research article evaluating the functionality of...

Research ArticleEvaluating the Functionality of Conjunctiva Usinga Rabbit Dry Eye Model

Yuan Ning12 Dhruva Bhattacharya2 Richard E Jones2 Fangkun Zhao1 Rongji Chen2

Jinsong Zhang1 and Mingwu Wang2

1Department of OphthalmologyThe Fourth Affiliated Hospital of ChinaMedical University Eye Hospital of ChinaMedical UniversityKey Lens Research Laboratory of Liaoning Province Shenyang 115001 China2Department of Ophthalmology and Vision Science University of Arizona College of Medicine Tucson AZ 85711 USA

Correspondence should be addressed to MingwuWang mingwuwanghotmailcom

Received 23 December 2015 Revised 17 February 2016 Accepted 17 February 2016

Academic Editor Lei Zhou

Copyright copy 2016 Yuan Ning et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Purpose To assess the conjunctival functionality in a rabbit dry eye (DE) model Methods Nictitating membrane lacrimal andHarderian glands were surgically excised frommale New Zealand white rabbits using minimally invasive surgery FluoresceinroseBengal staining of ocular surface (OS) and Schirmer test were done before (BE) and after excision (AE) The expression ofinterleukin- (IL-) 1120573 tumor necrosis factor- (TNF-)120572 andMUC5ACproteinswere estimated by immunoblotting fromconjunctivalimpression cytology specimens MUC5AC mRNA was quantified as well The effect of epithelial sodium channel (ENaC) blockerson tear production and potential differences (PD) of OSwere assessed under anesthesia in rabbits with andwithout surgery ResultsIncrease in corneal and conjunctival staining was observed 1 month AE compared to BE Schirmer tests failed to show decrease intear production Elevated IL-1120573 and TNF-120572 1 month AE indicated inflammation MUC5AC expression was elevated 1 month AEENaC blockers did not improve tear production in rabbit eyes AE but characteristic changes in PD were observed in rabbits withsurgery Conclusions DE biomarkers are important tools for OS assessment and MUC5AC expression is elevated in rabbit DE PDmeasurement revealed significant electrophysiological changes in rabbits with surgery

1 Introduction

Tear film (TF) constantly protects the exposed surface ofthe eye the cornea and the conjunctiva from environmentalstresses including desiccation temperature change physicalinjury and infections [1] By providing optimal concentra-tions of electrolytes proteins mucin and lipids the TFis critical in the maintenance of corneal transparency andgood vision [1] Dry eye disease (DED) is a multifactorialdysfunction of the TF resulting in symptoms of discomfortvisual disturbance and even loss of vision due to damageto the ocular surface [2] DED is generally acknowledgedto be in large part due to reduced secretion or increasedevaporation of the tear fluid resulting in subsequent increasein osmolarity and inflammation at the ocular surface [2]Since DED represents a diverse group of conditions thatmanifest as inadequate ocular surface lubrication restoration

of a sufficient tear volume remains themainstay of current dryeye (DE) treatment

Although lacrimal gland (LG) is considered the mainsource of tears [3] increasing evidence suggests that undercertain conditions conjunctival epitheliumhas the capacity tobe the primary source of TF [1] Removal of the main LG ofsquirrel monkeys does not lead to keratoconjunctivitis sicca(KCS) [4] In humans up to 86 of patients with epiphorawho underwent palpebral dacryoadenectomy (PDA) did notdevelop DE and in up to 50 of such patients the epiphorapersisted [5 6] Although accessory LGs were believed to bemostly responsible in these cases the conjunctiva certainlyplays a role as a compensatory tissueThe human conjunctivaoccupying 17 times more surface area than the cornea has thepotential to be the primary modulator of tear volume andcomponent [7]

Hindawi Publishing CorporationJournal of OphthalmologyVolume 2016 Article ID 3964642 9 pageshttpdxdoiorg10115520163964642

2 Journal of Ophthalmology

We are interested in understanding the physiology ofconjunctival epithelium so as to maximize its fluid secretioncapacity as an alternative to DED treatment A rabbit modelwith intact conjunctiva and equal DE phenotype bilaterallyis ideal in such research We created a DE model in rabbitsby surgical excision of the nictitating membrane (NM)Harderian gland (HG) and main LG [8] Surprisingly thetear secretion was not significantly reduced by these opera-tions Although DE associated ocular surface phenotype andinflammatory biomarkers elevated in the immediate postop-erative period they gradually decreased over 4-month dura-tion to near preoperative level without therapeutic interven-tion [8] These findings suggest that the rabbit ocular surfacecan potentially compensate for the loss of these seeminglyvital ocular surface structures including the main LG Theresults also indicate that in acute DE condition (as createdin our experiment) ocular surface injury and inflammationcan be mostly reverted To gain further insight into the exactmechanisms of conjunctiva mediated tear compensationthe present study further explored methods of conjunctivalcharacterization in this mixed mechanism rabbit DE model

2 Methods

21 Experimental Animals and Ethics Statement Male NewZealand white rabbits (119873 = 8 16 eyes Harlan Sprague Daw-ley Indianapolis IN USA) weighing 20ndash25 kg were used forthis studyThe rabbits were reared under standard laboratoryconditions (22 plusmn 2∘C 40 plusmn 5 relative humidity and a12-hour light-dark cycle) with free access to food and waterthroughout the experiment The study was conducted incompliancewith the Tenets of theDeclaration ofHelsinki andARVO statement for the use of animals in ophthalmic andvisual research The protocol was approved by the Universityof Arizona (Tucson AZ USA) Institutional Animal Careand Use Committee (protocol 14-511) All surgeries wereperformed by skilled surgeons (YN and MW)

22 Operative Procedure The surgical protocol for resectionofmain LG HG andNMwas published previously [8] whichwas modified from established procedures [9 10] Identicalprocedure was performed on the left and right eye

23 Evaluations The rabbits were assessed before excision(BE) and after excision (AE) To minimize slit lamp findingartifact from other tests the evaluations were carried out intwo days in the following sequence of each eye The first daybegins with corneal fluorescein test followed immediately byrose Bengal staining and CIC On the second day Schirmertests without (Schirmer I test SIt) and with anesthesia(Schirmer II test SIIt) were performed separately in themorning and afternoon

24 Corneal Fluorescein and Rose Bengal Staining TestsThe eyes of all rabbits were examined under a slit lampmicroscope (GR-54Gilras LLCMiami FL) by the same oph-thalmologist (YN) following protocol described previously[8]

25 Schirmer I and II Tests Both SIt and SIIt were carriedout in our study The SIt was performed as per the protocoldescribed previously [8] For SIIt one drop of 05 propara-caine hydrochloride (Bausch and Lomb Tampa FL USA)was placed and the excess fluid was blotted away with softpaper tissue prior to the insertion of the filter paper strips(Alcon Laboratories Inc FortWorth TX USA) in the lowerlateral one-third of conjunctival fornix and eyelids closed bygentle force for 5mins Both tests were performed three timesand the average score was used for analysis

26 Conjunctival Impression Cytology Conjunctival impres-sion cytology (CIC) was performed as per the protocol pub-lished [8] The filter paper discs were peeled off and immedi-ately placed in either 500120583L Trizol solution (Invitrogen CAUSA) for RNA isolation or 100 120583L of radio immunoprecip-itation assay (RIPA) buffer (Teknova CA USA) for proteinisolation

27 RNA Isolation and cDNA Synthesis Total RNA wasisolated from the CIC specimens in Trizol solution accordingto manufacturerrsquos instructions (Invitrogen CA USA) RNAconcentrations were estimated by NanoDrop ND-1000 Spec-trophotometer (NanoDrop Technologies Wilmington DEUSA) in 1 120583L volume Purity of the RNA was assessed by theratio of absorbance at 260280 nm A ratio of 19 to 2 wasconsidered to be good quality RNA specimen and used forfurther experiments The first strand of cDNA was synthe-sized with QuantiTect Reverse Transcription Kit (QiagenValencia CA USA) using 500 ng total RNA according to themanufacturerrsquos instructions

28 Reverse Transcriptase-Quantitative Polymerase ChainReaction (RT-qPCR) The RT-qPCR reactions were set usingSYBR Green PCR Master Mix (Applied Biosystems Fos-ter City CA USA) according to manufacturerrsquos instruc-tions The primer sequences for MUC5AC were as fol-lows MUC5AC-F CCCCAACGTCAAGAACAACT andMUC5AC-R TCAAACAGGCAGTTCGAGTG [11]The RT-qPCR was performed on StepOnePlus Real-Time PCRSystem (Applied Biosystems) with the following cycling con-ditions 15min 95∘C 40 cycles of 15 sec 95∘C and 30 sec 60∘CThe fluorescence was recorded during elongation step in eachcycle A melting curve analysis was performed at the end ofeach PCR by gradually increasing the temperature from 60 to95∘C while recording the fluorescence A single peak at themelting temperature of the PCR product confirmed primerspecificity To compare between different runs a fixed fluo-rescence threshold for derivation of CT value for all runs wasused Three technical replicates were performed to evaluatethe relative quantification

29 Relative Quantification ofmRNALevel Relative quantifi-cation of MUC5AC expression in rabbit CIC specimens wasperformed BE and 1 month AEThe fold change inMUC5ACexpression was relative to the internal housekeeping gene 120573-actin (endogenous control) Mean fold change in MUC5ACexpression was calculated using 2minusΔΔCT method where

Journal of Ophthalmology 3

ΔΔCT = (CTGene minus CTActin)After Excision minus (CTGene minusCTActin)Before Excision Difference between CT for MUC5ACand 120573-actin mRNA in each specimen was used to calculatelevel of target mRNA relative to that of 120573-actin mRNA in thesame specimen [8]

210 Immunoblotting Total cell lysate proteins were isolatedfrom CIC in radio immunoprecipitation assay (RIPA) bufferwith 1x HALT protease and phosphatase inhibitor singleuse inhibitor cocktail (Thermoscientific Rockford IL USA)by incubating on ice for 30min Protein concentration wasdetermined by Pierce BCA Protein Assay Kit (ThermoFisher Scientific NY) Specimens were mixed with Laemmlisample buffer (Bio-Rad laboratories Inc Hercules CAUSA)containing 120573-mercaptoethanol and heated at 95∘C for 10minSpecimens were then immunoblotted and analyzed as perthe protocol published previously [8] The primary rabbitmonoclonal antibodies to IL-1120573 TNF-120572 and MUC5AC(Abcam Cambridge MA USA) were used at a dilution of 1200

211 Effect of Epithelial Sodium Channel Blockers on Conjunc-tival Tear Secretion To test the effects of epithelial sodiumchannel (ENaC) blockers on tear secretion amiloride andbenzamil were administered topically to the right eyes (119899 = 8)of the operated rabbits 2 months AE A 01 of amilorideand benzamil [12] (both from Sigma-Aldrich Inc St LouisMO USA) were prepared in sterile buffered saline solution(BSS) and tested in separate experimentsThe right eyes wereallocated to ENaC blockers and left eyes to BSS as controlA 50120583L of ENaC blocker eye drops or BSS was instilled intothe lower conjunctival sac by a micropipette at the beginningof the experiments SIIt was performed before and at 5min15min 30min 60 and 90min after application of amilorideor benzamil

212 Open-Circuit Potential Difference Measurements at theRabbit Ocular Surface Potential difference (PD) is gener-ated by electrogenic Clminus secretion and Na+ reabsorptionacross superficial cell apical membrane of the corneal andconjunctival epithelia [13] PD measurement is a sensitivemodality in detecting transepithelial electrolyte conductanceat the ocular surface [14] Therefore to help delineate theunderlying physiological change which contributes to theincreased output of tears by the rabbit conjunctiva AEopen-circuit PD was measured with a method modifiedfrom a previously established protocol in mice [14] Brieflythe rabbits were anesthetized with 100mgkg ketamine and10mgkg xylazine (Sigma-Aldrich MO USA) and placed ona heating pad in a stereotaxic device with conjunctival andcorneal tissues exposed and faced upwards Two differentsolutions were perfused in series over the ocular surface at arate of 10mLmin using a pinch valve perfusion system (PS-8H Bioscience Tools San Diego CA USA) and peristalticpump (13-876-1 Fisher Scientific Pittsburgh PA USA) with11610158401015840 inner-diameter plastic tubing

First phosphate-buffered saline (1x PBS) was perfusedfor 5min to establish a stable baseline and then 100 120583Mamiloride (Sigma-Aldrich) prepared in 1x PBS was perfused

A low powered wall vacuum attached to 11610158401015840 ID tubingwas placed next to the fluid bolus covering the ocular surfaceto keep the volume constant and avoid fluid runoff ThePDs were measured with a high-impedance digital voltmeterIsoMilivolt Meter (World Precision Instruments SarasotaFL USA) with two AgAgCl electrodes connected througha 1M KCl agar bridge One probe was placed in contact withthe ocular fluid and the other was placed subcutaneously inthe rabbitrsquos mid-back The PDs were measured on operatedrabbit eyes (119899 = 4) 5 months AE and compared with normalrabbit eyes (119899 = 4) as controls

213 Data Analysis and Statistics Data in figures are pre-sented asmean Standard ErrorMethod the bars representingstandard errors Statistical significance between two groups(BE and AE) was evaluated using unpaired 2-tailed 119905-testA probability of 119875 equal to 005 was considered significant(where applicable lowast119875 lt 005 lowastlowast119875 lt 001 and lowastlowastlowast119875 lt0001) The Spearman correlation analysis was employed todetermine the correlation between every pair of the testsperformed BE and AE

3 Results

31 Modification of the Operative Procedure Chen et alextracted the HG through an inferior orbital rim incision [9]We found that this approach requires a long incision towardthe medial canthus In addition massive hemorrhage tendsto occur while excising the HG from between the medialrectus muscle and the anterior orbital wall Gelatin spongewas used to achieve hemostasis during their surgeries In ourstudy excision of the HG through the NM excision woundwas much less invasive Less hemorrhage and improvedvisibility of the surgical field ensured complete ablation of theHG (Figure 1) Li et al extracted HG using similar method[10] However a 5mm palpebral conjunctival incision wasmade in their study to extract lobes of the main LG In ourexperience such a small incision would not permit adequateaccess to all lobes of the LG especially the intraorbital lobewhich is deeply embedded beneath the lateral orbital rim andseparated by a dense membranous connective tissue from thesuperficial temporal lobe No additional conjunctival incisionwas necessary in our procedure and hence the entire con-junctival surface is preserved The skin incision only neededto cover the lateral two-thirds of the orbital rim in order tohave a good exposure to adequately remove the infraorbitaltemporal and intraorbital lobes of the LG A rabbit modelwith intact conjunctiva and equal DE phenotype bilaterally isideal for our research to comparatively assess modalities thatpotentially stimulate conjunctival fluid secretions

32 Ocular Surface Changes As compared to BE bothfluorescein and rose Bengal staining increased on the corneaand conjunctiva (Figure 2) 1 month AE Significantly higherstaining scores (119875 lt 00001 in both) demonstrated thepresence of DE phenotype at the ocular surface For all testsconducted there were no significant differences found as afunction of left versus right eye


(a) (b)

(c) (d)

Figure 1 Major surgical steps involved in creating our rabbit dry eye model (a) Nictitating membrane (NM) was removed at the base (b)Harderian gland was separated and ablated through same wound as excision of NM (this was done to reduce hemorrhage) (c) removal ofinfraorbital and temporal lobes of the lacrimal gland (d) removal of the deeply embedded intraorbital lobe of the lacrimal gland

BE AE0

5

10

15 Fluorescein staining

Stai

ning

scor

e

lowastlowastlowast

(a)BE AE

0

1

2

3

Stai

ning

scor

e

Rose Bengal staining

lowastlowastlowast

(b)

Figure 2Comparison of fluorescein and roseBengal staining of rabbit eyes before and 1month after surgeryTherewere significant differencesin fluorescein staining (a) and rose Bengal staining (b) (lowastlowastlowast119875 lt 00001) before excision (BE) and after excision (AE) Data are presented asmean Standard Error Method (SEM)


BE AE0

5

10

15

20

25Sc

hirm

errsquos

scor

e (m

m)

Schirmer test I

(a)BE AE

0

2

4

6

8

10

Schi

rmer

rsquos sc

ore (

mm

)

Schirmer test II

(b)

Figure 3 Comparison of Schirmer tests (I and II) before (BE) and 1 month after excision (AE) There were no significant differences inSchirmer scores BE and 1 month AE either without anesthesia (Schirmer I) or with anesthesia (Schirmer II) Data are presented as meanStandard Error Method (SEM)

AE51 kDa

37 kDa

TNF-120572

TNF-120572

Gapdh

BE BE AE35 kDaIL-1120573

IL-1120573

Gapdh 37 kDa

250 kDa

37 kDa

MUC5AC

Gapdh

AEBEMUC5AC

BE AE00

05

10

15

BE AE00

05

10

15

20

BE AE000510152025

MU

C5AC

Gap

dh

TNF-120572

Gap

dh

IL-1120573

Gap

dh

lowast

(a)

BE AE0

50

100

150

Fold

chan

ge in

expr

essio

n

BE AE0

5

10

15

Fold

chan

ge in

expr

essio

n

BE AE000510152025

Fold

chan

ge in

expr

essio

n

lowastlowast lowastlowast lowast

(b)

Figure 4 Quantification of mRNA and protein levels of inflammatory cytokines and MUC5AC in rabbit conjunctival impression cytologyspecimens BE and 1 month AE For the proteins the signal for the gene was normalized with the Gapdh signal from the same gene (a) FormRNA the fold change in expression of genes is relative to endogenous control 120573-actin (b) The data for mRNA of IL-1120573 and TNF-120572 (b)were referred from the previous publication [8]The upregulation of mRNA correlated with the increase at protein level for the inflammatorycytokines andMUC5AC Data are presented as mean Standard Error Method (SEM) For all graphs bars show standard error (SE) statisticaldifferences are shown (lowast119875 lt 005 lowastlowast119875 lt 001)

33 Schirmerrsquos Tests In our study large variations were notedin both Schirmer tests among eyes tested either BE or AEWith both SIt (119875 = 0104) and SIIt (119875 = 0478) no significantreduction in tear production was seen 1month AE (Figure 3)Therewas however significant difference between the SIt andSIIt (119875 lt 00001) either BE or AE with tear secretion beinglower under topical anesthesia

34 Upregulation of Dry Eye Biomarkers The protein levelsof DED associated inflammatory cytokines (TNF-120572 and IL-1120573) increased 1monthAE (Figure 4) which corroboratedwiththe mRNA levels of the inflammatory cytokines as reportedpreviously [8] Increase of conjunctival epithelium encodedgoblet cell-specific MUC5AC at mRNA and protein levelswere observed 1 month AE (Figure 4)


0

2

4

6

8

10

12

14

16

18

BenzamilControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 30998400 60998400 90998400

(a)

0

5

10

15

20

25

AmilorideControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 5998400 15998400 30998400 60998400

(b)

Figure 5 Effect of epithelial sodium channel blockers on rabbit DE model The application of epithelial sodium channel blockers benzamil(a) and amiloride (b) did not significantly increase the tear quantity in our rabbit DE model

35 Effect of Amiloride and Benzamil Treatment on Con-junctiva Secretion The two ENaC blockers did not increasetear secretion in our rabbit DE model as measured by SIIt(Figure 5)

36 Open-Circuit Potential Difference and Depolarizationafter Amiloride Treatment at the Ocular Surface of RabbitsThePDmeasurements for the 10 seconds before the perfusionsystem was switched from PBS to amiloride channel wereminus272plusmn6mV for rabbit eyes in the operated group (119899 = 4) andminus159 plusmn 3mV for the control group (119899 = 4) The difference inPDs was highly significant (119875 lt 0005) After the ocular sur-face was perfused with amiloride the 10-second average PDreached minus133 plusmn 4mV in the operated eyes and minus90 plusmn 4mV inthe control eyesThemagnitude of depolarizationwas statisti-cally larger (119875 lt 005) in the operated eyes than in the controleyes (Figure 6)

37 Statistical Correlations between Various Tests UsingSpearman correlation analysis higher SIIt scores are closelyassociated with lower rose Bengal test scores (negativelycorrelated correlation coefficient = minus057 119875 = 002) Addi-tionally RT-qPCR of IL-1120573 and TNF-120572were significantly cor-related (correlation coefficient = 072 119875 = 002) The changesof inflammatory biomarkers did not correlate with that ofthe clinical tests (fluorescein staining rose Bengal stainingand Schirmer tests)

4 Discussion

In our study as expected the rabbits showed increasedfluorescein and rose Bengal staining of the ocular surface 1

minus3500

minus3000

minus2500

minus2000

minus1500

minus1000

minus500

00 0 5 10 15 20 25 30 35 40

Time (s)

Pote

ntia

l diff

eren

ce (m

V)

Rabbit without surgery (control)Rabbit with surgery

Figure 6 Potential difference recordings of the rabbit eyes aftersurgery compared to rabbit eyes without surgery The potentialdifferences were recorded for the rabbit eyes 5 months AE (119899 = 4)and control eyes (119899 = 4) The perfusion channel was switched fromPBS to amiloride at 13 seconds with 2-3 seconds required for thenew solution to reach the ocular surface Data are presented asmeanStandard Error Method (SEM)

month AE characteristic of DE phenotype Interestingly nosignificant reduction was found in tear secretion by Schirmertests as compared to BE Possible explanations as to why nosignificant reduction in tear secretionwas seen after resectionof the LG HG and NM have been extensively discussed in aseparate publication [8]

It has been assumed that accessory LGs are responsiblefor the remaining tear secretion capacity in the absence of the


main LG [6 15] However increasing evidence supports thenotion that the conjunctiva can be an important contributor[5 8 16 17]The accessory LGs are embedded in the conjunc-tiva and hence the surface area of conjunctiva is substantiallylarger than the sum of secreting acinar cell surface area ofthe accessory LGs It is not unreasonable to assume that con-junctiva contributes substantial amount to the tear volumein the absence of main LG Significant difference betweenSIt and SIIt scores in our rabbit model suggests that sensoryregulation of the ocular surface plays an important role Sinceaccessory LGs have similar functions [18] and innervationsas the main LG [19] they are assumed to be under identicalreflex control [2] Although a local transcellular osmoticmechanism is believed to govern the fluid and electrolytetransport [20] fluid secretion by the conjunctiva can also bestimulated [1] The presence of parasympathetic nerves inrat conjunctiva [21] and increased conjunctiva Clminus and fluidsecretion by sympathomimetic agonists [22 23] suggests thatneural influence of conjunctiva secretion cannot be ruled outAnd if so local anesthesia of the secretory nerve terminalscould also suppress the secretion output of the conjunctivalepithelium Differences between scores of SIt and SIIt in ourstudy could reflect at least to a large extent the basal level tearsecretion from accessory LGs and the conjunctiva whereas itis very difficult if not impossible to determine the proportionof contributions from accessory LGs or conjunctiva to theremaining tear secretion capacity

Contemporary clinical assessments of DE in animalmodels have certain shortcomings Tear breakup time andcornealconjunctiva staining are extremely difficult to eval-uate objectively especially in small animals Schirmer testsresults provide no direct evidence of ocular surface damageOsmolarity test is expensive and has variable cutoffs [24]In human correlations between clinical symptoms signsof DE and diagnostic test results have been disappointingas well [25ndash29] In our study poor correlation among theclinical tests (fluorescein staining rose Bengal staining andSchirmer tests) is consistent with previous studies Molecularbiomarker based diagnostics on the other hand can offera standardized objective and precise measurement of thestatus of ocular diseases [30] and should be used as adjunctswhen possible

DED associated ocular surface inflammation [31] iscaused by increased level of inflammatory cytokines (IL-1IL-6 TNF-120572 and IL-17) in tear fluid cornealconjunctivalepithelia and increased infiltration of dendritic and T-cellsin conjunctiva [32] In our studies removal of main LGHG and NM led to inflammatory responses at the ocularsurface as depicted by increasedmRNA [8] and protein levelsof TNF-120572 and IL-1120573 Rabbits with sham surgeries did notshow significant increase in biomarker mRNA and protein(data not shown) suggesting that persistent elevation ofthese markers 1 month AE is not a direct result of surgicalprocedure itself Although there was no significant changein tear production at 1 month AE biomarker evaluationsconfirmed the increased inflammation which corroboratedwith the presence of DE phenotype at the ocular surfaceOur data is consistent with Solomon et al who demon-strated that DE is associated with increased production of

proinflammatory cytokines (IL-1 and TNF-120572) in conjunctiva[33] To the best of our knowledge overexpression of gobletcell-specificMUC5AC in response to acuteDE condition cre-ated by surgery is a novel finding in our study In associationwith the persistent normal level of tear secretion MUC5ACoverproduction likely contributed to the spontaneous recov-ery of ocular DE phenotype with time in our rabbit DEmodel [8] Gilbard et al noted reduced conjunctival gobletcell density in their rabbit DE model after cauterizing the LGexcretory duct and surgically removing the NM and HG [17]whereas with mucin-specific staining we were not able todiscern any changes in the number or morphology of gobletcells in CIC specimens BE and AE [8]The exact mechanismsof goblet cell mucin regulations in our rabbit DEmodel awaitfurther investigation

We isolated both total RNA and protein from CIC speci-mens a rapid convenient and minimally invasive techniqueto collect one to three layers of cells from bulbar conjunctivalsurface [34]The CIC has been widely performed on subjectsto confirm a variety of ocular surface diseases and monitorchanges at conjunctival surface Total RNA and protein iso-lated from CIC specimen detected subtle changes in mRNAand protein levels of the DED associated cytokines (TNF-120572 and IL-1120573) and MUC5AC Biomarkers provided objectiveand quantitative data that significantly enhanced the charac-terization of rabbit ocular surface pathology One CIC spec-imen per eye at a specific time point offered sufficient highquality total RNA and protein for analyzing several geneswithout sacrificing the animals This also enabled us to mon-itor these rabbits longitudinally and lowered experimentalcost [8]

ENaC has been shown to be present in rabbit conjunc-tiva [35] Shi and Candia concluded that the electrogenicNa+ reabsorption across rabbit conjunctiva was amiloride-insensitive [36] indicating the important roles played by Na+dependent cotransporters such as those carrying glucose andamino acids in series with the basolaterally located Na+-K+ pump Hara et al recently demonstrated increased tearsecretion as measured by Schirmer test after the applicationof amiloride at the rabbit ocular surface [37] However wewere not able to reproduce their results in our rabbit modelEven using more potent ENaC inhibitor benzamil [12] nosignificant increase in tear productionwas seen in the presentstudy We concluded that Schirmer test given its large varia-tion between measurements may not be sensitive enough todetect subtle change in tear productionTherefore we furtherassessed the baseline ocular PD and its response to the appli-cation of amiloride in rabbit eyes with and without surgerySignificantly higher (more negative) PD in the operatedrabbit eyes was noted in comparison to eyes without surgerySince electrogenic Clminus secretion and Na+ reabsorption acrosssuperficial cell apical membrane of the corneal and conjunc-tival epithelia contribute to the PD [13] the ocular surfacetissues must have reached a new equilibrium of higher Clminussecretion andor Na+ reabsorption Higher magnitude ofPD depolarization in the operated eyes in response to theapplication of amiloride indicates the presence of an elevatedamiloride-sensitive Na+ conductance (reabsorption) acrossthe epithelia Although amiloride-insensitive higher Na+


reabsorptionmechanism could not be measured in the studyit presumably exists Likewise a higher Clminus conductance(secretion)most probably is present aswell Our PDmeasure-ments demonstrate electrophysiological support of highertear output across the ocular surface in rabbit eyes withoutLG HG and NM

To summarize in this rabbit DED model althoughSchirmer tests were unchanged BE and AE analysis ofbiomarkers corroborated with the clinical examination find-ings and confirmed the development ofDE condition Assess-ing DED pertinent biomarkers enhanced the results obtainedfrom standard clinical tests and is a valuable addition to thetools of ocular surface evaluation It was interesting to notethe elevated MUC5AC expression in the acute DE conditioncreated by surgery but its mechanism requires further inves-tigation Nomeasurable increased tear secretionwas detectedwith Schirmer test with topical application of amiloride inrabbit eyes AE However the open-circuit PD measure-ment provided a sensitive modality to detect the underlyingelectrophysiological changes at the rabbit ocular surfaceAE

Competing Interests

The authors declare that they have no conflict of interests

Authorsrsquo Contributions

Yuan Ning and Dhruva Bhattacharya contributed equally

Acknowledgments

This work was supported by Research to Prevent BlindnessFoundation and Department of Ophthalmology and VisionScience at University of Arizona The authors are also thank-ful to the Animal Care Facility at the University of Arizonafor the care of their animals throughout this study

References

[1] D A Dartt ldquoRegulation of mucin and fluid secretion by con-junctival epithelial cellsrdquo Progress in Retinal and Eye Researchvol 21 no 6 pp 555ndash576 2002

[2] ldquoThe definition and classification of dry eye disease Report ofthe definition and classification subcommittee of the interna-tional Dry Eye WorkShop (2007)rdquo Ocular Surface vol 5 no 2pp 75ndash92 2007

[3] LThorig E J Van Agtmaal E Glasius K L Tan and N J VanHaeringen ldquoComparison of tears and lacrimal gland fluid in therabbit and guinea pigrdquo Current Eye Research vol 4 no 8 pp913ndash920 1985

[4] D Y Maitchouk R W Beuerman T Ohta M Stern and R JVarnell ldquoTear production after unilateral removal of the mainlacrimal gland in squirrel monkeysrdquoArchives of Ophthalmologyvol 118 no 2 pp 246ndash252 2000

[5] A Hornblass C Guberina and B J Herschorn ldquoPalpe-bral dacryoadenectomy for epiphorardquo Ophthalmic Plastic andReconstructive Surgery vol 4 no 4 pp 227ndash230 1988

[6] W Scherz and C H Dohlman ldquoIs the lacrimal gland dispens-able Keratoconjunctivitis sicca after lacrimal gland removalrdquoArchives of Ophthalmology vol 93 no 4 pp 281ndash283 1975

[7] M A Watsky M M Jablonski and H F Edelhauser ldquoCom-parison of conjunctival and corneal surface areas in rabbit andhumanrdquo Current Eye Research vol 7 no 5 pp 483ndash486 1988

[8] D Bhattacharya Y Ning F Zhao et al ldquoTear production afterbilateral main lacrimal gland resection in rabbitsrdquo InvestigativeOpthalmology amp Visual Science vol 56 no 13 pp 7774ndash77832015

[9] Z-Y Chen Q-F Liang andG-Y Yu ldquoEstablishment of a rabbitmodel for keratoconjunctivitis siccardquo Cornea vol 30 no 9 pp1024ndash1029 2011

[10] N Li X Deng Y Gao S Zhang M He and D ZhaoldquoEstablishment of themildmoderate and severe dry eyemodelsusing three methods in rabbitsrdquo BMC Ophthalmology vol 13no 1 article 50 2013

[11] T Nakamura Y Hata M Nagata et al ldquoJBP485 promotestear and mucin secretion in ocular surface epitheliardquo ScientificReports vol 5 Article ID 10248 2015

[12] A J Hirsh J R Sabater A Zamurs et al ldquoEvaluation of secondgeneration amiloride analogs as therapy for cystic fibrosis lungdiseaserdquo Journal of Pharmacology and Experimental Therapeu-tics vol 311 no 3 pp 929ndash938 2004

[13] M H Levin and A S Verkman ldquoAquaporins and CFTR inocular epithelial fluid transportrdquo Journal of Membrane Biologyvol 210 no 2 pp 105ndash115 2006

[14] M H Levin and A S Verkman ldquoCFTR-regulated chloridetransport at the ocular surface in living mice measured bypotential differencesrdquo Investigative Ophthalmology and VisualScience vol 46 no 4 pp 1428ndash1434 2005

[15] J P Gilbard S R Rossi K G Heyda and D A Dartt ldquoStim-ulation of tear secretion by topical agents that increase cyclicnucleotide levelsrdquo Investigative Ophthalmology and Visual Sci-ence vol 31 no 7 pp 1381ndash1388 1990

[16] M H I Shiue A A Kulkarni H J Gukasyan J B Swisher K-J Kim and V H L Lee ldquoPharmacological modulation of fluidsecretion in the pigmented rabbit conjunctivardquoLife Sciences vol66 no 7 pp PL105ndashPL111 2000

[17] J P Gilbard S R Rossi and K L Gray ldquoA new rabbit modelfor keratoconjunctivitis siccardquo Investigative Ophthalmology andVisual Science vol 28 no 2 pp 225ndash228 1987

[18] J L Ubels I K Gipson S J Spurr-Michaud A S Tisdale R EVan Dyken and M P Hatton ldquoGene expression in humanaccessory lacrimal glands of Wolfringrdquo Investigative Ophthal-mology and Visual Science vol 53 no 11 pp 6738ndash6747 2012

[19] P Seifert S Stuppi and M Spitznas ldquoDistribution patternof nervous tissue and peptidergic nerve fibers in accessorylacrimal glandsrdquo Current Eye Research vol 16 no 4 pp 298ndash302 1997

[20] O A Candia and L J Alvarez ldquoFluid transport phenomena inocular epitheliardquo Progress in Retinal and Eye Research vol 27no 2 pp 197ndash212 2008

[21] J D Rıos D Zoukhri I M Rawe R R Hodges J D Zieskeand D A Dartt ldquoImmunolocalization of muscarinic and VIPreceptor subtypes and their role in stimulating goblet cellsecretionrdquo Investigative Ophthalmology and Visual Science vol40 no 6 pp 1102ndash1111 1999

[22] H C Turner L J Alvarez and O A Candia ldquoCyclic AMP-dependent stimulation of basolateral K+ conductance in therabbit conjunctival epitheliumrdquo Experimental Eye Research vol70 no 3 pp 295ndash305 2000


[23] U B Kompella K-J Kim M H I Shiue and V H LLee ldquoCyclic AMP modulation of active ion transport in thepigmented rabbit conjunctivardquo Journal of Ocular PharmacologyandTherapeutics vol 12 no 3 pp 281ndash287 1996

[24] M Li C Du D Zhu M Shen L Cui and J Wang ldquoDaytimevariations of tear osmolarity and tear meniscus volumerdquo Eyeand Contact Lens vol 38 no 5 pp 282ndash287 2012

[25] B Caffery R L Chalmers H Marsden et al ldquoCorrelation oftear osmolarity and dry eye symptoms in convention attendeesrdquoOptometry and Vision Science vol 91 no 2 pp 142ndash149 2014

[26] M A Lemp A J Bron C Baudouin et al ldquoTear osmolarity inthe diagnosis and management of dry eye diseaserdquo The Amer-ican Journal of Ophthalmology vol 151 no 5 pp 792ndash798e12011

[27] B Sullivan ldquoChallenges in using signs and symptoms to evalu-ate new biomarkers of dry eye diseaserdquo Ocular Surface vol 12no 1 pp 2ndash9 2014

[28] B D Sullivan L A Crews E M Messmer et al ldquoCorrelationsbetween commonly used objective signs and symptoms for thediagnosis of dry eye disease clinical implicationsrdquo Acta Oph-thalmologica vol 92 no 2 pp 161ndash166 2014

[29] F Amparo Y Jin P Hamrah D A Schaumberg and R DanaldquoWhat is the value of incorporating tear osmolarity mea-surement in assessing patient response to therapy in dry eyediseaserdquoAmerican Journal of Ophthalmology vol 157 no 1 pp69ndash77e2 2014

[30] G N Foulks and S C Pflugfelder ldquoNew testing options fordiagnosing and grading dry eye diseaserdquo American Journal ofOphthalmology vol 157 no 6 pp 1122ndash1129 2014

[31] W Stevenson S K Chauhan and R Dana ldquoDry eye diseasean immune-mediated ocular surface disorderrdquo Archives ofOphthalmology vol 130 no 1 pp 90ndash100 2012

[32] C S De Paiva S Chotikavanich S B Pangelinan et al ldquoIL-17disrupts corneal barrier following desiccating stressrdquo MucosalImmunology vol 2 no 3 pp 243ndash253 2009

[33] A Solomon D Dursun Z Liu Y Xie A Macri and S CPflugfelder ldquoPro- and anti-inflammatory forms of interleukin-1 in the tear fluid and conjunctiva of patients with dry-eyediseaserdquo Investigative Ophthalmology andVisual Science vol 42no 10 pp 2283ndash2292 2001

[34] M Mrugacz I Kasacka A Bakunowicz-Lazarczyk M Kacz-marski andW Kulak ldquoImpression cytology of the conjunctivalepithelial cells in patients with cystic fibrosisrdquo Eye vol 22 no9 pp 1137ndash1140 2008

[35] U B Kompella K-J Kim and V H L Lee ldquoActive chloridetransport in the pigmented rabbit conjunctivardquo Current EyeResearch vol 12 no 12 pp 1041ndash1048 1993

[36] X-P Shi and O A Candia ldquoActive sodium and chloridetransport across the isolated rabbit conjunctivardquo Current EyeResearch vol 14 no 10 pp 927ndash935 1995

[37] S Hara A Hazama M Miyake et al ldquoThe effect of topicalamiloride eye drops on tear quantity in rabbitsrdquo MolecularVision vol 16 pp 2279ndash2285 2010

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom


We are interested in understanding the physiology ofconjunctival epithelium so as to maximize its fluid secretioncapacity as an alternative to DED treatment A rabbit modelwith intact conjunctiva and equal DE phenotype bilaterallyis ideal in such research We created a DE model in rabbitsby surgical excision of the nictitating membrane (NM)Harderian gland (HG) and main LG [8] Surprisingly thetear secretion was not significantly reduced by these opera-tions Although DE associated ocular surface phenotype andinflammatory biomarkers elevated in the immediate postop-erative period they gradually decreased over 4-month dura-tion to near preoperative level without therapeutic interven-tion [8] These findings suggest that the rabbit ocular surfacecan potentially compensate for the loss of these seeminglyvital ocular surface structures including the main LG Theresults also indicate that in acute DE condition (as createdin our experiment) ocular surface injury and inflammationcan be mostly reverted To gain further insight into the exactmechanisms of conjunctiva mediated tear compensationthe present study further explored methods of conjunctivalcharacterization in this mixed mechanism rabbit DE model

2 Methods

21 Experimental Animals and Ethics Statement Male NewZealand white rabbits (119873 = 8 16 eyes Harlan Sprague Daw-ley Indianapolis IN USA) weighing 20ndash25 kg were used forthis studyThe rabbits were reared under standard laboratoryconditions (22 plusmn 2∘C 40 plusmn 5 relative humidity and a12-hour light-dark cycle) with free access to food and waterthroughout the experiment The study was conducted incompliancewith the Tenets of theDeclaration ofHelsinki andARVO statement for the use of animals in ophthalmic andvisual research The protocol was approved by the Universityof Arizona (Tucson AZ USA) Institutional Animal Careand Use Committee (protocol 14-511) All surgeries wereperformed by skilled surgeons (YN and MW)

22 Operative Procedure The surgical protocol for resectionofmain LG HG andNMwas published previously [8] whichwas modified from established procedures [9 10] Identicalprocedure was performed on the left and right eye

23 Evaluations The rabbits were assessed before excision(BE) and after excision (AE) To minimize slit lamp findingartifact from other tests the evaluations were carried out intwo days in the following sequence of each eye The first daybegins with corneal fluorescein test followed immediately byrose Bengal staining and CIC On the second day Schirmertests without (Schirmer I test SIt) and with anesthesia(Schirmer II test SIIt) were performed separately in themorning and afternoon

24 Corneal Fluorescein and Rose Bengal Staining TestsThe eyes of all rabbits were examined under a slit lampmicroscope (GR-54Gilras LLCMiami FL) by the same oph-thalmologist (YN) following protocol described previously[8]

25 Schirmer I and II Tests Both SIt and SIIt were carriedout in our study The SIt was performed as per the protocoldescribed previously [8] For SIIt one drop of 05 propara-caine hydrochloride (Bausch and Lomb Tampa FL USA)was placed and the excess fluid was blotted away with softpaper tissue prior to the insertion of the filter paper strips(Alcon Laboratories Inc FortWorth TX USA) in the lowerlateral one-third of conjunctival fornix and eyelids closed bygentle force for 5mins Both tests were performed three timesand the average score was used for analysis

26 Conjunctival Impression Cytology Conjunctival impres-sion cytology (CIC) was performed as per the protocol pub-lished [8] The filter paper discs were peeled off and immedi-ately placed in either 500120583L Trizol solution (Invitrogen CAUSA) for RNA isolation or 100 120583L of radio immunoprecip-itation assay (RIPA) buffer (Teknova CA USA) for proteinisolation

27 RNA Isolation and cDNA Synthesis Total RNA wasisolated from the CIC specimens in Trizol solution accordingto manufacturerrsquos instructions (Invitrogen CA USA) RNAconcentrations were estimated by NanoDrop ND-1000 Spec-trophotometer (NanoDrop Technologies Wilmington DEUSA) in 1 120583L volume Purity of the RNA was assessed by theratio of absorbance at 260280 nm A ratio of 19 to 2 wasconsidered to be good quality RNA specimen and used forfurther experiments The first strand of cDNA was synthe-sized with QuantiTect Reverse Transcription Kit (QiagenValencia CA USA) using 500 ng total RNA according to themanufacturerrsquos instructions

28 Reverse Transcriptase-Quantitative Polymerase ChainReaction (RT-qPCR) The RT-qPCR reactions were set usingSYBR Green PCR Master Mix (Applied Biosystems Fos-ter City CA USA) according to manufacturerrsquos instruc-tions The primer sequences for MUC5AC were as fol-lows MUC5AC-F CCCCAACGTCAAGAACAACT andMUC5AC-R TCAAACAGGCAGTTCGAGTG [11]The RT-qPCR was performed on StepOnePlus Real-Time PCRSystem (Applied Biosystems) with the following cycling con-ditions 15min 95∘C 40 cycles of 15 sec 95∘C and 30 sec 60∘CThe fluorescence was recorded during elongation step in eachcycle A melting curve analysis was performed at the end ofeach PCR by gradually increasing the temperature from 60 to95∘C while recording the fluorescence A single peak at themelting temperature of the PCR product confirmed primerspecificity To compare between different runs a fixed fluo-rescence threshold for derivation of CT value for all runs wasused Three technical replicates were performed to evaluatethe relative quantification

29 Relative Quantification ofmRNALevel Relative quantifi-cation of MUC5AC expression in rabbit CIC specimens wasperformed BE and 1 month AEThe fold change inMUC5ACexpression was relative to the internal housekeeping gene 120573-actin (endogenous control) Mean fold change in MUC5ACexpression was calculated using 2minusΔΔCT method where









3 Results




(a) (b)

(c) (d)


BE AE0

5

10


Stai

ning

scor

e

lowastlowastlowast

(a)BE AE

0

1

2

3

Stai

ning

scor

e


lowastlowastlowast

(b)



BE AE0

5

10

15

20

25Sc

hirm

errsquos

scor

e (m

m)

Schirmer test I

(a)BE AE

0

2

4

6

8

10

Schi

rmer

rsquos sc

ore (

mm

)

Schirmer test II

(b)


AE51 kDa

37 kDa

TNF-120572

TNF-120572

Gapdh


IL-1120573

Gapdh 37 kDa

250 kDa

37 kDa

MUC5AC

Gapdh

AEBEMUC5AC

BE AE00

05

10

15

BE AE00

05

10

15

20

BE AE000510152025

MU

C5AC

Gap

dh

TNF-120572

Gap

dh

IL-1120573

Gap

dh

lowast

(a)

BE AE0

50

100

150

Fold

chan

ge in

expr

essio

n

BE AE0

5

10

15

Fold

chan

ge in

expr

essio

n

BE AE000510152025

Fold

chan

ge in

expr

essio

n


(b)





0

2

4

6

8

10

12

14

16

18

BenzamilControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 30998400 60998400 90998400

(a)

0

5

10

15

20

25

AmilorideControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 5998400 15998400 30998400 60998400

(b)





4 Discussion


minus3500

minus3000

minus2500

minus2000

minus1500

minus1000

minus500

00 0 5 10 15 20 25 30 35 40

Time (s)

Pote

ntia

l diff

eren

ce (m

V)















Competing Interests




Acknowledgments


References












































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
























3 Results




(a) (b)

(c) (d)


BE AE0

5

10


Stai

ning

scor

e

lowastlowastlowast

(a)BE AE

0

1

2

3

Stai

ning

scor

e


lowastlowastlowast

(b)



BE AE0

5

10

15

20

25Sc

hirm

errsquos

scor

e (m

m)

Schirmer test I

(a)BE AE

0

2

4

6

8

10

Schi

rmer

rsquos sc

ore (

mm

)

Schirmer test II

(b)


AE51 kDa

37 kDa

TNF-120572

TNF-120572

Gapdh


IL-1120573

Gapdh 37 kDa

250 kDa

37 kDa

MUC5AC

Gapdh

AEBEMUC5AC

BE AE00

05

10

15

BE AE00

05

10

15

20

BE AE000510152025

MU

C5AC

Gap

dh

TNF-120572

Gap

dh

IL-1120573

Gap

dh

lowast

(a)

BE AE0

50

100

150

Fold

chan

ge in

expr

essio

n

BE AE0

5

10

15

Fold

chan

ge in

expr

essio

n

BE AE000510152025

Fold

chan

ge in

expr

essio

n


(b)





0

2

4

6

8

10

12

14

16

18

BenzamilControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 30998400 60998400 90998400

(a)

0

5

10

15

20

25

AmilorideControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 5998400 15998400 30998400 60998400

(b)





4 Discussion


minus3500

minus3000

minus2500

minus2000

minus1500

minus1000

minus500

00 0 5 10 15 20 25 30 35 40

Time (s)

Pote

ntia

l diff

eren

ce (m

V)















Competing Interests




Acknowledgments


References












































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















(a) (b)

(c) (d)


BE AE0

5

10


Stai

ning

scor

e

lowastlowastlowast

(a)BE AE

0

1

2

3

Stai

ning

scor

e


lowastlowastlowast

(b)



BE AE0

5

10

15

20

25Sc

hirm

errsquos

scor

e (m

m)

Schirmer test I

(a)BE AE

0

2

4

6

8

10

Schi

rmer

rsquos sc

ore (

mm

)

Schirmer test II

(b)


AE51 kDa

37 kDa

TNF-120572

TNF-120572

Gapdh


IL-1120573

Gapdh 37 kDa

250 kDa

37 kDa

MUC5AC

Gapdh

AEBEMUC5AC

BE AE00

05

10

15

BE AE00

05

10

15

20

BE AE000510152025

MU

C5AC

Gap

dh

TNF-120572

Gap

dh

IL-1120573

Gap

dh

lowast

(a)

BE AE0

50

100

150

Fold

chan

ge in

expr

essio

n

BE AE0

5

10

15

Fold

chan

ge in

expr

essio

n

BE AE000510152025

Fold

chan

ge in

expr

essio

n


(b)





0

2

4

6

8

10

12

14

16

18

BenzamilControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 30998400 60998400 90998400

(a)

0

5

10

15

20

25

AmilorideControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 5998400 15998400 30998400 60998400

(b)





4 Discussion


minus3500

minus3000

minus2500

minus2000

minus1500

minus1000

minus500

00 0 5 10 15 20 25 30 35 40

Time (s)

Pote

ntia

l diff

eren

ce (m

V)















Competing Interests




Acknowledgments


References












































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















BE AE0

5

10

15

20

25Sc

hirm

errsquos

scor

e (m

m)

Schirmer test I

(a)BE AE

0

2

4

6

8

10

Schi

rmer

rsquos sc

ore (

mm

)

Schirmer test II

(b)


AE51 kDa

37 kDa

TNF-120572

TNF-120572

Gapdh


IL-1120573

Gapdh 37 kDa

250 kDa

37 kDa

MUC5AC

Gapdh

AEBEMUC5AC

BE AE00

05

10

15

BE AE00

05

10

15

20

BE AE000510152025

MU

C5AC

Gap

dh

TNF-120572

Gap

dh

IL-1120573

Gap

dh

lowast

(a)

BE AE0

50

100

150

Fold

chan

ge in

expr

essio

n

BE AE0

5

10

15

Fold

chan

ge in

expr

essio

n

BE AE000510152025

Fold

chan

ge in

expr

essio

n


(b)





0

2

4

6

8

10

12

14

16

18

BenzamilControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 30998400 60998400 90998400

(a)

0

5

10

15

20

25

AmilorideControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 5998400 15998400 30998400 60998400

(b)





4 Discussion


minus3500

minus3000

minus2500

minus2000

minus1500

minus1000

minus500

00 0 5 10 15 20 25 30 35 40

Time (s)

Pote

ntia

l diff

eren

ce (m

V)















Competing Interests




Acknowledgments


References












































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















0

2

4

6

8

10

12

14

16

18

BenzamilControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 30998400 60998400 90998400

(a)

0

5

10

15

20

25

AmilorideControl

Time (min)

Schi

rmer

rsquos te

st sc

ore (

mm

)

0998400 5998400 15998400 30998400 60998400

(b)





4 Discussion


minus3500

minus3000

minus2500

minus2000

minus1500

minus1000

minus500

00 0 5 10 15 20 25 30 35 40

Time (s)

Pote

ntia

l diff

eren

ce (m

V)















Competing Interests




Acknowledgments


References












































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of


























Competing Interests




Acknowledgments


References












































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
















research article evaluating the functionality of...

Documents