The pH tolerance of rabbit and human
`corneal endothelium
`
`Russell Gonnering, Henry F. Edelhauser, Diane L. Van Horn, and William Durant
`
`The endotheliums of rabbit corneas were pe1fusecl i11 cm in vitro perf11sio11 specular microscope
`up to 3 hr with solutions varying in pH from 3.5 to 10.0. Corneal thickness was mo11itored
`throughout the expeliment, and at appropriate times the corneas were prepared for SEM and
`TEM. Analysis of the corneal thickness data and i11terpretatio11 of the electron micrographs
`reveals that outside of the pl-I range of 6.5 to 8.5, structural a11cl fu11ctional alteratio11s occur.
`Direct cellular damage, as well as disruption of junctional complexes, lead to a l1reakclown in
`the barrier function of the corneal endothelium. The extent of this breakdow11 is dependent
`upon both the 11wgnitude of the pH change and the exposure time. Further experime11ts on
`banked l11111w11 eyes support this finding.
`
`Key words: corneal endothelium, pH, human corneal endothelium, specular
`microscope, intraocular drugs
`
`with the advent of new surgical tech(cid:173)
`
`niques such as vitrectomy, phakoemulsifica(cid:173)
`tion, and intraocular lens implantation, the
`anterior chamber of the eye is exposed to an
`increasing number of solutions, drugs, pre(cid:173)
`servatives, vehicles, and implants. The pHs
`2
`of these materials range from 3.6 1 to 8.2. 1
`•
`The effect of these agents upon the pH of
`the anterior chamber contents is, of course,
`dependent upon the manner in which they
`are used. For example, during a surgical pro(cid:173)
`cedure such as anterior chamber reconstruc(cid:173)
`tion, the aqueous humor will be, for all in-
`
`From the Departments of Ophthalmology and Physiol(cid:173)
`ogy, The Medical College of Wisconsin, and the Re(cid:173)
`search Service, Wood Veterans Administration Cen(cid:173)
`ter, Milwaukee, Wis.
`This study was supported in part by National Eye Insti(cid:173)
`tute Research Grants EY-00933 and I-P30-01931, by a
`contract from Alcon Laboratories, Inc., and by the
`Medical Research Service of the Veterans Adminis(cid:173)
`tration.
`Submitted for publication March 22, 1978.
`Reprint requests: Henry F. Edelhauser, Ph.D., De(cid:173)
`partment of Physiology, The Medical College of Wis(cid:173)
`consin, P.O. Box 26509, Milwaukee, Wis. 53226.
`
`tents, replaced by the irrigating solution, and
`thus the pH of the anterior chamber will be
`that of the irrigating solution . Alterations of
`aqueous humor pH produced by exogenous
`substances also depend on a number of other
`factors such as rate of aqueous humor forma(cid:173)
`tion, buffering capacity of the ocular fluids,
`initial pH, buffering capacity of the exoge(cid:173)
`nous substance, and the length of time the
`substance remains in the anterior chamber.
`There are other clinical situations in which
`the pH of the anterior chamber is profoundly
`altered over extended periods of time. Or(cid:173)
`lowski and Wekka 1 first showed that the pH
`of the anterior chamber increases to 10 fol(cid:173)
`lowing a 1 min alkaline burn with IM NaOH
`and remains at that level for up to 1 ½ hr.
`Feller and Grauper and their co-workers 2
`7
`-
`Subsequently described the influence of con(cid:173)
`centration, duration of contact, and various
`treatment modalities upon the pH of the an(cid:173)
`terior chamber following experimental acid
`and alkali burns. Similar experiments by
`Paterson et al. 8 have further explored these
`interrelationships.
`The purpose of this study was to determine
`
`373
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 1 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`374 Gonnering et al.
`
`Invest. Opl1thalmol. Visual Sci.
`A/Jri/ 1979
`
`Table I. Composition of perfusion media
`
`Component compounds
`
`NaCl
`KCI
`CaCl, · 2H,O
`MgCl, · 6H,O
`Sodium acetate (C,HaNaO, · 3H,O)
`Sodium citrate (Cofl. 5Na 3O 7 • 2H 2O)
`NaH,PO4
`Glycine (NH,CH,COOH)
`
`Solutions
`
`A
`pH 3.5-8.5
`(mmol/L)
`
`109.5
`10.l
`3.3
`1.5
`28.7
`5.8
`
`B
`pH 7.4-10
`(mmol/L)
`
`143.8
`4.8
`0.8
`0.8
`
`0.9
`20.0
`
`Table II. Mean corneal swelling rates for
`rabbit experiments
`
`Mean swelling rates (mean + S.E.M.) (µm!hr)
`
`Solution A
`( citrate-acetate)
`>230 (n = 3)
`62.8 ::!: 8.9 (n = 4)
`58.l ::!: 2.2 (n = 4)
`36.3 ::!: 1.2 (n = 11)
`36.0 ::!: 3.8 (n = 4)
`30.4 ::!: 0.9 (n = 23)
`34.3 ::!: 1.6 (n = 3)
`31.4 ::!: 1.3 (n = 4)
`
`pH
`
`3.5
`5.5
`6.0
`6.5
`7.0
`7.4
`8.0
`8.5
`9.0
`9.5
`10.0
`
`Solution B
`(glycine)
`
`41.2 ::t 1.6 (n = 11)
`
`53.5 ::!: 2.4 (n = 4)
`69.4 ::!: 4.3 (n = 4)
`252.5 ::!: 11.9 (n = 3)
`
`the pH range in which corneal endothelial
`function and ultrastructure are safely main(cid:173)
`tained. These data could then serve as a
`guideline for a more rational formulation of
`drugs, vehicles, and irrigating solutions as
`well as give new insight into the pathophysi(cid:173)
`ology of chemical burns of the cornea.
`
`Materials and methods
`Rabbit experiments. New Zealand albino rab(cid:173)
`bits, weighing 2 to 3 kg, were sacrificed, their eyes
`were enucleated, and the corneas were mounted
`in a dual-chambered perfusion specular micro(cid:173)
`scope. 9· 1° Corneal endothelial perfusions with ex(cid:173)
`perimental solutions at 37° C were performed with
`a Harvard pump at a flow rate of 0.0097 ml/min at
`a pressure of 15 mm Hg. Corneal thickness was
`measured at intervals of 5 to 15 min. At the con(cid:173)
`clusion of each experiment, the corneas were fixed
`in 2. 7% glutaraldehyde in phosphate buffer (pH
`7.2, 330 mOsm) for at least 8 hours at 4° C . The
`
`corneas were then cut in half and postfixed in 2%
`osmium tetroxide for 2 hr. Small pieces of half of
`each cornea were embedded in a low-viscosity
`epoxy medium for transmission electron micros(cid:173)
`copy (TEM). The other half of each cornea was
`prepared for scanning electron microscopy (SEM)
`of the endothelium according
`to a modified
`method of Cleveland and Schneider. 11 They were
`penetrated with the low-viscosity resin , poly(cid:173)
`merized overnight at 37° C and for 48 hr at 60° C,
`glued to stubs, sputter-coated with gold-palladi(cid:173)
`um, and viewed with an AMR 1000 scanning elec(cid:173)
`tron microscope.
`The composition of the perfusing solution for
`the first set of experiments is given as solution A in
`Table I. It consisted of a five-salt solution with a
`citrate-acetate buffer and had an osmolarity of 312
`to 320 mOsm. The composition of this solution was
`similar to that of vehicles currently used for in(cid:173)
`traocular drugs and did not include glucose,
`glutathione, and bicarbonate ion that are present
`in "ideal" irrigating solutions . The pH was ad(cid:173)
`justed to experimental levels immediately before
`each experiment by the addition of NaOH or HCl
`and was measured on a Corning Model 12 pH
`meter. One cornea of each pair was perfused at the
`experimental pH; the other was pe1-fused at pH
`7.4 to serve as a control. The pH of the effi ux
`perfusing solution was also monitored to ensure
`maintenance of the proper pH throughout the ex(cid:173)
`periment.
`The pH of solution A could be adjusted and
`maintained through a range of3.5 to 8.5; however,
`above pH 8.5, the solution did not retain its buf(cid:173)
`fering capacity. Solution B (utilizing a glycine buf(cid:173)
`fer but also lacking glucose, glutathione, and bi(cid:173)
`carbonate ion) was therefore used (Table I) to ex(cid:173)
`pand the alkaline pH range up to 10.0.
`Corneal thickness measurements of all exper-
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 2 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`Volllme 18
`Number 4
`
`pH tolerance of rabbit, human corneal endothelium 375
`
`iments at each pH were subjected to linear re(cid:173)
`gression analysis for swelling rates are listed in
`Table II.
`Human experiments. Human eyes were ob(cid:173)
`tained from the Wisconsin Lions Eye Bank. The
`age of the donors was 64.9 ± 8.5 years; time from
`death to enucleation was 3.5 :±: 0. 7 hr; time from
`enucleation to experiment was 29.5 :±: 7.6 hr. The
`eyes were transported and stored in moist cham(cid:173)
`bers at 4° C prior to experimentation.
`The human eyes were mounted, perfused, and
`fixed in a manner identical to that used for the
`rabbit corneas. Three pairs of eyes were perfused
`at each experimental pH value. One eye of the
`pair was perfused with solution A (Table I) at the
`experimental pH; the other served as a control and
`was perfused with BSS Plus , a modified glutathi(cid:173)
`one bicarbonate Ringer's solution . 12
`Although BSS Plus contains glutathione , bicar(cid:173)
`bonate, and glucose, which is not present in solu(cid:173)
`tion A, this protocol was dictated by the varied
`biological state of the human corneas (type of
`death, elapsed time , eye bank c.-onditions, etc.), a
`factor not present in the rabbit experiments. By
`observing the response of the control cornea to
`perfusion with a solution known to maintain the
`integrity of human corneal endothelium, one can
`ascertain the initial biological state of the pair and
`separate experimental results due to the experi(cid:173)
`mental conditions from those due to pre-existing
`endothelial damage.
`Because of this variability inherent in experi(cid:173)
`ments using banked human eyes, the thickness
`measurements were pooled and standardized by
`comparing each individual c.-ornea to its mated
`wntrol. The change in thickness vs. time was de(cid:173)
`termined for both the experimental cornea and its
`control in each experiment, and the difference be(cid:173)
`tween the two was calculated. If the expe1imental
`eye increased in thickness more than the control,
`the difference had a positive sign. The mean dif~
`ferences for each experimental group were ob(cid:173)
`tained and subjected to linear regression analysis.
`The swelling rates therefore represent differences
`in changes in thickness compared to c.-ontrols.
`
`Results
`Rabbit experiments
`Changes in thickness. Control corneas per(cid:173)
`fused with solution A at pH 7.4 swelled at
`30.4 ± 0.9 µm/hr (Table II). The rates of
`swelling of corneas perfused at pH 6.5, 7.0,
`8.0, and 8.5 varied from 30.4 to 36.3 µm/hr,
`whereas the rates of swelling at pH 5.5, 6.0,
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`300
`
`UJ
`
`~ 40
`
`(!)
`Z
`
`30 i 20
`
`Solution 8
`( Glycine Buffer I
`
`Solution A
`(Citrate-Acetate Suffer)
`
`0
`
`2
`
`4
`
`6
`pH
`
`8
`
`10
`
`12
`
`14
`
`Fig. l. Logarithmic plot of the corneal swelling
`rate vs . pH of the endothelial perfusion media for
`rabbit experiments . The minimal swelling oc(cid:173)
`curred over a pH range of 6.5 to 8.5.
`
`and 3.5 were much greater than that at pH
`7.4 and illustrate that the endothelial barrier
`fonction has been
`lost. Extremely rapid
`swelling (greater than 230 µm/hr) occurred
`with perfusion at pH 3.5.
`Control corneas perfused with solution B
`at 7.4 swelled at 41.2 ± 1.6 µ.m/hr. Increas(cid:173)
`ing the alkalinity of the perfusate to 9.0, 9.5,
`and 10.0 resulted in swelling rates that were
`significantly greater than the control value
`(p > 0.001). Perfosions at pH 10.0 resulted
`in a swelling rate of 252. 5 ± 11. 9 µ.m/hr.
`When these corneal swelling rates at the
`various experimental pHs are plotted, a
`parabolic curve results (Fig. 1) which illus(cid:173)
`trates that the least amount of swelling occurs
`within the range of pH 6.5 to 8.5. On either
`side of this range, a rapid increase in control
`swelling occurs which can be associated with
`progressive ultrastructural alterations in the
`corneal endothelial cells and is dependent
`upon the exposure time.
`Electron microscopy. Control corneas per(cid:173)
`fused at pH 7.5 for 3 hr with either solution A
`(Fig. 2, A) or solution B (Fig. 3, A) retained
`their normal endothelial mosaic pattern on
`SEM. TEM revealed that ultrastructural
`morphology was normal in the endothelial
`cells of these corneas, except for a minimal
`amount of swelling in the basal cytoplasm
`(Figs. 2, B and 3, B).
`
`Slayback Exhibit 1057, Page 3 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`376 Gormerir1g et al .
`
`/11uesr. Oplttha/1110/. Vis11al Sr.i.
`April 1979
`
`Fig. 2. A, Scanning electron micrograph of the endothelium of a rabbit cornea perfused with
`solution A (citrate-acetate buffer) at pH 7.40 for 3 hr. The mnsaic-like pattern is normal.
`( x 1000.) 13 , Trnnsmission electron micrograph of e ndothelial cells from the same cornea show(cid:173)
`ing that normal ultrastructure is maintained exc.-ept for some . light swelling of the basal
`cytopla.~m. ( Xl0,600.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 4 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`Vo/11111 c 18
`N,onl.u.,-r 4
`
`71H tolern11ce of rabbit , human corneal enclothelimn 377
`
`Fig. 3. A Scanning electron micrograph of the endothelium of a rabbit cornea perfused with
`solution B (glycine huffer) at pH 7.40 for 3 hr. The normal mosaic-like cell11 l.1r pattern is
`preserved. ( xJO0O.) B Transmission electron micrograph of endothelial cells from the same
`cornea displaying normal ultrastructure except for some ch1rification of the basal cytoplasm .
`( xl0,600.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 5 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`378 Gonnering et al.
`
`/,west . Oplit/10/1110/. Vi,•11al Sci.
`April J979
`
`Fig. 4. A, Scanning e lectron microgrnph of the endothelium of a rabbit cornea perfused with
`solution A at pH 3.50 for 5 min . The e ndothelial cells appear to be swollen , forming prominent
`clefts in the apical surface. ( x 1000.) B, Transmission electron micrograph of endothelial cells
`from the same cornea showing C)1toplasmic swelling and vacuolization, di stortion of the nu(cid:173)
`cleus , clumping of the nuclear chromatin, dilation of the rough endoplasmic reticulum inter(cid:173)
`ce llular space , disruption of the mitochondria, and irregular oute r plasma membranes .
`(X9200.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 6 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`Vo/111110 18
`Number 4
`
`711-J tof.era11ce of rabbit , h11man corneal endothelium 379
`
`Fig. 5. A, Scanning e lectron micrograph of the endothelium or a rabbit cornea perfused with
`solution A at pH 3 .50 for 15 min . The cells appear to be collapsed a nd disphly disruptio ns in the
`pos te rior surfac.-e . ( x 1000.) B, Transmission e lectron micrograph or the same endothelium
`revealing completely necrotic c:eUs . The re are no rec.-ognizable organelles and the outer pl.lSma
`me mbrane shows many discontinuities. ( X 10,500.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 7 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`380 Gon11eri11g et al.
`
`Invest . Oplitlin/1110/. Visual Sci .
`1\11ril 1979
`
`Fig. 6. A and B, For legend, see facing page.
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 8 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`Volume 18
`Numln-r 4
`
`pH tolerance of rabbit, human corneal endothelium 381
`
`Five minutes of endothelial perfusion at
`pH 3.5 produced alterations in the surface
`morphology of the endothelial cells (Fig. 4,
`A). TEM examination showed cytoplasmic
`swelling, vacuolization, distortion of the nu(cid:173)
`cleus, clumping of nuclear chromatin, dila(cid:173)
`tion of the rough endoplasmic reticulum,
`widening of the intercellular spaces, disrup(cid:173)
`tion of mitochondria, and very irregular outer
`plasma membranes with prominent apical
`clefts (Fig. 4, B) . These changes progressed
`rapidly to cell lysis after 15 min of exposure
`(Fig. 5, A and B).
`The endothelium of corneas perfused at
`pH 5.5 for 30 min also showed changes in
`endothelial morphology by SEM (Fig. 6, A).
`Upon TEM examination there was marked
`cytoplasmic and nuclear swelling, clumping
`of the nuclear chromatin, dilation of the
`rough e ndoplasmic reticulum, condensation
`of the mitochondria, irregular outer plasma
`membranes , and straightened intercellular
`junctions with apical flaps . Extending the
`pe1fosion time to 1 hr at pH 5. 5 revealed lysis
`and cell death similar to those occurring at
`pH 3.5. Corneas perfused at pH 6.0 showed
`similar ultrastructural changes at 2 hrs , with
`time
`that
`lysis beyond
`endothelial cell
`period.
`Corneal endothelium pe rfused at pH 6.5
`showed a variable appearance after 3 hr.
`Normal e ndothelial morphology was main(cid:173)
`tained in half of th e corneas, wh ereas in
`others the endothelial cells appeared swollen
`with cytoplasmic blebbing, straight cellular
`junctions with apical flaps, irregular outer
`plasma membranes, and central blebs com(cid:173)
`posed of vesiculated cytoplasm (Fig. 7, A
`and B).
`Normal endothelial morphology was main(cid:173)
`tained in those corneas perfused for 3 hr at
`pH values of7.0, 8.0, 8.5, and 9.0. After 3 hr
`of perfi.tsion at pH 9.5, however, numerous
`
`microvilli were seen with SEM and TEM re(cid:173)
`vealed the presence of basal cytoplasmic
`swelling, cytoplasmic vacuoles, and conden(cid:173)
`sation of the rnitochond1ia (Fig. 8, A and B).
`Increasing the pH to 10.0 revealed a
`dramatic effect on cellular morphology. Af'ter
`15 min of perfusion , the cells were swollen
`and had pulled apart at the apical junction
`(Fig. 9, A) . Thes e structural changes are
`consistent with the high rate of corneal swell(cid:173)
`ing (See Table II) . With TEM the cells them(cid:173)
`selves were seen to be necrotic, with a loss of
`organelles and discontinuity of the outer
`plasma membranes (Fig. 9, B) .
`Human experiments
`Changes in thickness. Corneas perfosed at
`pH 5.5, 6.5, and 8.4 increased in thickness at
`rates of93.4 ± 18.5, 17.8 ± 3.7, and 18.8 ±
`3.5 µm / hr, respectively over their BSS Plus
`controls (Fig. 10).
`The use of an experimental perfusate dif(cid:173)
`fering from the control only in pH , though
`desirable , was impossible for these studies.
`The bicarbonate necessary for the mainte(cid:173)
`nance of normal cell function precludes the
`possibility of adjusting the pH. The use,
`however, of a control solution without bicar(cid:173)
`bonate would not allow for a check on the
`physiological integ1ity of the corneal pair,
`in(cid:173)
`thus pre-existing damage could
`and
`validate the results.
`In this expe1imental model, one cannot as(cid:173)
`cribe all the changes seen in the experimen(cid:173)
`tal corneas to only differences in pH as com(cid:173)
`pared to controls . However, even if a direct
`comparison to controls, as in the rabbit ex(cid:173)
`periments, is impossible, a comparison of the
`swelling rates of the experimental groups to
`each other is instructive. The human corneas
`perfo sed at pH 6.5 and 8.4 behaved si milarly
`to each other, with a relatively low swelling
`rate , whereas the corneas perfused at pH 5. 5
`exhibited a significantly higher rate of swell-
`
`Fig. 6. A, Scanning electron micrograph of the endothelium of a rabbit cornea perfused with
`solution A at pH 5.50 for 30 min . The ce lls appear to be greatly swollen, and the junctions
`appear altered. ( XlOOO.) B, Transmission electron micrograph of endothelial cells from the
`same cornea revealing marked cytoplasmic and nuclear swelling, clumping of the nuclear
`chromatin, dilation of th e rough e ndoplasmic reticulum , condensation of the mitochondtia,
`irregular outer plasma membranes, and straight cellular junctions with apical Aaps . ( x l0,000.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 9 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`382 Gonnering et al.
`
`!,west. 011Mliu/1110/. \lia,10/ Sci.
`A11ril 1979
`
`Fig. 7. A, Scan ning electro n micrograph of the endothe lium of :t r.ihbit co rnea perfuse d with
`solution A at pH 6.50 for 3 hr . The cells appe<1r to be swollen, with a central bleb protruding
`from a majority of the m. Pits are present between ·ome cells. ( Xl,000.) B, Trnnsmission
`e lectron micrograph of the endothelial cells from the same cornea displaying cytoplasmic
`swelling, straight cellular junctions with apical !laps , and very irregular outer plasma mem(cid:173)
`branes. Blebs composed of vesicu lated, condensed cytoplasm project outward from th e nuclear
`region of mos t cells. ( X9,300.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 10 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`Volu me 18
`N1,mbcr 4
`
`1JH tolerance of rabbit , lwman corneal endotheliu m 383
`
`Fig. 8. A, Scanning e lectron micrograph of the endothe lium of a rabbi t cornea perfused with
`solution Bat pH 9.50 for 3 hr. The endothe li.il cells are in L'lct, wilh the normal mosiac pattern .
`{x l0OO .) B, Transmissiou electron micrograph showing numerous micnwilli , some vacuoliza(cid:173)
`tion , and condensation of the mitochondria. ( Xl4,000.)
`
`Downloaded from iovs.arvojournals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 11 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`384 Go1m ering et al.
`
`lri cat. 011/11/,almol. Vi,110/ Sci .
`April 1979
`
`Fig. 9. A, Sca11ni11g electron micrograph of the e nclotlie lium of a rabbit cornea perfused with
`solution B at pH 10:00 for 15 min . The c:ells appear to be swollen and to have separa ted .
`( XlOOO.) B, Transmission electron mi crogr;1ph of t·he same endothelium revealing completely
`necrotic cells. There are no rec.-og nizable organelles, and the other plasma membrane shows
`man y <list'O ntinuities and is very irregular due to the collapse of the junc:tional complexes .
`(xll,600.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 12 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`Volume 18
`Number 4
`
`pH tolerance of rabbit, human corneal endothelium 385
`
`ing. These results support the pH range of
`6.5 to 8.5 found in the rabbit experiments.
`Electron microscopy. The endothelium of
`human corneas perfused with the control so(cid:173)
`lution (BSS Plus) at pH 7.4 retained a normal
`ultrastructure throughout a 3 hr period (Fig.
`11, A and B) . The only abnormality was slight
`basal cytoplasmic swelling, which was prob(cid:173)
`ably due to postmortem time and indicated
`that none of the corneas had pre-existing en(cid:173)
`dothelial damage . In contrast to this, the cells
`exposed to pH 5.5 for 1 hr showed necrosis,
`with loss of cellular organelles and disruption
`of the outer plasma membrane (Fig. 12, A
`and B). When the pH of the perfusion media
`was increased to 6.5 and the time of perfusion
`extended to 3 hr, the cells were able to main(cid:173)
`tain their integ,ity except for cytoplasmic and
`mitochrond,ial swelling, clumping of the
`nuclear chromatin, and partial disruption of
`the junctional complexes (Fig. 13, A and B).
`TEM and SEM of the endothelium of
`human corneas perfused at pH 8.5 revealed
`varying degrees of cytoplasmic swelling. The
`cellular junctions were partially disrupted,
`with large spaces between many of the cells
`(Fig. 14, A and 8) .
`
`Discussion
`Throughout a 3 hr period, functional and
`ultrastructural integrity was maintained in
`the endothelium of rabbit corneas perfused
`with solutions ranging in pH from 6.5 to 8.5.
`On either side of this pH range, corneal
`thickness increased dming the perfusion and
`in
`endothelial ultrastructure deteriorated
`proportion to both the severity of the pH
`change and the length of exposure.
`The rabbit experiments were conducted
`with a salt solution purposely chosen to be of
`similar composition to vehicles used for in(cid:173)
`traocular drugs. The basal swelling rate of 30
`to 40 µ,m/hr seen in those corneas perfused
`with solutions in the pH range 6.5 to 8.5 is
`therefore not surprising. This basal swelling
`rate could have been prevented if a more
`complete medium had been used . 13 How(cid:173)
`ever, the use of such a medium , containing
`glucose and bicarbonate, would not have al(cid:173)
`lowed a direct application of the results to th e
`
`HUMAN CORNEAS
`
`~ 1~0
`z
`0 i 80
`;;/, 60
`.... z
`a: 8 40
`;:::;
`ff] 20
`Cl z O
`~
`<J -20
`
`0
`
`2
`HOURS
`
`3
`
`Fig . 10. Changes in human corneal thickness
`when perfused at pH 5.5, 6.5, and 8.5 compared
`to control (dashed line). The solid lines at each pH
`are statistically fit regression lines. "N" for each
`group is 3.
`
`clinical situation , and therefore it was not
`employed.
`These findings were confirmed in the sec(cid:173)
`ond experimental system, in which human
`corneas were perfused with an identical so(cid:173)
`lution . The use ofa complete medium, i.e.,
`BSS Plus , for the perfusion of the human con(cid:173)
`trol corneas was necessary because we were
`unable to assess the viability of the en(cid:173)
`dothelium prior to the perfusions . Such fac(cid:173)
`tors as age, postmortem time, and pre(cid:173)
`existing endothelial disease all contribute to
`the marked heterogeneity of banked human
`corneas as compared to rabbit corneas. Per(cid:173)
`fusion of the controls with the complete
`medium enabled us to determine that the
`endothelium was viable and that changes ob(cid:173)
`served during the course of the experiment
`were indeed caused by the perfusion solution
`and were not merely a consequence of pre(cid:173)
`existing damage.
`The clinical implications of this study are of
`some importance. The results indicate that
`through a 3 hr exposure time , no deleterious
`effects upon the corneal endothelium will re(cid:173)
`sult from exposure to intraocular solutions,
`vehicles, and drugs if they have a pH of from
`6.5 to 8.5, with the necessary ions for main(cid:173)
`tenance of endothelial function , i.e., Na+,
`K+, Cl-, Ca++ , and Mg++ . Manipulation of
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 13 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`386 Go,mer·ing el al.
`
`l11cu t . Oplo tlrnlmol. Vis11al Sci .
`A,,ril 1979
`
`Fig. 11 . A, Scanning e lectron microgrnph of the e 11clotheli11m of a human cornea perfused with
`BSS Plus , pH 7.2, for 3 hr. The 11ormal mosaic-like cellular pattern is prese1ved . ( XlOOO.) B,
`Transmission electron micrograph of endothe lial cells from the same <.'Ornea displaying normal
`ultrash·ucture except for some clarification of the cytoplasm between the nucleus an<l Desce(cid:173)
`met's membrane. ( x8600 .)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 14 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`Vul1111te 18
`Numbc,· 4
`
`JJfl toleran ce of rabhit, himum corneal endothelium 387
`
`Fig. 12. A, Scann ing electron micrograph of the endothelium of a human cornea perfused with
`solution A at pH 5.50 for l hr. The nuclei of most cells are exposed due to the collapse of the
`oute r phmna membrane . ( xlOOO.) B, Tra ns mission electron mic.:rograph of e ndothelial cells
`from the s:rn1e wrnea displaying co mplete!)' necrotic cell s. There are no ret'Ognizable organell(cid:173)
`es, ,md the outer plasma membrane is com pl etely disrupted . ( x ll ,200.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 15 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`388 Gonnering et al.
`
`t11 ucst. 01,l11ha/111o/. Visual Sci .
`A 1iril 1979
`
`Fig. 13. A, Scanning electron micrograph of the endothelium of a h uman <.-ornea perfused with
`solu tion A at pH 6.50 for 3 hr. The cells appear to be swolle n, .ind pits are present at nrnny of
`the junctions. ( X 1000.) B, Transmission electron micrograph of endothelial c.-ells Ii-om the same
`cornea disph,ying both cytoplasmic and mitochondrial swelling and some dumping of the
`nuclear chromatin . The junctional t-omplexes have partially broken down , leaving large spaces
`be tween many cells. ( X8600.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 16 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`Volume 18
`Numlu.,· 4
`
`pH tolerance of rnbbit, human corneal e,ulothelium 389
`
`©
`
`Fig. 14. A, Scanning e lectron rnicrograph oft he endothelium of a human cornea perfused \,~th
`solution A at pH 8.50 for 3 hr. Some or the cells appear to be swoll en , whereas others appear
`quite normal. Pits are present at many of the cellular junctions. (XlOOO.) 8, Tran smission
`electron micrograph of the endothelial cells from the same cornea showing normal ul(cid:173)
`trnsh·ucture except for varyi ng degrees of cytoplasmic swelling. The junctional complexes have
`partially broken down , leaving large spaces between many cells. (x6800.)
`
`Downloaded from iovs.arvojoumals.org on 09/09/2022
`
`Slayback Exhibit 1057, Page 17 of 18
`Slayback v. Eye Therapies - IPR2022-00142
`
`

`

`390 Gonnering et al.
`
`Invest. Oplit/,a/1110/. Vis,,a/ Sci.
`1\pril 1979
`
`the pH of the anterior chamber outside this
`range is probably still acceptable, as long as
`the exposure time is sufficiently shortened to
`balance the magnitude of the alteration. The
`clinician must, of course, be mindful of al(cid:173)
`tered aqueous humor dynamics as well as
`pre-existing endothelial disease, both of
`which would figure into the balancing of the
`pH change and exposure time.
`This study is also of interest in the man(cid:173)
`agement of severe chemical burns of the eye.
`Successful therapy might be that which not
`only normalizes the pH of the external ocular
`tissues but also attends to the severe distur(cid:173)
`in
`the anterior chamber
`bance of pH
`Paracentesis of the anterior chamber, alone
`or with an irrigation solution containing a buf(cid:173)
`fer, has been shown in the laboratory to be
`the most rapid and certain method for the
`normalization of the anterior chamber pH. 7
`8
`•
`On the basis of the present study, which doc(cid:173)
`uments the destructive damage to the cor(cid:173)
`neal endothelium that occurs at nonphysio(cid:173)
`logical pH, a method of bringing the pH of
`the aqueous humor into the range of endo(cid:173)
`thelial tolerance should be advocated.
`
`We thank Harlan J. Pederson and Linda Mader for
`their technical assistance during this study. The scanning
`electron photomicrographs were taken by Mr. Marion
`C. Hatchell at Mid-American Microanalysis Laboratory,
`Inc .• Milwaukee, Wisc.
`
`REFERENCES
`1. Orlowski, W.J., and Wekka, Z.: La concentration
`des ions d'hydrogene clans l'humeur aqueuse des
`yeux 1,..;_,les par la soude, Ophthalmologica 137:244,
`1959.
`2. Feller, K., and Craupner, K.: Die Wasserstoffion(cid:173)
`konzentration in der Vorderkammer des Kanin(cid:173)
`chenauges nach experimenteller Veratzung mit
`Siiure and L.~uge, Albrecht von Craefes Arch. Klin.
`Exp. Ophthalmol. 170:373-376, 1966.
`
`3. Feller, K . . and Craupner, K.: Die Wasserstoffion(cid:173)
`konzentration in der Verderkammer des Kanin(cid:173)
`chenauges nach experimenteller Veriitzung mit
`Calcium hydroxyd und deren BeeinRussung durch
`verschiedene PufTersubstanzen, Albrecht von Crae(cid:173)
`fes Arch. Klin. Exp. Ophthalmol. 173:71, 1967.
`4. Craupner, O.K.: Die Anderung des pH-Wertes in
`der Vorderkammer des Kaninchenauges nach Ver(cid:173)
`atzung and seine BeeinHussung durch therapeu(cid:173)
`tische Massnahmen, Acta Biol. Med. Germ. 21:871,
`1968.
`5. Craupner, O.K., and Hausmann, C.M.: Die An(cid:173)
`derung des Kaninchen.wges nach Venitzung mit
`kleinsten Mengen labortiblicher Konzentrationen
`VOil Saure and L.~uge, Albrecht VOil Craeles Arch.
`Klin. Exp. Ophthalmol. 180:60, 1971.
`6. Craupner, 0. K. : Die Bedeutung des Hornhautepi(cid:173)
`thels fiir die Anderung des pH-Wertes in der Vor(cid:173)
`derkammer des Kaninchenauges nach experimen(cid:173)
`teller Veriitzung mit Siiure, Albrecht von Craefes
`Arch. Klin. Exp. Ophthalmol 181:65, 1970.
`7. Craupner, O.K.: Die Wirkung einer Ablassung des
`Kammerwassers nach Veriitzung mit Siiure und
`Lauge auf die Wasserstoffionkonzentration in der
`Vorderkammer des Kaninchenauges, Albrecht von
`Craeles Arch. Klin. Exp. Ophthalmol. 186:67, 1973.
`8. Paterson, C.A. , Pfister, R.R., and Levinson, R.A.:
`Aqueous humor pH changes after experimental al(cid:173)
`kali burns, Am. J. Ophthalmol. 79:414, 1975.
`9. Dikstein, S., and Maurice, D. M.: The metabolic
`basis to the Ruid pump in the cornea, J. Physiol.
`221:29, 1972.
`10. McCarey, B. E., Edelhauser, H.F., and Van Horn,
`D.L.: Functional and structuml changes in the cor(cid:173)
`neal endothelium during in vitro perfusion, INVEST.
`OPIITflALMOL. 12:410, 1973.
`11. Cleveland, P.H .. and Schneider, C. W.: A simple
`method of preserving ocular tissue for scanning elec(cid:173)
`tron microscopy, Vision Res. 9:1401, 1969.
`12. Edelhauser, H.F., Connering, R.S., and Van Horn,
`D.L.: Intraocular irrigating solutions: a comparative
`study of BSS Plus and lactated Ringers solution,
`Arch. Oph