation through each single layer,
`the permeability coefficient of the intermediate layer was higher than those of other single
`layers. However in the case of flurbiprofen, the permeability coefficient of the ventral layer
`was higher than other single layers. The diffusion coefficients of 5-fluorouracil and
`flurbiprofen in the dorsal layer were the lowest of any single layer. The drug concentration
`in each layer was estimated using each respective permeation parameter. The drug
`concentration in the nail plate was observed to be dependent on the solubility and the
`flux of the drug.
`From these findings, we suggest that the human nail plate behaves like a hydrophilic gel
`membrane rather than a lipophilic partition membrane and that the upper layer functions as
`the main nail barrier to drug permeation through its low diffusivity against the drugs.
`
`Although the human nail plate is generally con-
`sidered to be composed of either one layer or three
`layers,
`the three—layer model
`is more widely
`accepted (Spearman 1978; Dawber 1980). The
`“PPCI,
`the middle and the lower layers of the
`human nail plate are called the dorsal, intermediate
`and Ventral nail plates, respectively (Figure 1). In
`addition, the tissue under the nail plate, called the
`“all bed, consists of viable epidermis. Information
`regarding the distribution of fungi
`in the human
`D3-ll plate is required for the treatment of onycho-
`nlycosis trichophytica. It has been reported that the
`dorsal nail plate, the ventral nail plate,
`the sub-
`lmgllal keratin and the eponychium are infected by
`ngl Such as Trichophyton rubrum and TI"iC/’lO-
`
`Sc.C”1TESp0ndence: Y. Morimoto, Faculty of Pharmaceutical
`fences, Josai University, 1-1 Kcyakidai, Sakado, Saitama
`500295. Japan.
`
`phyton mentagrophytes (Sagher 1948; Jillson &
`Piper l957; Akiba 1971). Of these nail compo-
`nents, it is the ventral nail plate and the subungual
`keratin that are found to be infected in most cases.
`
`Therefore, it is of great importance to understand
`the specific properties of each layer in the human
`nail plate.
`Onychomycosis trichophytica has been treated
`mainly with oral antifungal medication (Piepponen
`et al 1992; Villars & Jones 1992). With systemic
`treatment, an antifungal drug may be delivered
`from the blood in the dermis under the nail bed and
`
`the nail matrix to the ventral nail plate and sub-
`ungual keratin (Matthieu et al 1991). On the other
`hand, antifungal drugs are not delivered to the
`ventral nail plate and the subungual keratin fol-
`lowing topical treatment because of low nail—plate
`permeability. Each layer has unique physical
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 - 4/11
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 4/11
`
`

`
`YOICHI KOBAYASHI ET AL
`
`_
`
`Eponychium
`
`Dorsal nail plate
`Intermediate nail plate
`¢
`\
`
`¢
`
`Ventral nail plate
`
`Figure 1.
`
`Schematic diagram of the human nail plate.
`
`drug permeabilities
`therefore,
`and,
`properties
`through each layer may be different. Only a few in-
`vitro drug permeation studies have been performed
`on the human nail plate. Mertin & Lippold (l997a)
`reported that the human nail plate and the keratin
`membrane from bovine hooves (the model mem-
`brane of
`the human nail plate), behave like
`hydrophilic gel membranes. Walters et al (1983,
`I985) also suggested that
`the human nail plate
`behaves like a hydrophilic gel membrane. Fur-
`thermore, they suggested that it has an additional
`lipophilic route for permeation.
`ln the present study, the distribution of lipids in
`the human nail plate was examined in detail. In
`order to investigate the nail permeability of a water
`soluble model drug, 5—fluorouracil, and a poorly
`soluble model drug, flurbiprofen, a modified side-
`by—side diffusion cell was used. The drug perrnea—
`tion characteristics of each single layer were
`investigated by filing upper or lower layers, or both,
`of the human nail plate. 5—Fluorouracil and flurbi-
`profen perrneations, from the ventral to the dorsal
`nail plate, were compared with those from the
`dorsal to the ventral nail plate. The amounts of 5-
`fluorouracil and flurbiprofen in the nail plate were
`also compared with those calculated from the per-
`meation parameters of each drug.
`
`Materials and Methods
`
`Materials
`
`5—Fluorouracil was obtained from Tokyo Kasei
`Kogyo Co., Ltd (Tokyo, Japan). Flurbiprofen was
`supplied by Kaken Pharmaceutical Co., Ltd (Chiba,
`Japan). Triolein (practical grade), cholesterol and
`Vaniline (for measuring total
`lipid) and Sudan
`Black B (for lipid staining) were purchased from
`Wako Pure Chemical
`Industries, Ltd (Osaka,
`Japan). Other reagents were obtained from com-
`mercial sources.
`
`Preparation of nail plates
`Tip nail pieces were obtained from the fingers of
`healthy volunteers using nail clippers. A lot of nail
`pieces, clipped from the same volunteer (ma1e
`aged 24 years) were used in the permeation study
`and for microscopic observation. Forty nail pieces
`obtained from other healthy volunteers (10 males,
`mean age 24 years) were used in the measurement
`of total lipid. The thicknesses of nail pieces were
`measured with a micrometer
`(Mitutoyo Corp”
`Japan) equipped with pointed metal attachments,
`The thickness ratio of each layer of the human nail
`plate (dorsal : intermediate 2 ventral) was assessed
`to be 3 : 5 : 2 after examining several reports (Jillson
`& Piper 1957; Spearman 1978; Stiittgen & Bauer
`1982). The dorsal—filed nail plate, the ventral—filed
`nail plate and the dorsal— and ventral—filed nail plate
`were prepared to known thicknesses with sand
`paper.
`
`Determination of solabilities and nail plate/ vehicle
`partition coeficients of drugs
`5—Fluorouracil and flurbiprofen aqueous suspen-
`sions were mixed with a magnetic stirrer at 37°C.
`After 24 h, each suspension was subjected to fil-
`tration (Ekicrodisc 3; German Sciences Japan, Ltd,
`Tokyo). The filtrate was immediately diluted with
`distilled water or methanol to obtain samples for
`analysis.
`The finger nail pieces (full—thickness nail plate,
`ventral—filed nail plate, dorsal—filed nail plate and
`the
`dorsal—and—ventral—filed
`nail
`plate) were
`weighed with an electronic balance (JL—200, Chyo
`Balance Corp.). They were immersed in half—c0n—
`centration solutions (1 mL) of 5—fluorouracil Or
`flurbiprofen solubility at 37°C for 48 h. After
`removal of the nail piece from each solution, the
`concentration of the solution was measured. T116
`drugs
`in the nail pieces were extracted with
`methanol (10 mL) three times. The extracted S01-
`vent was evaporated under nitrogen gas at 60°C-
`Subsequently, 1mL of distilled water or methaI101
`was added to obtain samples for analysis. We
`assumed that the weight (volume) ratio of Bach
`layer of the nail plate was 3:5 :2. The appflfem
`partition coefficient of the drugs was calculated a5
`the concentration ratio in the nail plate/vehicle at
`37°C.
`
`.
`,
`_
`Permeation studies
`To overcome large variations in nail permeab1l1t1€5
`due to individual differences in barrier pf0P"m6Sd’
`nail pieces (20-35 mg) from the third finger
`the fifth finger of the same volunteer were used In
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 - 5/11
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 5/11
`
`

`
`DRUG PERMEATION THROUGH HUMAN NAIL PLATE
`
`273
`
`the permeation studies of 5—fluorouracil and flur-
`biprofen, respectively. The full-thickness nail plate,
`the dorsal-filed nail plate, the ventral—filed nail plate
`and the dorsal—and-ventral—filed nail plate were
`Sandwiched between 2 adapters made of poly-
`propylene with an 2O—shaped ring (effective diffu-
`sion area, 0049 cm ) and mounted in a side—by—side
`diffusion cell
`(2-5mL or 1-5 mL) with a water
`jacket connected to a water bath at 37°C. 5—Fluor-
`0uracil~ or flurbiprofen—suspension was applied to
`me dorsal nail plate side of the diffusion cell, and
`the ventral nail side was filled with distilled water
`or 40% polyethylene glycol 400 (to maintain the
`sink condition). In the case of vehicle application to
`the ventral nail plate side, 5—fluorouracil— or flur-
`biprofen-suspension was placed in the ventral nail
`plate side compartment, and the dorsal nail plate
`side was filled with distilled water or 40% poly-
`ethylene glycol 400. No preservative was added
`because the receiver solution was clear even at the
`end of the experiment. The permeation was mea-
`sured by sampling the receiver side solution at
`predetermined times. The experimental period was
`7 days (5—fluorouracil) or 19 days (flurbiprofen)
`because of the low nail permeability of drugs.
`
`Measurement of total lipid
`Total lipid was measured according to the method
`of Knight et al (1972). After nail pieces from the
`fingers of healthy volunteers were prepared into
`full-thickness, ventral—filed, dorsal-filed and dorsal-
`and—ventral—filed layers, they were immersed in a
`chloroform—methanol (3:1) mixture for 24h to
`extract lipids. The extracted solvent was evaporated
`under nitrogen gas at 50°C. Sulphuric acid (300 ,uL)
`was added to the sample and heated in boiling
`Water for 5 min. After cool ing to room temperature,
`2mL of phosphovaniline test solution (0-6% w/v
`aqueous vaniline solution/phosphoric acid; 1:4)
`was added and the mixture was incubated for
`15min at 37°C. Then, the absorbance was mea-
`Sllred at 535nm using a spectrophotometer (UV-
`190A, Shimadzu Seisakusho, Kyoto, Japan). A
`II11xture of cholesterol and triolein (3: 1) was used
`38 a reference standard.
`
`Observation of lipid distribution
`A Hail piece from the first finger of a healthy
`Vfllunteer was used to observe nail plate lipid dis-
`‘Hbution. After the nail plate was frozen in an
`embedding medium (Tissue—Tek; Sakura Fine-
`lefshnicai Co., Ltd, Tokyo),
`it was sliced with a
`mlcfotorne
`(IEC Model Minotome Microtome
`Tyostat; International Equipment Company). A
`
`section of this nail plate was stained with Sudan
`Black B in isopropyl alcohol
`(5 mgmL’1) for
`30 min. After being washed with 50% methanol for
`about 30 s,
`it was observed under a microscope
`(New VANOX, Olympus, Tokyo).
`
`Analytical method
`5—Fluorouracil and flurbiprofen levels were deter-
`mined by high—performance liquid chromatography
`(HPLC). Sample solutions were injected into the
`HPLC instrument, which was composed of a pump
`system (LC—10A, Shimadzu Seisakusho, Kyoto,
`Japan), a UV detector (SPD-10A, Shimadzu), a
`fluorescence detector (RF—10AXL, Shimadzu), a
`Chromatopack (C—R5A, Shimadzu), a system con-
`troller
`(SCL—10A, Shimadzu), an auto injector
`(SIL—10A, Shimadzu) and a reverse phase column
`(Inertsil ODS 250mm><4-6 mm i.d., GL Sciences
`Inc., Tokyo). The mobile phase for flurbiprofen was
`0-1% phosphoric acid/acetonitrile (40:60),
`the
`flow rate was 1 mLmin_‘ and fluorescence detec-
`tion was conducted at an excitation wavelength of
`260nm and an emission wavelength of 313nm.
`The mobile phase for 5—fluorouracil was 0-1%
`phosphoric acid/acetonitrile (98 :2), the flow rate
`was 1 mLmin‘ , and UV detection was conducted
`at a wavelength of 270nm.
`
`Results and Discussion
`
`Lipid distribution in the nail plate
`The lipid content in the human nail plate was found
`to be much lower than that in the stratum comeum
`of skin (Walters & Flynn 1983). However, various
`lipids such as long—chain fatty acids,
`free fats,
`cholesterol, squalene and phospholipids are present
`in the human nail plate (Spearman 1978; Hirose et
`al 1990). It is thought that the lipid distribution and
`concentration in the human nail plate may affect
`the drug permeation, particularly the partition of
`drugs into the membrane.
`Table 1 shows the total—lipid concentration (CL)
`in the ventral—filed, dorsal—filed, dorsal—and—ventral—
`filed and full—thickness layers. The ventral—filed and
`dorsal-filed layers had a high total—lipid con-
`centration; about l5 and 3-0 times those observed
`in the dorsal—and-ventral—filed layer. The total—lipid
`concentrations of each single layer were calculated
`from the experimental data of the ventral—filed,
`dorsal-filed
`and
`dorsal—and-ventral—filed
`layers
`(Table 1). The total—lipid concentration in the
`ventral layer was the highest in the human nail
`plate. In contrast, the intermediate layer had a low
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 - 6/11
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 6/11
`
`

`
`274
`
`YOICHI KOBAYASHI ET AL
`
`Table 1. Experimental (a) and calculated lipid (b) concentration in the nail plate.
`
`Experimental Layer
`(thickness ratio)
`
`Ventral—filed
`(3 : 5)
`
`CL (,ugmgT1)
`
`4.33:0.45
`
`Calculated Layer
`(thickness ratio)
`
`CL uigmg“)
`
`Dorsal
`(3)
`
`6-82
`
`Dorsal—filed
`(5 2 2)
`
`8A76:l:1-15
`
`Intermediate
`(5)
`
`2-83
`
`Dorsal—and-ventral-filed
`(5)
`
`2-83 :: 0-28
`
`Ventral
`(2)
`
`236
`
`Full-thickness
`(3 : 5 :2)
`
`7~53i1.20
`
`Full-thickness
`(3 ; 5 ; 2)
`
`8-50
`
`CL :total lipid concentration. Each value in Table 1 represents the mean :: s.e.m. (n : 10).
`
`total—lipid concentration. The rank order of the total
`lipid concentration in each single layer was ventral
`layer > dorsal layer > intermediate layer. The total
`lipid concentration in the full—thickness nail plate
`(Table 1) was calculated using those of each single
`layer. This lipid concentration was very similar to
`the value for the full—thickness nail plate obtained
`from the experiment.
`Figure 2 shows a micrograph of a human nail
`plate cross section. The lipids in the nail plate were
`stained with Sudan Black B. Sudan Black B can
`
`stain most lipids, such as neutral fats and phos-
`pholipids. Most lipids were observed in the upper
`and lower parts of the human nail plate. Moreover,
`uneven parts of the ventral side contained a lot of
`lipids, whereas no lipids were observed in the
`intermediate part of the nail plate.
`From these results, we suggest that the dorsal and
`ventral layers in the human nail plate contain some
`lipids, whereas the intermediate layer, which is the
`main nail body of the human nail plate, contains
`few lipids.
`
`Figure 2. Micrograph of a human nail plate. D: dorsal nail
`plate side. V: ventral nail plate side. This nail plate section was
`stained with Sudan Black B.
`
`Permeation parameters of each layer
`The data analysis is based on Fick’s first law:
`
`dQ/dt = DmAC/h
`
`(1)
`
`where dQ/dt is the steady state permeation rate, DH,
`is the diffusion coefficient of the drug in the
`membrane, AC is the concentration differential of
`the drug in the membrane, and h is the membrane
`thickness. In the case of a sink condition, AC in
`Equation 1 can be replaced with the product of the
`solubility of the drug in the donor solvent (CV) and
`the membrane/donor vehicle partition coefficient
`of the drug (Km):
`
`dQ/dt = DmKm CV /h
`
`(2)
`
`The permeability coefficient of the drug (P) is
`given by:
`
`P = DmKm /h
`
`(3)
`
`Figure 3 shows the permeation profiles of 5—fluor-
`ouracil and flurbiprofen through the Ventral—filed,
`dorsal—filed,
`dorsal—and—ventral—filed
`and
`full-
`thickness nail layers. The rank orders for the 5-
`fluorouracil and flurbiprofen fluxes were both:
`dorsal—and—ventral—filed layer > dorsal-filed layer >
`ventral—filed layer>full—thickness layer. The 13%
`times for 5—fluorouracil and flurbiprofen pemlfia‘
`tions through t'ull—thickness nail plates were 21130111
`2-5 and 11 days, respectively. Mertin & Lipp01d
`(1997a) also suggested that steady—state perflleaf
`tions of nicotinic acid esters through human {I311
`plates were obtained after
`l0—80 h.
`In 3.dCl1t10nv
`they reported that
`steady—state permeation Of
`chloramphenicol, from an aqueous suspension of 3
`nail lacquer, through the human nail plate occnfffid
`after a lag time of 200h and 400 h,
`respeCt1Ve1y
`(Mertin & Lippold 1997b).
`Table 2 shows the 5—fluorouracil and flurbiPr0fen
`permeation parameters of the ventral—filed, d0T5a1'
`filed,
`dorsal—and—ventral—filed and full-thiCk_I“35:
`nail layers. The apparent permeability coefficlentr
`of each drug through these nail plates were 031°"
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 - 7/11
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 7/11
`
`

`
`DRUG PERMEATION THROUGH HUMAN NAIL PLATE
`
`c» OOO
`
`
`
`
`
`5~f|uorouraci|permeatedlugcm‘2)
`
`
`
`Flurbiprofenpermeated(Mgcmez)
`
`
`
`
`
`:3 o
`
`01 O
`
`N.i>ooooo
`
`4
`
`Time (days)
`
`10
`
`15
`
`Time (days)
`
`Figure 3. Permeation profiles of 5—fluorouracil (a) and flurbiprofen (b) through ventral—filed (A), dorsal—filed (
`ventral—filed (O) and full-thickness (O) nail layers. Each Value represents the mean :: s.e.m. (n: 3-5).
`
`), dorsal—and—
`
`lated using Equations 2 and 3. In all permeation
`studies, CV Values are given in saturated con-
`centrations of 5—fluorouracil (17-1mgmL’1) and
`flurbiprofen
`(277 ,ug mL—‘)
`at 37°C, because
`donor vehicles are aqueous suspensions. Although
`the ventral—filed nail layer had approximately the
`same 5—fluorouracil permeability coefficient as the
`dorsal—filed layer, the dorsal—and—Ventral—filed layer
`had a higher permeability coefficient
`than those
`layers. On the other hand, the fiurbiprofen perme-
`ability coefficient of the dorsal—and—Ventral—filed
`layer was also higher than those of the ventral—filed
`and dorsal—filed layers. The nail plate/vehicle par-
`tition coefficients of each drug were calculated as
`Concentration ratios in the ventral—filed, dorsal-
`filed,
`dorsal—and—Ventral—filed and full—thickness
`layers/water at 37°C, respectively. The nail plate/
`Vehicle partition coefficients of 5—fluorouracil and
`flllrbiprofen were not very different between the
`Ventral—filed, dorsal—filed and dorsal—and—ventral—
`filed layers. The diffusion coefficients of each drug
`111 the Ventral—filed, dorsal—filed, dorsal—and—Ventral—
`filed and full—thickness nail layers were calculated
`“Sing Equation 3. The diffusion coefficients of 5-
`fll10r0uracil and flurbiprofen in the ventral—filed
`layer were lower than those in the dorsal—filed and
`d0ISal—and—ventral—filed layers.
`
`The permeation parameters of each single layer
`(dorsal,
`intermediate or Ventral) were computed
`using those of the Ventral—filed, dorsal—filed and
`dorsal—and—ventral—filed layers. Flynn et al (1974)
`proposed that the diffusional resistance, R,-,
`in the
`ith layer can be defined by:
`
`R = 1/P. = h./<D.K.>
`
`(4)
`
`In multiple layers, the total diffusional resistance
`(RT) may be computed by:
`
`RT =1/PT = hl/(D1K1)‘i‘h2/(DZKZ)
`
`+ -
`
`-
`
`- + hn/(DnKn)
`
`(5)
`
`Table 3 shows the permeation parameters of 5-
`fluorouracil and flurbiprofen for each single nail
`layer. The permeability coefficient of each single
`layer was calculated using Equation 5. For 5-
`fluorouracil,
`the intermediate layer had a high
`permeability coefficient compared with the other
`single layers. In contrast, the Ventral layer had a
`high permeability coefficient for flurbiprofen. The
`permeability coefficients
`for 5—fluorouracil and
`flurbiprofen through the dorsal layer were low. The
`nail plate/Vehicle partition coefficients of the drugs
`for each single layer were calculated using those
`of the Ventral—filed, dorsal—filed and dorsal—and—
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 - 8/11
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 8/11
`
`

`
`.
`
`1
`
`276
`
`YOICHI KOBAYASHI ET AL
`
`Table 2. Permeation parameters of 5—fluorouracil and flurbiprofen through the human nail plate.
`
`Drug Parameter
`
`
`
`Layer (thickness ratio)
`
`m F
`
`Dorsal-filed (5 : 2)
`
`Dorsal—and—ventral-filed (5)
`
`ull—thickness (3 ; 5 :2)
`
`5—F1uorouracil
`
`1
`h (#1107
`P (x 10 cm s“ )
`Km
`Dm (X108 crn2s_1)
`Flurbiprofen
`I
`h (HUI) 6
`P (><10 ems’ )
`Km
`Dm (X108 cm2 s
`
`1)
`
`Ventral—filed (3 : 5)
`
`396.0
`2-51 :|: 0-18
`0-53 :: 0-04
`1-87
`
`347-2
`l-83 :: 0-11
`1-07 :1: 0-15
`5-92
`
`346-5
`3-07 :1: 0-22
`0-49 :: 0-01
`2-18
`
`303-8
`2-76 :: 0-43
`1- 13 : 0-24
`7-42
`
`247-5
`7-12:: 0-65
`0-56:l:0-02
`3-13
`
`217-0
`3-59 :: 1-28
`0-98 :: 0-22
`7-96
`
`495-0
`1-49:
`0-54:
`1-37
`
`434-0
`1-45 :1: 0-54
`1 -47 :: 0-32
`4-29
`
`Each value represents the mean:l: s.e.m. (n : 3-5). Apparent permeability coefficients were calculated using Equations 2 and 3_
`h is the nail membrane thickness; P is the permeability coefficient; Km is the membrane/donor vehicle partition coefficient of the
`drug; Drn is the diffusion coefficient of the drug in the nail membrane.
`
`ventral—filed nail layers and the weight ratio of each
`layer
`(3 : 5 :2). The partition coefficient
`for 5-
`fluorouracil in the ventral
`layer was lower than
`those in the dorsal and intermediate layers. The
`rank order of the partition coefficients for 5 —fluoro—
`uracil was
`intermediate layer > dorsal
`layer >
`ventral layer, whereas the rank order of the flurbi-
`profen partition coefficients was the reverse of this.
`In addition, rank orders of these partition coeffi-
`cients for each drug are correlated with those of the
`total lipid concentrations in the three layers of the
`human nail plate. The diffusion coefficients of the
`drugs in each single layer was calculated using
`Equation 3. The dorsal
`layer had low diffusion
`coefficients
`for 5—fluorouracil
`and flurbiprofen
`compared with the other single layers. The diffu-
`sion coefficients for 5—fluorouracil and flurbiprofen
`in the intermediate layer were the highest of those
`for any single layer. The permeation parameters of
`the full—thickness nail plate (as shown in Table 3)
`
`were calculated by substituting those of each single
`layer, according to Equation 5. They agreed with
`the experimental permeation parameters of the full-
`thickness nail plates obtained from the permeation
`studies for both drugs (as shown in Table 2).
`From these results,
`the drug permeation char-
`acteristics of each single layer can be summarized
`as follows: the dorsal layer is characterized by a
`low diffusivity of drugs; the intermediate layer is
`characterized by low lipophilicity; and the ventral
`layer is characterized by high lipophilicity. We
`suggest that the main nail barrier to drug permea-
`tion may be the low diffusivity of drugs in the
`dorsal layer. The difference between the nail plate/
`vehicle partition coefficients for 5—fluorouracil and
`flurbiprofen was small because of the low lipid
`content
`in the human nail plate. Therefore,
`this
`suggests that the human nail plate behaves like a
`hydrophilic gel membrane rather than a lipophilic
`partition membrane. This suggestion agrees with
`
`Table 3. Calculated permeation parameters of 5—fluorouracil and flurbiprofen through the human nail plate.
`Layer (thickness ratio)
`
`Drug Parameter
`
`5—Fluorouracil
`
`1
`h (pm) 7
`P (><l0 cm s_ )
`Km
`Drn (><l08cm2s_1)
`Flurbiprofen
`I
`h (#111) 6
`P (X10 ems’ )
`Km
`,
`Dm (><103cm2s'
`
`')
`
`Dorsal
`(3)
`
`148-5
`3-87
`0-48
`1-21
`
`130-2
`3-71
`1-22
`263
`
`Intermediate
`(5)
`
`247-5
`7-12
`0-56
`3-13
`
`217-0
`3-59
`0-98
`7-96
`
`Ventral
`(2)
`
`99-0
`5-39
`0-30
`1-80
`
`86-8
`11-90
`1-51
`3-42
`
`Full-thickness
`(3 : 5 :2)
`
`Each value represents the meanzl: s.c.m. (n : 3—5).Permeability coefficients of each single layer were calculated using Eq“
`5. h is the nail membrane thickness; P is the permeability coefficient; Km is the membrane/donor vehicle partition C0efl'lCl61'1t0
`drug; Drn is the diffusion coefficient of the drug in the nail membrane.
`
`atiofl
`f the
`
`CFAD V. Anacor, |PR2015-01776
`ANACOR EX. 2190 - 9/11
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2190 - 9/11
`
`

`
`DRUG PERMEATION THROUGH HUMAN NAIL PLATE
`
`277
`
`the findings of Walters et al
`Mertin & Lippold (l997a).
`
`(1983, 1985) and
`
`Estimation. of drug concentration in each layer
`For the treatment of fungal infections in the nail
`plate,
`information concerning antifungal drug
`concentrations in the nail plate, particularly the
`ventral nail plate, is of great importance. However,
`it
`jg very difficult to obtain detailed information
`about these concentrations. Consequently, the drug
`concentration in each layer was estimated from the
`permeation parameters of each drug, obtained in
`earlier studies.
`In order
`to obtain information
`regarding the treatment of fungal infection in the
`nail plate by oral antifungal medication, 5—fluoro—
`uracil and flurbiprofen permeations from the back
`of the nail plate were also investigated.
`and
`5—Fluorouracil
`(9-54j:0-72 /1% cmT2h_])
`flurbiprofen (0-13 :E O-03 pg cm_‘hT 1) fluxes from
`the ventral to the dorsal nail plate were approxi-
`mately the same values as those (5-fluorouracil:
`9-1921: 1-63 ,ug cm_2h"]; flurbiprofen: 0-l4:|:0-05
`,ugcmT2h’1) from the dorsal to the ventral nail
`plate.
`The
`amount
`of
`5—fluorouracil
`(13-98:l:l-25 ,ug)
`in the full—thickness nail plate
`from the ventral to the dorsal nail plate was also
`nearly the same as that (14-33:l:0-69 pg) from the
`dorsal
`to the ventral nail plate. However,
`the
`amounts of flurbiprofen in the full—thickness nail
`plates were not detectable in either case.
`We can calculate the drug amount in each layer
`from the permeation parameters of the drug. The
`flux of a drug through each single layer can be
`defined as follows:
`
`dQ/dt = DA(CA — CA)/hA = DB(CB — Ci3)/hB
`
`= Dc(Cc — C/C)/hC
`
`(6)
`
`where (CA ~ CA), (CB — CB) and (CC — cc) are
`the concentration differentials of that drug in each
`layer (A = dorsal; B I: intermediate; C = ventral).
`In a sink condition, C C can be assumed to be equal
`to zero. Therefore, the total amount of a drug (QT)
`in the full—thickness nail plate can be defined as
`follows:
`
`QT/S = (CA ‘ C/A)hA/2 + (CB ‘ 'l3)hB/2
`
`+ Cchc/2
`
`(7)
`
`where S is the surface area of the nail plate. We
`calculated the mean drug concentration in each
`single layer using Equations 6 and 7.
`Table 4 shows the amounts and mean con-
`
`centrations of 5—fluorouracil and flurbiprofen in
`each layer, as estimated f