Heterogeneity in infrastructure and Elemental
Composition of Perinuclear Lens Retrodots
Gijs F. J. M. Vrensen,* Ben Willekens,* Paulus T. V. M. De Jong,f
G. Adrien Shun-Shin,X Nicholas P. Brown,X and Anthony J. Bron%
Purpose. To unravel the cataractogenic process(es) leading to the birefringent lenticular bodies
known as perinuclear retrodots.
Methods. Ten human lenses containing biomicroscopically verified perinuclear retrodots were
systematically screened and analyzed using scanning electron microscopy and energy dispersive x-ray microanalysis to verify their ultrastructure and elemental composition.
Results. Three types of retrodots were distinguished, different in size, ultrastructure, and
origin. Two of them contained calcium phosphate, the third probably contained calcium oxalate. All three types were separated from surrounding normal fibers and the crystalline inclusions were sequestered within membrane-lined bodies.
Conclusions. Because of these observations and data found in the literature it is postulated that
elevated free calcium is the initiating factor in the formation of retrodots, trapped by either
oxalate or phosphate and sequestered in the retrodots. It is suggested that the oxalate is
derived from ascorbate because of impaired protection against oxidative stress in the older
lens. Phosphoric acid is believed to be released by calcium-induced hydrolysis of membrane
phospholipids. Invest Ophthalmol Vis Sci. 1994; 35:199-206
1 he biomicroscopic features of perinuclear retrodots in human lenses have been extensively described
by Bron and Brown1 using focal, slit, and red reflex
examination and specular microscopy. Retrodots
proved to be common perinuclear features of human
lenses after the fifth decade of life and to be significantly associated with central nuclear scatter.2 They
are round, oblong, or oval bodies with apparent sizes
between 80 and 500 /urn and show birefrigence. They
were found in all cortical layers and their numbers
varied between a few and several hundred per lens.
Because of the biomicroscopic resemblance with the
"Spharolithen" of the German literature1 and the birefringent bodies observed by Harding et al3 by X-ray
From the * Department of Morphology, The Netherlands Ophthalmic Research
Institute, Amsterdam, The Netherlands, the ^Department of Ophthalmology,
Erasmus University Rotterdam, Rotterdam, The Netherlands, and the %Clinical
Cataract Research Unit, Nuffield Laboratory of Ophthalmology, University of
Oxford, Oxford, United Kingdom.
This study was carried out within the framework of the European Community
Concerted Action on Cellular Aging and Diseases "EURAGE".
Submitted for publication: March 5, 1993; revised May 26, 1993; accepted June 7,
1993.
Proprietary interest category: N.
Reprint requests: Prof. Dr. Gijs F.J.M. Vrensen, Department of Morphology, The
Netherlands Ophthalmic Research Institute, P.O. Box 12141, 1100 AC
Amsterdam, The Netherlands.
microanalysis it was proposed that retrodots also contain calcium oxalate crystals. It was argued1 that the
oxalic acid moiety of the crystals mainly stems from the
breakdown of ascorbic acid as a result of the less effective protection against oxidative stress of the elderly
lens.
Data on ultrastructure and elemental composition
of lens inclusions are rather scarce,3"6 and much of the
evidence in favor of calcium oxalate as the crystalline
inclusion in retrodots is circumstantial. The purpose
of the current study was to investigate the ultrastructure and elemental composition of retrodots in a series
of well-screened human lenses. Scanning electron microscopy (SEM) and energy dispersive x-ray (EDX) microanalysis were used as analytical tools. EDX systems
with windowless detectors have become available,
which, in addition to the detection of phosphoros (P),
sulphur (S), and calcium (Ca), also allow the detection
of carbon (C), nitrogen (N), and oxygen (O). Theoretically this enables discrimination between organic and
inorganic salts of calcium. The study was performed
on lenses obtained from extracapsular cataract extraction and screened in vivo for the presence of retrodots
(Oxford lenses) and on donor lenses and extracapsular
Investigative Ophthalmology & Visual Science, January 1994, Vol. 35, No. 1
Copyright © Association for Research in Vision and Ophthalmology
199
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933177/ on 06/15/2017
200
Investigative Ophthalmology & Visual Science, January 1994, Vol. 35, No. 1
cataract extraction lenses obtained from the Cornea
Bank Amsterdam and the Department of Ophthalmology of the Erasmus University, Rotterdam showing
identical inclusions.
MATERIALS AND METHODS
Six lenses obtained after extracapsular cataract extraction and examined at the slit lamp for the presence
of retrodots using the Oxford criteria1 (Fig. 1) were
used in this study. In addition, two donor lenses obtained from the Cornea Bank Amsterdam and two extracapsular cataract extraction lenses from Rotterdam
were used, which proved to contain identical inclusions on biomicroscopic and SEM screening. The extracapsular cataract extraction lenses had additional
cataractous features. The donor lenses were free of
these.
The six lenses from Oxford were immediately
fixed in a 0.1 M phosphate-buffered solution of 5%
glutaraldehyde (P-GL), pH 7.3. The lenses from Rotterdam and the donor lenses were fixed in a 0.08 M
cacodylate buffered solution of 1.25% glutaraldehyde
and 1% paraformaldehyde (C-GL/P), pH 7.3. Fixation
was for several weeks. The lenses were dissected and
pieces containing retrodots and minimal other disturbances were selected for further treatment. These
pieces were dehydrated in a graded series of ethanols
and critically point dried using CO2 as the intermediate. The uncoated specimens were inspected and
analyzed in a Philips SEM 505 scanning microscope
(Philips Industries, Eindhoven, The Netherlands)
equipped with an EDAX PV 9800 microanalysis system using a windowless ECON detector (EDAX, PV
9760/26; EDAX-Itl:, Mahwah, NJ). A high tension of
15 kV was used throughout and EDX signals were obtained from spot measurements (spot size 100 to 200
nm). After completion of the EDX analysis the specimens were coated with a thin (5 nm) layer of platinum
and the EDX-analyzed sites or identical structures
were photographed.
RESULTS
The results of the systematic SEM and EDX inspection
of the retrodots from all ten lenses are summarized in
Figures 2 to 9. On the basis of these observations three
types of retrodots can be distinguished. Apart from
size differences no other differences could observed
on biomicroscopy.
Type 1 retrodots (Figs. 2, 3, 4) are characterized
by the smooth lining of their dome or discus-shaped
(Fig. 2A), cavity studded with ridges that are continuous with membranous elements crisscrossing the cavity (Figs. 2C, 3B, 3C). In addition they contain globular elements of irregular form (Figs. 2B, 4B, 4C). In
many instances this globular content is lost during
fracturing for SEM. In comparison to EDX spectra of
normal control regions (insert in Fig. 2A) the Type 1
retrodot spectra [insert in Fig. 2B) exhibit enhanced S
and Na peaks and a lower O peak. In addition, pronounced P and Ca peaks appear. The smooth lining
and the fibers immediately underlying it (insert Fig.
2C) also show increased S and Na peaks and a somewhat lower O peak. A significant Ca peak is present
but no P peak was observed. The observations illustrated in Figure 4 tentatively allow the conclusion that
Type 1 retrodots originate from wormlike membrane
aberrations on individual fibers, which accumulate or
form small globular elements that grow into mature
retrodots. The EDX spectra of these early Type 1 retrodots are similar to that illustrated in Figure 2B (insert).
FIGURE l. In vivo retroillumination photograph of a human
lens with retrodots. The rounded oval features are the retrodots of variable size. Note the reversal of background illumination within the retrodots, ie, background lighter on the
right, retrodots lighter on the left. The bright so-called corneal reflex is an artifact from the flash.
Type 2 retrodots (Figs. 5, 6, 7) are dome- or discus-shaped and their cavity lining has a rough texture
(Figs. 5B, 6B). They contain numerous (Fig. 6A) complex crystals, varying in size between 15 and 30 nm
consisting of mutually anchored platelike elements
(Figs. 5B, 6B). EDX spectra (insert Fig. 5A) of these
crystals clearly show very pronounced Ca and P peaks,
and small Na, O, and C peaks (at high Ca concentrations two K,a-peaks at 3.69 and 4.04 keV were observed). With respect to their pathogenesis Figure 7
allows the conclusion that they probably originate
from coalescing, disrupting membranes of a number
of stacked fibers (Fig. 7; A, B, and C), which subsequently form a multilamellar pocket within which the
crystals are formed (Fig. 7, D and E). The early stages
of Type 2 retrodots (Fig. 7C, insert) exhibit pronounced Ca (double peaks) and P peaks but in addi-
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933177/ on 06/15/2017
Ultrastructure and Elemental Composition of Retrodots
201
accumulations (Fig. 8, A and B) of irregular pillarlike
elements (Fig. 8C). EDX spectra (Fig. 8C, insert) revealed that these elements only have pronounced Ca,
O, and C peaks. No ultrastructural bodies representing early stages of this type of retrodots were found.
Incidentally (Fig. 9) one or two closely adjacent
retrodots were observed having ultrastructural and
EDX characteristics of Type 2 and Type 3 retrodots.
In keeping with the biomicroscopic observations
the SEM observations revealed that all three types of
retrodots were well-demarcated lenticular bodies separated from the surrounding normal fibers. They are
lined by what apparently seem to be membranes that
segregate the content from neighboring unaffected
fibers. With respect to their size, Type 1 retrodots are
the smallest, ranging from 25 to 50 ixm. Type 2 retrodots range from 100 > 500 fim and Type 3 retrodots
have a maximal size of approximately 300 ^m. Of the
ten lenses systematically surveyed, five proved to contain only Type 1 retrodots, and two contained only
Type 2 retrodots. Types 1 and 2 were observed together in one lens and two lenses contained all three
FIGURE 2. Panel illustrating different ultrastructural aspects
of Type 1 retrodots. (A) Survey micrograph illustrating the
dome-shaped appearance of unfractured Type 1 retrodots
(arrowheads) surrounded by normal fibers. SL = suture line.
(90001,P-GL). (B) Intermediate Type 1 retrodot (approximately 30 ^m) containing irregularly shaped material
surrounded by fibers with normal membranes. (90001,PGL). (C) Large Type 1 retrodot (approximately 55 ^m) the
content of which is lost during SEM preparation and which
is surrounded by typical membranes of deep cortical fibers
studded with grooves and ridges (arrowheads). The retrodot
cavity is lined by a smooth membrane, discontinuous with
the underlying normal fibers (white arrow) and studded with
ridges. (91112,P-GL). The x-ray spectra of normal regions
(black and white dots, A and B) show carbon (C), oxygen (O),
sodium (Na), and sulphur (S) peaks. The lining of the cavity
and the underlying fibers exhibit an enhanced S peak and a
Ca peak (asterisks, C). The Type 1 retrodot content has in
addition to C, O, Na, S, and Ca peaks a pronounced phosphor os (P) peak, (asterisks, B).
tion still have significant S and Na peaks. The C and O
peaks are less prominent as in control material, but are
higher than in the pure crystals.
Type 3 retrodots (Fig. 8) are dome-shaped structures lined by smooth radially coursing fiberlike elements (Fig. 8A) most likely pinched off from neighboringfibers.The content of these retrodots, which is also
often lost during SEM preparation, consists of large
FIGURE 3. In cross section (A) Type 1 retrodots proved to be
spindle-shaped and its cavity to be filled with irregularly
shaped material (arrows, A and C) intermingled with membranous elements (open arrows, A and C). The ridges on the
lining of the cavity (arroiuheads, B) proved to be continuous
with these membranous elements (arrowheads, C). Note that
the fibers adjacent to the retrodots have normal groove-andridge studded membranes (asterisks, B and C). (A: 90001,PGL; B: 90001,P-GL; C:88260,C-GL/P).
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933177/ on 06/15/2017
202
Investigative Ophthalmology 8c Visual Science, January 1994, Vol. 35, No. 1
content is sequestered from the adjacent fibers. The
segregation of these aberrant lenticular inclusions is
largely analogous to that found in early radial opacities
described by Vrensen and Willekens10 and Vrensen et
al11 and may be taken as an additional support of the
view of Bron and Brown12 that there is a strong tendency in the lens to seal off damaged portions from
undamaged portions, i.e., those that have an annealing
mechanism.
A number of authors913"16 have argued that the
main component of these crystalline inclusions is calcium oxalate. Harding et al,3 using the same analytical
tools as in the current study, (SEM and EDX analysis),
concluded that the birefringent lenticular bodies contain calcium oxalate. Our Type 3 retrodots have similar EDX spectra as those described by Harding et al,3
FIGURE 4. Panel illustrating the possible origin of Type 1 retrodots. In regions of these retrodots fibers are observed
with wormlike imprints (arrowheads, A) closely associated
with regions containing small irregular elements (asterisks,
A). In a further stage small accumulations of irregularly
shaped material (asterisks, B and C) lined by smooth surfaced
membranes (arrowhead, B) are found. X-ray microanalysis
shows peaks identical to that observed in Type 1 retrodots
(cf. Fig. 2B). (A: 89205,P-GL; B: 90001,P-GL; C: 90199,CGL/P).
types of retrodots. All retrodots were located in the
intermediate and deep cortex and were not found in
the nucleus. Types 1 and 2 retrodots were observed in
the posterior, anterior, and equatorial regions of the
lens, whereas Type 3 was restricted to the equatorial
region.
DISCUSSION
The current systematic SEM study reveals that the perinuclear retrodots described previously,1 which are
identical to the "Spharolithen" of the German literature7"9 and the calcium containing birefringent bodies
of Harding et al,3 may be subdivided into at least three
subtypes of varying size: Type 1, 25 to 50 ju.m; Type 3,
50 up to 300 nm; and Type 2 100 to more than 500
(im. Biomicroscopic observations as illustrated in Figure 1 only show the retrodots to vary in size. They are
clearly different with respect to their ultrastructural
appearance and two of them may have a different
pathogenic origin. In keeping with the biomicroscopic
observations all three have a dome- or discus-shaped
outline and are well demarcated from the surrounding
unaffected lens parts. The SEM observations allow the
conclusion that they are segregated by membranes
from the neighboring unaffected fibers and that the
FIGURE 5. Ultrastructure of Type 2 retrodots. In the survey
picture (A) of a partly fractured retrodot the discoid shape
can be appreciated. The cavity is partly filled by a rough,
textured surface (arrowheads, B) and is surrounded by normal fibers (A, B). In cross section (C) the space crossing
appearance of the complex crystals can be seen. The retrodots of this Type 2 range from approximately 100 to more
than 500 )im. All x-ray spectra of the crystals exhibit pronounced Ca and P peaks and minor C, O, and Na peaks. At
high concentrations Ca exhibits two Ka peaks at 3.69 and
4.04 keV (A: 88260.C-GL/P; B: C-GL/P; C: 90198,
C-GL/P).
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933177/ on 06/15/2017
Ultrastructure and Elemental Composition of Retrodots
203
fixed in a cacodylate-buffered fixative (Rotterdam and
donor lenses). Second, the calcium phosphate crystals
of Types 1 and 2 retrodots have completely different
appearances. Finally, within one lens two or three
types occur simultaneously even in one or two adjacent retrodots (Fig. 9).
The nature of the content of Type 1 retrodots is
the most complex of all. The pronounced P peak and
the enhanced Ca signal indicate that a calcium salt of
phosphorous is certainly one of the constituents. However, the relative low Ca peak as compared to the P
peak is puzzling because the calcium phosphate of
Type 2 crystals have about equally intense peaks for
both elements. When taking into account the elevated
S peak and the presence of crisscrossing elements in
FIGURE 6. Ultrastructure of a very large (more than 1 mm)
Type 2 retrodot from which the upper lining has been fractured away during the SEM procedure. The rough surfaced
lining {arrowheads, B) is occupied by complex crystals (size
approximately 15 to 20 nm). SL = suture line. (A: 91089,PCL; B: 90218, C-GL/P).
ie, high Ca, low S, and no P peaks. Because of the use
of a windowless ECON detector we additionally found
C and O peaks. However, this elemental composition
can be ascribed to a number of other insoluble calcium
salts as, for example, calcium carbonate, calcium formate, calcium hydroxide, etc. The microchemical analysis carried out by Harding et al3 and the laser microprobe analytical observations of Pau and Kaufmann9
of birefringent spheroliths strongly favor the calcium
oxalate nature of this material. Calcium oxalate crystals give similar EDX spectra. The calcium oxalate type
of retrodots is rather infrequent and only few were
observed in the equators of two lenses.
The type 2 retrodots clearly contain calcium salts
of phosphorous. Because a quantitative interpretation
of the EDX signals is very complicated, especially for
the low energy oxygen signal, it cannot be determined
which form of calcium phosphate is present. Crystals
of different calcium phosphate salts yield identical
EDX spectra. Because of the poor solubility of the
dibasic Ca-HPO4, this is the most likely candidate. A
critical technical point to be raised here is whether the
crystals are artifacts caused by fixation with a phosphate-buffered solution. If this were the case the calcium would have to have been present as free calcium
in high concentrations or as a salt, which would be
changed to calcium phosphate, eg, calcium oxalate.
Our observations do not favor this view. First, identical calcium phosphate crystals were observed in lenses
FIGURE 7. Possible origin of Type 2 retrodots. In regions
close to Type 2 retrodots small (approximately 10 //m) circular elements are observed (asterisks, A), which seem to be
formed by coalescence of membranes of a group of stacked
fibers (B and C) and are covered by disintegrating membranes (asterisks). In a further stage the disintegrating membranes seem to form a cavity in which the typical crystals are
formed (arrowheads, D and E). The x-ray spectra of these
early Type 2 retrodots show in addition Lo the pronounced
Ca (double peak due to high concentration), P, Na, C, and O
Peaks a significant S peak. (A: 88260,C-GL/P; B: 88260,
C-GL/P; C: 90198, C-GL/P; D: 9O218,C-GL/P; E:
90198,C-GL/P)
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933177/ on 06/15/2017
204
Investigative Ophthalmology & Visual Science, January 1994, Vol. 35, No. 1
gests that calcium phosphate is the most prevalent
form of retrodot. The presence of a calcium oxalate
retrodot is more inferential. It seems likely that an
elevated calcium level within the fiber is the initiating
factor for the pathological formation of retrodots and
that the excessive calcium is eliminated either in the
form of calcium phosphate or calcium oxalate, which
are subsequently sequestered in membrane-bound
bodies: the retrodots.
The crucial role of calcium in cataract is now generally accepted17 and the presence of calcium phosphate and calcium oxalate in Morgagnian and hypermature cataracts has long been noted.18 Experimentally it has been shown that calcium induces
aggregation of crystallins.19 The observation of ShunShin et al,2 that the presence of retrodots in the human
lens is significantly associated with increased nuclear
light scatter, is in line with this. Recently Duncan et
al20 showed that the free calcium content of clear human lenses significantly increases with age and could
account for the enhanced optical density and light
scatter. In addition, there is ample evidence that altered calcium levels stimulate the activity of proteolytic enzymes in the lens21-22 and that both lens crystalFIGURE 8. Type 3 retrodots are characterized by their round
lins and cytoskeletal proteins are substrates for the
shape, their lining by radially running elements {arrowheads, calcium-activated proteases. In conjunction with the
A) and a pillarlike content (asterisks, A, B, and C). Their size
notion that crystallins, cytoskeletal proteins and the
varies between 100 and 300 /xm. The x-ray spectra of Type 3
lipid bilayer are firmly associated23 it is not surprising
retrodots content exhibit peaks only for Ca, C, and Q. (A:
that low (< 0.5 raM) and high (> 1.3 mM) levels of
90198.C-GL/P; B: 88260, C-GL/P; C: 88260,C-GL/P).
calcium alter the membrane structure and can even
lead to complete disruption.24 It has further been
shown25 that the eye lens exhibits phospholipase activthe cavity it can be argued that the more pronounced
ity and that the hydrolysis of phospholipids is a comP peak is also due to the presence of numerous memplex process regulated by protein kinase C and stimubranes enclosing residual protein. An elevated P peak
lated
by calcium.26 In view of this evidence a tentative
caused by membranous elements is also found by
45
hypothesis on the origin of the calcium phosphate deHarding et al. ' However, they did not observe an
increased Ca peak in their inclusion bodies. The elevated Na peak especially in the type 2 retrodot is puzzling as well. Because we did not take care to fix this
ion and because it is completely absent in pure Type 3
and Type 2 crystals it is not likely to be a residue of the
rinsing solution (sodium cacodylate). Most likely it represents sodium bound to protein. Its presence in normal regions supports this view.
Regarding the origin of the Type 3 retrodots,
which are presumptively calcium oxalate, Bron and
Brown1 summarized the available biochemical literature and concluded that ascorbic acid, present in large
amounts in the normal human lens, would be a strong
candidate for and an initiating factor in the formation
of such retrodots. As a result of continuous challenge
FIGURE 9. Incidentally one retrodot or two closely adjacent
to the lens by reactive oxygen species and an impaired
retrodots were observed, which reveal the ulirastructural
protection against oxidative damage in older age, oxaand x-ray characteristics of Type 2 (black and white dots) and
late would be formed in abundance, and would be
Type 3 retrodots (asterisks). Note that the membranes of
eliminated and sequestered within calcium-oxalate—
surrounding fibers have a normal ultrastructural appearance. (901892, P-GL.)
containing retrodots. The current study strongly sug-
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933177/ on 06/15/2017
Ultrastructure and Elemental Composition of Retrodots
posits can be formulated: Elevated free calcium levels
locally stimulate the activity of proteolytic enzymes
and of phospholipases leading to membrane disintegration and the liberation of crystallin-bound calcium.
Phosphoric acid is set free during the hydrolysis of the
phospholipids and is immediately trapped by the calcium. As a result of this trapping the calcium content
eventually will drop to more physiological levels. As a
consequence of this the stimulating effect on proteolysis, hydrolysis of phospholipids and aggregation of
crystallins will be diminished and the cataractogenic
process will be slowed down or stopped. This explains
why despite the presence of numerous retrodots the
retrodot lens is not opaque. As outlined earlier, retrodots are well-demarcated, isolated features, restricted
to the perinuclear region but absent in the nucleus. It
has been shown,27'28 that the phospholipid content of
lens membranes dramatically decreases from outside
to inside and that deep cortical and nuclear membranes have a cholesterol to phospholipid ratio of 5 or
more and are impermeable to ions. Moreover, it has
been found25 that the phospholipase activity in the
lens shows a decrease in activity from cortex to nucleus. Furthermore, freeze fracture structure studies11'28 allow the conclusion that deep cortical and nuclear fibers have an aberrant perimembranous protein
organization. This would explain the absence of retrodots in the lens nucleus. High calcium levels can lead
to the uncoupling of cells ultrastructurally indicated
by the presence of square arrays and it has been
shown10'29 that isolated early opacities in the human
lens are surrounded by membranes densely populated
with these square arrays. Because of this it is tempting
to postulate that the elevated calcium levels locally
lead to uncoupling of parts of fibers by formation of
square arrays and that therefore the retrodots remain
restricted to relative small parts of lens fibers. It
should be noted that spheroliths are said to occur in
the lens nucleus in Morgagnian cataract.14 It may be
that retrodots can be formed in the nucleus, but less
frequently than in the deep cortex, or that the definition of "nucleus" in some reports, also includes perinuclear cortex.
Finally, two puzzling questions remain to be answered. Why are there two types of retrodots containing calcium phosphate and why is the calcium trapped
by phosphate in two types and by oxalate in the other?
It has been shown that free calcium is not evenly distributed throughout the normal animal lens17'30 but
that it is relatively low in the most superficial regions,
increases toward the deeper cortex, and is low in the
nucleus. If this were the case for the human lens as
well, it would mean that most superficially the calcium
level would not be as such that it inevitably leads to a
breakdown of proteins and phospholipids and that oxalate would be able to trap the only slightly increased
205
calcium. Whether Type 1 retrodots are just early
stages of Type 2 retrodots, still containing remnants of
lens fibers, remains to be investigated.
Key Words
human lens, retrodots, ultrastructure, elemental composition, calcium
Acknowledgments
The authors thank Dr. E. Pels and her coworkers for collecting the donor lenses and Mrs. H. Fopma-Bonnes for secretarial assistance. They also thank Mrs. M. Danzmann, Mr. N.
Bakker, and Mr. T. Put for photographic assistance. The
EDX equipment was made available through a donation by
the "Stichting Blinden-Penning" at Amsterdam, The Netherlands.
References
1. Bron AJ, Brown NAP. Perinuclear lens retrodots: a
role for ascorbate in cataractogenesis. BrJ Ophlhalmol.
1986;71:86-95.
2. Shun-Shin GA, Bron AJ, Brown NP, Sparrow JM. The
relationship between central nuclear scatter and perinuclear retrodots in the human crystalline lens. Eye.
1992;6:407-410.
3. Harding CV, Chylack LT Jr, Susan SR, Lo W-K, Bobrowski WF. Calcium-containing opacities in the human lens. Invest Ophthalmol Vis Sci. 1983; 24:11941202.
4. Harding CV, Susan SR, Worgul B, Solway A, Maser
M. Energy dispersive X-ray analysis of human lens cataracts. Ophthalmic Res. 1979; 11:159-163.
5. Harding CV, Chylack LT Jr, Susan SR, Lo W-K, Bobrowski WF. Elemental and ultrastructural analysis of
specific human lens opacities. Invest Ophthalmol Vis
Sci. 1982;23:1-13.
6. Harding CV Jr, Susan SR, Lo W-K, Bobrowski WF.
The structure of the human lens. In: Maisel H, ed. The
Ocular Lens. New York: Marcel Dekker Inc;
1985:367-404.
7. Vogt A. Klinischer und anatomischer Beitrag zur
Kenntnis der Cataracta senilis insbesondere zur Frage
des subcapsularen Beginnes derselben. Graefes Arch
Klin Exp Ophthalmol. 1914; 88:329-391.
8. Vogt A. Weitere Ergebnisse der Spaltlampen mikroskopie des vorderen Bulbusabschnittes IV. Praesenile
und senile Linsentriibungen. Graefes Arch Klin Exp
Ophthalmol. 1922; 108:192-218.
9. Pau H, Kaufmann R. Laser microprobe analysis
(LAMMA) of spheroliths in human cataract lenses.
Arch Ophthalmol. 1983; 101:1935-1937.
10. Vrensen G, Willekens B. Biomicroscopy and scanning
electron microscopy of early opacities in the aging human lens. Invest Ophthalmol Vis Sci. 1990; 31:15821591.
11. Vrensen G, Van Marie J, Van Veen H, Willekens B.
Membrane architecture as a function of lens fibre maturation: a freeze fracture and scanning electron microscopic study in the human lens. Exp Eye Res.
1992;54:433-446.
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933177/ on 06/15/2017
206
Investigative Ophthalmology & Visual Science, January 1994, Vol. 35, No. 1
12. Bron AJ, Brown NAP. Lens structure and forms of
cataract. In: Duncan G, ed. The Lens: Transparency and
Cataract. Top Aging Res Europe. 1986; 6:3-11.
13. Zimmerman LE, Johnson FB. Calcium oxalate crystals
within ocular tissues. Arch Ophthalmol. 1958; 60:372383.
14. Bron AJ, Habgood JO. Morgagnian cataract. Trans
Ophthalmol Soc UK. 1976; 96:265-277.
15. Flocks N, Li twin CS, Zimmerman LE. Phacolytic glaucoma. Arch Ophthalmol. 1955; 54:37-45.
16. Goldberg MF. Cytological diagnosis of phacolytic
glaucoma utilizing millipore filtration of aqueous. BrJ
Ophthalmol. 1967; 51:847-853.
17. Duncan G, Jacob T. Calcium and the physiology of
cataract. In: Human Cataract Formation. Ciba Found
Symp 106. London: Pitman; 1984:132-152.
18. Van Heyningen R. The human lens I. A comparison of
cataracts extracted in Oxford (England) and Shikarpur (W. Pakistan). Exp Eye Res. 1972; 13:136-147.
19. Delaye M, Danford-Kaplan ME, Clark JI, Krop B, Gulik-Kryzwicki, Tardieu A. Effect of calcium on the calf
lens cytoplasm. Exp Eye Res. 1987;44:601-616.
20. Duncan G, Hightower KR, Gandolfi SA, Tomlinson J,
Maraini G. Human lens membrane cation permeability increase with age. Invest Ophthalmol Vis Sci.
1989;30:1855-1859.
21. David LL, Shearer TR. Role of proteolysis in lenses: a
review. Lens Eye Tox Res. 1989; 6:725-747.
22. Marcantonio JM, Duncan G, Rink H. Calcium-induced opacification and loss of protein in the organcultured bovine lens. Exp Eye Res. 1986;42:617-630.
23. Bloemendal H, Berbers GAM, De Jong W, et al. Interaction of crystallins with the cytoskeletal-plasma membrane complex of the bovine lens. In: Nugent Y, Whelan Y, eds. Human Cataract Formation. Ciba Found
Symp 106. London: Pitman; 1984:177-190.
24. Clark JI, Mengel L, Bagg A, Benedek GB. Cortical
opacity, calcium concentration and fiber membrane
structure in the calf lens. Exp Eye Res. 1980; 31:399410.
25. Broekhuyse RM. Biochemistry of membranes. In:
Duncan G, ed. Mechanisms of Cataract Formation in the
Human Lens. London: Academic Press; 1981:151191.
26. Exton JH. Mechanisms of action of calcium-mobilizing agonists: some variations on a young theme. FASEBJ. 1988; 2:2670-2676.
27. Li L-K, So L, Spector A. Membrane cholesterol and
phospholipid in consecutive concentric sections of
human lenses. J Lipid Res. 1985; 26:600-609.
28. Van Marie J, Vrensen G, Van Veen H. Maturing human eye lens fibre membranes and filipin cytochemistry. In: Vrensen GFJM, Clauwaert J, eds. Eye Lens
Membranes and Aging. Top Aging Res Europe
1991;15:123-134.
29. Vrensen G, Van Marie J, Willekens B, Van Veen H.
Square arrays in early cortical opacities. Invest Ophthalmol Vis Sci. 1990;30:2476-2481.
30. Harding CV, Unakar NJ, Bagchi M, et al. Elemental
and structural studies of the rat galactose cataract.
Lens Eye Toxic Res. 1989; 6:477-501.
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933177/ on 06/15/2017