Gene Expression and Genetic Engineering in the Lens


Dec 10, 2012 (5 years and 7 months ago)


Gene Expression and Genetic Engineering in the Lens
Friedenwald Lecture
Joram Piatigorsky
I began to study the transparent eye lens of the
chicken as a postdoctoral fellow in the laboratory of
Dr. Alfred J. Coulombre at the National Institutes of
Health in 1967, because of the many special features
making this tissue advantageous for investigating cel-
lular differentiation in eukaryotic cells.
The verte-
brate lens is an avascular, non-innervated epithelial
tissue physically surrounded by a capsule (Fig. 1). Lens
cell differentiation involves the cessation of cell divi-
sion, marked cell elongation during fiber cell formation,
and loss of cellular organelles, including the cell nucleus
(reviewed elsewhere
). Of particular interest to me was
that a group of specialized proteins (crystallins) could
be recognized as markers for specific gene expression
during lens differentiation. Little did I realize that I
was embarking on a winding journey that would trace
poorly defined lens crystallins through to the regulatory
elements controlling the expression of their genes, and
that traveling this path would consider, simultaneously,
evolutionary and developmental time scales. Jonas
Friedenwald predicted such a course for scientific re-
search already in 1922 at the beginning of his career,
when he wrote:
The road of science is a tortuous one, that twists
and turns and not infrequently crosses some of
the most ancient footpaths.
When I started my postdoctoral work, it was known
that crystallins appear already when the presumptive
lens cells are induced by the optic vesicle, and that the
crystallins accumulate during development until they
constitute the major soluble proteins of the lens.
Moreover, it was becoming evident that synthesis of
different crystallins was temporally and spatially reg-
ulated in the developing lens.
However, it was also
evident that, despite the potential usefulness of the lens
From the Laboratory of Molecular and Developmental Biology,
National Eye Institute, National Institutes of Health, Bethesda,
Submitted for publication: July 10, 1986.
Reprint requests: Joram Piatigorsky, PhD, Building 6, Room 201,
National Eye Institute, National Institutes of Health, Bethesda, MD
for studying the regulation of gene expression during
differentiation, we still knew too little about the crys-
tallins to explore the molecular basis for their synthesis.
Studies were revealing a growing number of crystallins
containing numerous polypeptides,
and iden-
tification was often based on immunological criteria
alone, without distinction between primary gene prod-
ucts and their post-translationally modified forms (re-
viewed elsewhere
). Thus, the crystallins had to be
defined at the level of their mRNAs and genes before
it would be possible to investigate their regulated
expression during development.
Here, I will summarize how combined studies using
the techniques of protein sequencing, X-ray crystal-
lography, and recombinant DNA have advanced our
understanding of the crystallins. I will then present the
progress that has occurred on our understanding of
crystallin gene expression during development, and
show how these studies have initiated, astonishingly
quickly, genetic engineering in the visual system.
Crystallin Heterogeneity
It was known as early as 1894 that the lens contains
high concentrations of heterogeneous, structural pro-
Numerous studies since then have established
that there are four immunologically distinct, major
classes of crystallins (a, j8, 7, and 8). Reviews covering
the a-, |8- and 7-crystallins
and 5-crystallin
been written. There are also a small number of so-
called minor crystallins in vertebrates, such as e-
and r-crystallin,
among others.
The crystallins are not distributed uniformly
throughout the vertebrates (Fig. 2). All vertebrate lenses
contain a- and /3-crystallins. 5-Crystallin, however, is
present only in lenses of birds and reptiles. 7-Crystallin
is found in fish, amphibians and mammals, but is ab-
sent from birds and reptiles. The physiological signif-
icance of having 5-crystallin substitute for 7-crystallin
in birds and reptiles is not known. It may be related
to its greater hydration or a-helical content, which may
aid accomodation in these soft lenses (reviewed
). There are other differences between 7-
and 5-crystallin, as described in the reviews cited above,
Vol. 28
Fig. 1. A sagittal paraffin section of a 5-day-old embryonic chicken
lens. The cornea is present at the top of the photograph. The cuboidal
epithelial cells overlie the elongated fiber cells. The cell nuclei will
disintegrate in the central region of fiber mass as the lens matures.
This photograph was made from a slide given to me by Dr. A. J.
Coulombrein 1968.
which may also satisfy specific physiological needs of
the different species.
Each crystallin class has characteristic native mo-
lecular weights, and, thus, can be separated by gel fil-
The approximate molecular weights are
800,000 for a-crystallin, 100-200,000 for ^-crystallin,
20,000 for 7-crystallin, and 200,000 for 8-crystallin.
Subsequent reduction and electrophoresis of the pu-
rified crystallins in a denaturing polyacrylamide gel
Fig. 3. Electrophoresis in SDS-urea-polyacrylamide gels of primary
polypeptides derived from purified crystallins. The a, 0, and y crys-
tallins were isolated from 5- to 10-day-old mice and were fractionated
by column chromatography using Sephadex G-200 (taken from C.
R. King, Doctoral Dissertation, Johns Hopkins University, 1983,
with permission of the author). S-Crystallin was purified by isoelectric
focusing. The murine crystallins were analyzed on a 15% and S-crys-
tallin on a 10% polyacrylamide gel containing 0.1% sodium dodecyl
sulfate and 8 M urea; the gels were stained with Commasie blue R.
Some 0-crystallins are contaminating the 7-crystallin fraction.
No. of
all vertebrates
all vertebrates
fish, amphibians,
birds, reptiles
Drosophila small
heat shock proteins
M.xanthus spore
coat protein S,
membrane protein?
appear unlinked
linked; one member
may be unlinked
Fig. 2. Brief accounting of the major crystallin polypeptides of
vertebrates. The possible ancestral relationships were taken from In-
golia and Craig"
for the a-crystallins, Wistow and coworkers"
Xanihus spore cost protein S) and Crabbe'" (c-myc) for the /Jy-crys-
tallins, and Williams and coworkers
(r-crystallin) for 5-crystallin.
I have suggested that o-crystallin may be ancestrally related to a
membrane protein because of its unusually high leucine and isolencine
content (see Piatigorsky"). References concerning organization of
the crystallin genes can be found elsewhere.
Recent reports on
mapping human crystallin genes have indicated that the aA gene is
on chromosome 21 and the aB gene on chromosome 16,
the /SA3/
Al gene is on chromosome 17,
and the 0B2 (0Bp) gene on
chromosome 22 (Gorin, personal communication), and the 7-crys-
tallin gene cluster is on chromosome 2.
(Fig. 3) demonstrate that each crystallin class is com-
posed of several polypeptides with molecular weights
ranging from 20,000 (a and 7) to 50,000 (6). The 7-
crystallins can be further fractionated into a series of
extremely similar polypeptides by isoelectric focus-
Thus, a major cause for the native molecular
weight differences among crystallins is the aggregation
state of their polypeptides.
Extensive post-translational modifications and
changes occurring with age further complicate the
identification of the different primary crystallin poly-
However, the heterogeneity shown in
Figure 3 can be achieved by in vitro translation of lens
This suggests that there is a separate gene
for each crystallin polypeptide. As we shall see below,
while this conclusion is essentially correct, there are
some surprises.
Crystallin Genes
Chicken 5-crystallin was the first crystallin cDNA to
be cloned.
This consequently led to our first glimpse
of a crystallin gene.
cDNA:DNA hybridization
No. 1 LENS GENE EXPRESSION / Piorigorsky
and Southern blot hybridizations
that there are probably only two chicken 5-crystallin
genes. Examination of cloned fragments of these genes
in the electron microscope revealed that they are very
Our first photomicrograph of a hetero-
duplex showed a minimum of seven introns in one of
the cloned 5-crystallin gene fragments (Fig. 4). The
complicated structure of the 5-crystallin gene raised
immediate questions concerning the nature of the in-
trons and whether all crystallin genes have numerous
We subsequently cloned and characterized cDNAs
for aA,
(now called 0A3/A1),
and four y-
of the murine lens in order to examine
other crystallin genes. In the last few years, other cloned
crystallin cDNAs have been constructed.
Many of
these have been indispensable for characterizing crys-
tallin genes in different species.
A schematic ex-
ample of gene structure from each crystallin family is
illustrated in Figure 5. Inspection of Figure 5 shows
that the representative member of each crystallin class
has a very different gene structure, with the 5-crystallin
gene being much more complex than the a-, &-, or y-
crystallin gene. The references cited above show that
all members of a crystallin class have similar structures
which have been conserved throughout evolution (see
for brief review).
Fig. 4. Electron micrograph showing a heteroduplex generated by
hybridizing the cloned chicken 5-crystallin genomic fragment gSCr-
1 with 5-crystallin mRNA.
The arrows denote the introns which
are seen as loops between the hybridized exons. The small stretch of
5' 5-crystallin mRNA contains sequences derived from exons 1 and
2 which did not hybridize to the cloned gene fragment; the non-
hybridized 3' half of 5-crystallin mRNA is seen extending from the
cloned DNA, since their sequences are not present in gSCr-1. The
introns have been labeled C through J to fit our present knowledge
of the 51 gene structure." The long stretch of double-stranded DNA
contiguous with the 5' region of the 61 -crystallin gene fragment belongs
to a second DNA fragment, not containing S-crystallin sequences,
that was cloned in the same Charon X 4A recombinant bacterio-
This is the first electron photomicrograph taken of a crystallin
gene in the project subsequently reported by Bhat and coworkers.'"
1 2 3 45 6
1 2
3 4 5 6 7 8 9 1011 12131415 16 17
1 1 I I I I I I • I I I I I • chicken.
Fig. 5. Schematic representatives of one member from each major
family of crystallin genes. Numbers refer to exons. The open boxes
are non-coding sequences, the solid boxes are protein-coding se-
quences, and the lines are introns. Breaks in the intron mean the
sequencing is incomplete in that region. The following sources have
been used: «A (Thompson, Hawkins and Piatigorsky, in preparation),
£B1 (Thompson, Hawkins and Piatigorsky, in preparation), y4 (Lok
and coworkers),
and 51 (Nickerson and coworkers).
Relationship Between Protein and Gene
Structures For the 0- and 7-Crystallins
Investigations from BlundelPs, Bloemendal's, and
my laboratory have converged to link gene structure
with protein structure in the fi- and 7-crystallin fami-
lies. X-ray crystallographic studies from BlundelFs
laboratory showed that 7ll-crystallin from the calf lens
contains two domains that are joined by a connecting
peptide, and that each domain is composed of two very
similar structural motifs.
The structural motifs were
called "Greek keys," due to their resemblance to the
decorations on Greecian urns. The 7-crystallin poly-
peptide showed the highest internal symmetry of any
protein examined previously by X-ray crystallography.
Another important finding was from Bloemendal's
laboratory. They demonstrated an unexpected se-
quence similarity between calf 7II and j8Bp (0B2) from
the bovine lens.
Moreover, both the 7II and 0Bp
sequence have an internal repeat corresponding to the
two domains of the tertiary structure of 711, and each
repeat is itself composed of an internal repeat sequence
corresponding to the two structural motifs of each do-
main. Subsequent studies involving protein
sequencing have demonstrated that
the similarities between and internal duplications
within the primary structures of the /3- and 7-crystallin
polypeptides extend to the other members of these two
families of proteins. It is interesting to note that the
greatest variation in sequence among the different 7-
crystallins is found in the third motif.
These re-
sults indicated that the /?- and 7-crystallins are evolu-
tionarily related and form a /?7-superfamily of proteins.
Further discussion of the structure of the jS7-crystallins
can be found elsewhere.
Motif 4
Motif 1
Motif 3
Motif 1
Motif 3
7-crystallin polypeptides.
The paths of the polypeptide
chains are traced by joining
Motif 4 the positions of consecutive
t i f 2
s\ Motif 2 »,_
, Qa atoms. The structure of
•yll is taken from the coor-
dinates determined by X-ray
crystallography at 2.6 A res-
The molecule is
viewed from a direction per-
pendicular to the pseudo-
dyad which relates the N- and
C-terminal domains, em-
phasizing the intra-molecular
symmetry. The model of 0Bp
) is based on the same
coordinate set, modified by
Calf yll Calf 0 Bp interactive computer graph-
The model has been
rotated slightly, relative to the view of 7II, in order to show the N- and C-terminal "arms." This conformation is that predicted for the "extended"
model for j8Bp dimerization.
1 am grateful to Dr. G. J. Wistow for help in preparing this figure.
Since the three-dimensional structure of 7ll-crys-
tallin was known and a sequence similarity between
7II and jSBp had been demonstrated, it was possible
to predict the structure of the /3Bp polypeptide by an
interactive computer graphics program.
As shown in
Figure 6, the predicted structure of /3Bp is extremely
similar to that of 7II, except that #Bp has both an N-
and C-terminal peptide extending from the two domain
core of the protein. With exception of the j8s polypep-
tide, which could as well be considered a member of
the 7-crystallins,
all j8-crystallin polypeptides appear
to have arms extending from their N- and (sometimes)
It has been suggested that the terminal
extensions of the |8-crystallin polypeptides play an im-
portant role in the protein-protein interactions of the
The N-terminal arms of the 0-crys-
tallins are evolving faster and differ much more from
each other than the globular domains of these pro-
t d n s
Finally, we showed that the 0A3/A1 (formerly £23)
polypeptide from the murine lens contains introns
separating exons which encode the predicted structural
motifs of the protein.
This is diagrammed in Figure
7. At that time, we thought that the N-terminal arm
was coded for by the same exon as that which encodes
the first structural motif. We have shown recently,
however, that there are two additional exons which
encode this extension.
The /3B1 gene in the rat
and the /3A3/A1 gene in the human
similarly have
separate exons for each structural motif. The N-ter-
minal extension of the rat
and chicken (Fig. 5) j8Bl
gene is encoded entirely on the second exon, while the
C-terminal extension is encoded on the last exon along
with the fourth structural motif. Analysis of 7-crystallin
genes from the rat,
and human
that, in contrast to the j8-crystallin genes, each domain
of the protein (rather than each structural motif) is
coded for by an exon, although an intron still separates
the exons encoding the protein domains (Fig. 7). In
Figure 7, the regions coding for motifs 1 and 3 are
given similar markings, as are the regions coding for
motifs 2 and 4, in both the 0- and 7-crystallin gene,
Mouse 0A3/A1
Mouse y4
Pr o t e i n
Pr o t e i n
Fig. 7. Relationship between the structures of the /8A3/A1 and 74
genes from the mouse and their encoded proteins. The /3A3/A1 gene
structure has been constructed from the studies of Inana and
and Peterson and Piatigorsky;
74 gene structure is from
Lok and coworkers.
The general idea that exons encode structural
motifs of proteins was first proposed by Gilbert
and by Blake.
See text and Figure 5 for further explanations. The solid boxes encode
N-terminal amino acids extending from the first structural motif of
the protein. The exons (j8A3/Al) or exon regions (74) which are
believed to be related by duplication are indicated by dots or hatches.
No. 1 LENS GENE EXPRESSION / Piorigorsky
since it appears that these form structurally related
and arose by a process of intragenic dupli-
cation in this 0y superfamily of proteins.
A Functional Role For Introns
The heterogeneity of primary polypeptides within
the jSY-crystallins is due largely to several gene dupli-
cations followed by separate evolutionary changes
within the individual genes. The two primary a-crys-
tallin polypeptides, «A
and aB
, also appear to have
arisen by gene duplication.
However, in the rodent
families Muridae (rat, mouse) and Cricetidae (hamster,
gerbil), there is a third a-crystallin polypeptide, aA
which has a 22 (rat) or 23 (mouse, hamster) amino
acid sequence inserted between residues 63 and 64 of
the aA
-crystallin polypeptide.
Interestingly, intron
1 of the a A gene separates codons 63 and 64 in all
species examined, including mice,
chicken (Thompson, Hawkins, and Piatigorsky, in
preparation), and humans.
The aA
insert peptide of
and hamsters
is neatly encoded within this
intron, as illustrated by the open box in Figure 8. The
RNA derived from this small (69 base pairs) insert exon
is spliced into mature mRNA 10-20% of the time the
gene is transcribed; this splicing reaction is not regu-
lated during development.
By contrast, the aA
mRNA is generated by splicing together the RNA se-
quences derived from exons 1 and 2 80-90% of the
time the gene is transcribed, eliminating the sequences
derived from the insert exon. We do not know yet the
molecular basis for the infrequent inclusion of RNA
from the insert exon. One possibility is that its 3'
splice junction is GC rather than the more common
GT (Fig. 8).
The alternative splicing of the aA-crystallin gene in
rodents may produce a specialized polypeptide tailored
to the needs of the lens in these species; another pos-
sibility is that the aA-crystallin splicing variant is an
example of evolution testing a modification of this gene
without altering the original gene or its product. In any
event, alternative RNA splicing of the aA-crystallin
gene shows how an intron can have a functional, as
well as a structural role.
Another example of function embedded within an
intron is the presence of an enhancer in immunoglob-
ulin genes.
Brief reviews on enhancers (regulators
of gene activity) can be found elsewhere.
So far,
no sequences controlling gene expression have been
uncovered within introns of crystallin genes.
It is not possible to relate the gene structure for aA-
crystallin to the structure of its encoded protein, as has
been done for the /fy-crystallins, since the three-di-
mensional structure of aA-crystallin is not known.
5' GT GC
Fig. 8. Alternative splicing of the murine aA-crystallin gene. The
ak gene has been taken from King and Piatigorsky.
Exon 3 was
added by analogy with the chicken (Fig. 5) and hamster
genes, and knowledge of the murine aA-crystallin cDNA.
The actual
3' region of the murine aA-crystallin gene has not been isolated. The
exons were differentially marked only for clarity in visualization of
the mRNAs. Note the different 3' splice junction (GC) for the insert
Nonetheless, interesting speculations have been made,
and a two domain structure for the aA-crystallin poly-
peptide has been suggested (see Wistow
for further
discussion and references).
Surprising Features of 6-CrystaIlin
Gene Expression
As shown in Figure 3, there are two chicken 5-crys-
tallin polypeptides (50 K and 48 K).
Only two tryptic
peptide differences have been noted between these 5
Of particular interest is the fact that
the ratio of synthesis of the 48 K to the 50 K polypep-
tides is strongly affected by the intracellular concen-
tration of ions (especially Na
, K
and Cl~) in cultured
embryonic lenses.
Since both 5 polypeptides are
synthesized in a reticulocyte or a wheat germ lysate
supplemented with 5-crystallin mRNA,
or in cul-
tured monkey kidney cells injected with 5-crystallin
mRNA (see Piatigorsky
), it appeared as if each 5
polypeptide had its own mRNA. This idea was sup-
ported by the observations that the 51 and 52 promoters
both function in a Hela cell extract,
and by transient
expression experiments using cultured embryonic
chicken lens epithelia transfected with the pSVO-CAT
expression vector containing the 51 or 52 promoter.
In addition, the extreme similarity of the polypeptides
encoded in the 51 and 52 genes (91% identity) suggests
strongly that both genes are expressed and subjected
to constraints by evolution.
The simplest hypothesis
was that each of the two similar 5-crystallin genes (Fig.
9) encodes one of the two polypeptides.
Despite the similarity of the two 5-crystallin genes,
however, all the cDNAs ever reported were derived
from the 51 gene.
The EcoRl site in exon 13
of 51 (Fig. 9) is absent from the 52 gene,
which has
allowed cloned 5-crystallin cDNAs to be identified as
Vol. 28
12 3 4 5 6 7 89 1011 12131415 16 17
5' -H D (HHHHHHHl—H-H-fl—D "Kb spacer
4' 5' 6' 7'8' 9'10'11' 12'13'14'15'16' 17'
-W—O-M-O-O-O—^HHHH}—D- 3-
expression of the chicken 5-
crystallin gene locus. The
gene structures and organi-
zation have been pub-
l i shed.
4 5 7 3
" Exon 13 is rep-
resented as a solid box in the
51 gene to indicate that it
contains an EcoRl site lack-
ing in the corresponding exon
of the 52 gene, maki ng it pos-
sible to distinguish cDNAs
derived from each gene. The
| autoradiogram of an SDS-
polyacrylamide gel shows the
~ two polypeptides synthesized
in a reticulocyte lysate sup-
• plemented with 51 mRNA
~ and immunoprecipitated
with 5-crystallin antiserum.
51 mRNA was derived from a 51 cDNA subcloned into an SP6 vector.
We are investigating whether the 52 gene generates an mRNA.
51 products (Nickerson, unpublished). Surprisingly, as
shown in Figure 9, mRNA derived from a cloned 51
cDNA synthesizes both the 48 K and 50 K polypeptides
in a reticulocyte lysate.
As with authentic 5-crystallin
mRNA, the ratio of synthesis of the 48 K to 50 K
polypeptides is affected by ions. We do not yet know
the mechanism by which the 51 mRNA produces both
polypeptides, but our current experiments indicate that
it is not by use of two different AUG initiating codons
(Warwrousek, unpublished), as appears to occur with
the 0A3/A1 mRNA.
Future studies must also
determine whether the 52 gene produces a functional
mRNA in the lens. An interesting possibility is that
the 52-crystallin gene is expressed in non-lens cells,
since some tissues outside of the lens have been re-
ported to contain 5-crystallin RNA sequences.
The surprising nature of 5-crystallin synthesis illustrates
the variety of mechanisms used to create crystallin het-
erogeneity and emphasizes the importance of under-
standing the molecular genetics of each polypeptide.
Temporal and Spatial Regulation
of Crystallin Gene Expression
Figure 10 relates schematically the early develop-
ment of the chicken lens with the initial appearances
of a, j8, and 5-crystallins. 5-Crystallin is the first crys-
tallin present and can be detected already in the lens
placode, before the formation of a lens vesicle.
early appearance of 5-crystallin resulted in its being
called FISC (first important soluble crystallin);
the name was changed to 5-crystallin.
We have sug-
gested on the basis of extrapolated cDNA:RNA hy-
bridization data
that 5-crystallin mRNA begins to
accumulate by 42 hours of development, which is just
a few hours after the optic vesicle interacts with the
lens ectoderm. It is possible, however, that 5-crystallin
mRNA is present before this time. Immunofluores-
cence studies have shown that 0- and traces of a-crys-
tallin do not appear in the embryonic chicken lens until
approximately 60 hours of development.
It is inter-
esting that the order of appearance of different crys-
tallins differs among species (see Piatigorsky
for re-
view). The reason for the specific order of crystallin
synthesis in different species is not known.
The regulation of crystallin gene expression is not
limited to the early development of the lens. In chick-
ens, 5-crystallin synthesis predominates in the embry-
35 45 49 52 55 60
Fig. 10. Time of appearance of different crystallins during early
development of the chicken lens. —, not detectable; +/-, just de-
tectable. This figure was composed with consideration of the im-
munofluorescence data of Zwaan and Ikeda.
I am grateful to Dr.
James W. Hawkins for this figure.
No. 1
onic lens and 5-crystallin accumulates to approximately
70% of the total protein present.
After hatching,
5-crystallin synthesis is selectively reduced,
the /3-crystallins become the principal proteins in the
adult lens.
In vitro translations indicate that the
mRNA for 5-crystallin is the major mRNA in the em-
bryonic lens,
but that it is lost from the chicken
lens between 3 and 5 months after hatching.
The 0-
crystallin mRNAs accumulate in the post-hatched
chicken lens.
In addition to the temporal regulation of crystallin
gene expression, there is also a marked spatial regu-
lation of crystallin synthesis within the lens. It has been
known for some time that 7-crystallin is present only
in the lens fiber cells.
More recently it has been
demonstrated in embryonic rats that /3-crystallin, like
7-crystallin, is present only in the fibers.
immunofluorescence experiments show also that the
/?-crystallins are detected in the peripheral, elongating
fiber cells and the deeper cells of the fiber mass, while
the 7-crystallins are found only in the more elongated
posterior fiber cells.
Spatial regulation occurs within, as well as between,
the different families of crystallin genes. Quantitative
and qualitative differences in synthesis of a- and 0-
crystallin polypeptides in the calf lens have been re-
viewed elsewhere.
In chickens, we showed that the
mRNA for the 0B1 polypeptide (formerly 035) appears
as the cells begin to elongate into lens fibers, while the
other 0-crystallin mRNAs are present both in the
epithelial and fiber cells.
Figure 11 provides a
quantitative estimate for 5-crystallin and for different
0-crystallin mRNAs in different regions of the 15-day-
old embryonic chicken lens.
These data show graph-
ically that each mRNA displays a characteristic distri-
bution in the lens at this stage of development.
Thus, present evidence indicates that the crystallin
genes are differentially expressed both temporally and
spatially in the developing lens, and that this differential
expression results in the uneven distribution of different
crystallins within the lens. Although it is possible that
the relative amounts of crystallin mRNAs are con-
trolled post-transcriptionally at the level of RNA pro-
cessing or degradation (for example, see Bower and
), we assume that transcriptional regula-
tion contributes in a major way to the expression of
the crystallin genes in the eye lens. Since the levels of
the different mRNAs composing a family of crystallin
genes appear to be regulated independently, it is likely
that each gene contains its own regulatory elements.
This does not eliminate the possibility that there are
also regulatory elements controlling coordinately more
than one crystallin gene. We began, therefore, to in-
vestigate the molecular basis for transcription of crys-
tallin genes.
Central Epithelium
Equatorial J
Epithelium |
Fig. 11. Spatial distribution of crystallin mRNAs in the 15-day-
old embryonic chicken lens. The different crystallin mRNAs were
quantified by hybridization with cloned cDNAs.
1 am grateful to
Auran P. Piatigorsky for help in preparing this figure.
Strategy For Transcriptional Control Studies
We decided to focus initially on the murine aA-
crystallin gene, since we had cloned the 5' region of
this gene with its associated flanking sequences and
were ready to analyze it in greater detail.
There is
only a single copy of the aA-crystallin gene,
and it is
highly conserved throughout evolution.
strategy was to identify functionally the promoter of
the gene by using it to drive a foreign gene in an expres-
sion vector; promoters consist of numerous regulatory
elements within the 5' flanking sequence of genes.
We chose the pSVO-CAT expression vector as a test
described below. By this type of modular
approach, we could add whatever sequences we wished
from the aA-crystallin gene or its flanking regions to
the expression vector in order to identify putative con-
trol elements. In addition, this strategy facilitates later
comparisons of regulatory elements from different
crystallin genes by allowing us to determine their ac-
tivity using the same assay.
I •
Fig. 12. A, Schematic drawing of the murine aA-crystallin gene
and its 5' promoter. +1 defines the initiation site of transcription
(RNA synthesis). B, Diagram of the pSVO-CAT expression vector.
The indicated promoter fragments from the murine aA-crystallin
gene were inserted into the Hind III site of the vector, (a) indicates
that the promoter was inserted in the proper orientation and (b)
indicates that the promoter was inserted in the inverse orientation.
These constructions have been described by Chepelinsky and
except that the position numbers have been corrected
(Chepelinsky, Sommer, and Piatigorsky, submitted).
Figure 12A shows a diagrammatic representation of
the aA-crystallin gene and its promoter. Two different
lengths of the putative promoter were examined. In
both cases, the 3' cut was made at nucleotide position
+46. This includes 46 base pairs of 5' untranslated se-
quence from exon 1. For the longer piece, a cut was
made at nucleotide position —346 in the 5' flanking
region, and, for the shorter piece, a cut was made at
nucleotide position —88. Minus numbers indicate base
pairs upstream from the initiation site of transcription.
These two fragments were inserted separately into the
pSVO-CAT expression vector in each possible orien-
tation, as shown in Figure 12B. The pSVO-CAT vector
contains the bacterial chloramphenicol acetyl trans-
ferase (CAT) gene, as well as portions of the simian
virus 40 (SV40) genome. The latter donates a poly-
adenylation site and splicing signals to the CAT pri-
mary transcript. This construct makes is possible to
test for aA-crystallin promoter function by determining
CAT enzyme activity, which is a very sensitive assay.
Since animal cells do not contain a CAT gene, any
enzymatic activity observed must be generated by the
expression of the CAT gene in the vector.
Two methods were used for testing the function of
the aA-crystallin promoter, as diagrammed in Figure
13. These consisted of transfecting cultured lens
and creating transgenic mice.
Transfecting Cultured Lens Epithelia
From Chicken Embryos
Phillpott and Coulombre
discovered that the cells
in explanted lens epithelia from 6-day-old chicken em-
bryos elongate when cultured with fetal calf serum.
Subsequently, we showed that these elongating lens
cells synthesize crystallins and differentiate into fiber-
like cells.
Since numerous studies had shown
that crystallin synthesis arrests in cultured lens cells,
we examined the possibility that crystallin promoters
could function when introduced into primary explants
of the lens epithelia. Multiple copies of the super-coiled
expression vector containing the a-CAT fusion gene
were introduced into the cells of explanted 14-day-old
embryonic chicken lens epithelia.
The method for
obtaining the lens epithelial explants for the transfec-
tion experiments is diagrammed in Figure 14 A. Trans-
fections were performed by co-precipitating the plas-
mids with calcium phosphate
1 day after explan-
tation, and CAT assays were performed 3 days later.
Cells remaining in the original explant elongate as they
begin to differentiate into lens fibers during the culture
period, while the peripheral cells that migrated from
the explant remain as an epithelial monolayer (Fig.
14B-D). We do not know at present whether all the
cells of the explants take up and express the expression
vectors, or how many copies of the plasmids are in the
cells after transfection. Our finding that foreign genes
can be introduced into the cells of explanted lens ep-
ithelia has made this system a valuable tool for the
study of regulatory sequences of crystallin genes.
j PBR322 | Promote
Transfection Transgenic
Calcium phosphate
Cultured cells
nonintegrated DNA
Fig. 13. Strategy for crystallin gene expression studies using the
pSVO-CAT expression vector containing the aA-crystallin promoter
sequences described in Figure 12.
No. 1 LENS GENE EXPRESSION / Piorigorsky 17
Fig. 14. A, Explanation of
a lens epithelium. The lens is
placed on a collagen-coated
dish with the anterior capsule
facing down (a). The poste-
rior capsule is torn and
opened (b). The fiber mass
(F) is removed, leaving equa-
torial (E) and central (C) ep-
ithelial cells attached to the
lens capsule (c). The epithe-
lium (d) is cut with a scalpel
and fixed onto the dish with
forceps (e). B, Low-magnifi-
cation photomicrographs of
a 14-day-old embryonic
chicken lens explant after 3
days of culture. C, Piece of
explant after culture for 1
day. D, Explant and cellular
outgrowth after 4 days of
culture. Taken from Chepe-
linsky and coworkers.
Production of Transgenic Mice
We were fortunate to collaborate with Drs. Heiner
Westphal, Paul Overbeek, Jaspal Khillan, and Kathy
Mahon (Laboratory of Molecular Genetics, National
Institute of Child Health and Human Development,
NIH) for the transgenic mice experiments. Initially, in
these experiments the a-CAT fusion gene was removed
from the pBR322 sequences of the expression vector
and microinjected into one of the pronuclei of the fer-
tilized mouse egg.
The method is diagrammed in
Figure 15. The microinjected eggs were placed into a
pseudopregnant foster mother. The litter developing
from the microinjected eggs is called the Fo generation.
DNA was taken from a snip of tail and tested by hy-
bridization for the presence of the foreign gene (the a-
CAT fusion gene). Positive mice are mated to normal
mice to test for transmission of the a-CAT gene; their
progeny are the Fl generation. If 50% of the Fl progeny
carry the foreign gene, one can conclude that it is being
carried in the germ line of the Fo mouse. The transgenic
experiments contrast with the transfection experiments
in that the a-CAT fusion gene becomes integrated into
the chromosomes of the host cells. Thus, the conditions
Fig. 15. Production of transgenic mice. Stippled mice contain the
foreign gene in their germ line. This is indicated by the positive dot-
blot hybridization result (solid circle) using DNA from the tail. See
text for further explanation. I am grateful to Dr. Ana B. Chepelinsky
for preparation of this figure.
Vol. 28
— cm-Ac 3
— cm
Fig. 16. Thin-layer chromatogram showing CAT activity in lens
epithelia transfected with pa366a-CAT, but not with the other vectors.
SVO, pSVO-CAT; 366a, p«366a-CAT; 88a, p«88a-CAT; 366b,
p«366b-CAT; 88b, pa88b-CAT. cm, chloramphenicol; cm-Aci,
chloramphenicol 1-acetate; cm-Ac
, chloramphenicol 3-acetate.
Taken from Chepelinsky and coworkers,'
except that the position
number has been changed from 364 to 366 (Chepelinsky, Sommer,
and Piatigorsky, submitted).
for expression of foreign genes in transgenic mice
should mimic closely those for gene expression in the
normal in vivo environment.
Transient Expression of the
a-CAT Fusion Gene
The data shown in Figure 16 show that lens epithelia
transfected with the paA366a-CAT vector (aA pro-
moter inserted in proper orientation) expressed CAT
activity, indicating aA-crystallin promoter function.
By contrast, only a trace amount of CAT activity was
present when the epithelia were transfected with
paA366b-CAT (aA promoter inserted in opposite ori-
entation) or with pSVO-CAT (no promoter present).
Of particular interest was the fact that neither paA88a-
CAT nor paA88b-CAT promoted CAT activity in the
transfected lens epithelia. Thus, sequences between —88
and -366 appear essential for the proper functioning
of the murine aA-crystallin promoter in the explanted
embryonic chicken lens epithelia. Additional experi-
ments showed that paA366a-CAT did not function in
cultured embryonic chicken fibroblasts, indicating that
this promoter behaves in a tissue-specific fashion.
Our current experiments have shown that a fragment
consisting of nucleotide positions -111 to +46 are suf-
ficient to promote CAT activity in the transfected lens
In addition, we have identified a regulatory
element within this region that functions only in the
proper orientation or in both orientations, depending
on its surrounding sequences (Chepelinsky, Sommer,
and Piatigorsky, in preparation).
Transient expression experiments have also been
performed using the chicken aA-crystallin gene pro-
moter fused to the 5-crystallin gene.
These investi-
gators microinjected a hybrid a/5-crystallin gene into
the nuclei of murine lens epithelial cells in primary
culture and assayed immunologically for 5-crystallin
synthesis. The a/5 hybrid gene was expressed in the
murine lens cells, but not in primary cultures of fibro-
blasts or in L-cells, indicating that the chicken aA-
crystallin promoter, like that of the mouse, is tissue-
specific. Deletion experiments located 53 base pairs
between nucleotide positions -242 and -189 that were
critical for promoter function in the lens cells. These
sequences displayed enhancer-like properties by being
able to function in either orientation and, although
less efficiently, when placed approximately 1.7 kilo-
bases downstream in the second intron of the hybrid
Crystallin Promoters Can Function
in Foreign Species
Experiments from several laboratories demonstrat-
ing that crystallin genes and their promoters can func-
tion in lens cells of foreign species are summarized in
Figure 17. Kondoh and coworkers
showed by an
immunoperoxidase assay that the cloned 51-crystallin
gene from chickens functions well after microinjection
into the nucleus of cultured lens epithelial cells from
10-day-old mice. Either very low levels or no 5-crys-
tallin expression was observed when non-lens cells were
microinjected with the 51 gene. Further experiments
indicated that the region between positions —51 and
- 80 is critical for the lens-specific function of the 51
promoter in the murine lens cells.
It is important to
underscore that mice do not have 5-crystallin
Figure 17 also shows transfection experiments using
various crystallin promoters inserted into the pSVO-
CAT expression vector and explanted embryonic
chicken lens epithelia. First, Borras and coworkers
demonstrated that the chicken 51 (sequences —344 to
+23) and 52 (sequences -346 to +22) promoters func-
tion in the homologous lens epithelial cells.
as described above, Chepelinsky and coworkers
showed that the murine aA-crystallin promoter (se-
quences —366 to +46) is active in explanted chicken
lens epithelia. This result complements the microin-
jection experiments given above, showing that chicken
aA-crystallin promoter sequences can function in mu-
rine lens epithelial cells, and is not surprising, since
both chickens and mice have an aA-crystallin gene.
More unexpected are the findings by Lok and cowork-
They showed that the murine 72-crystallin pro-
moter (sequences -392 to +42) functions very well in
No. 1
Fig. 17. Summary of experiments
demonstrating that crystallin promoters
can function in lens cells of foreign species.
The experiments are discussed in the text.
In the experiment by Kondoh and co-
the black cell on the left shows
a positive immunoperoxidase reaction for
S-crystallin while the transparent cells on
the right show a negative reaction. The
numbers refer to the extent of 5' flanking
base pairs on the specified promoters in
the pSVO-CAT expression vector. COS
cells used in the mouse P72-CAT
were derived from monkey
cells and contain SV40 T-antigen.
chicken p6-CAT experiment is from Bor-
ras and coworkers,
and the mouse paA-
CAT experiment is from Chepelinsky and
Mouse lens epithelial cells
Chicken lens epithelium
<J1 gene
etal. (1983)
Chicken I Mouse \ Mouse
pd-CAT JpaA-CAT \py2-CAT
Borras Chepelinsky Lok
etal. (1985) etal. (1985) etal. (1985)
expianted embryonic chicken lens epithelia, but not in
non-lens cells. Deletion of the yl 5' flanking sequences
to nucleotide position — 171 resulted in loss of promoter
activity in the transfected lens epithelia (Fig. 17).
Moreover, recent experiments indicate that the 72-
crystallin promoter contains an enhancer-like element
between positions —226 and —31 and two domains
(-190 to -125 and -105 to +45) for optimal promoter
function in the chicken lens epithelia.
Again, it is
important to note that chickens appear to lack 7-crys-
tallin genes.
These data suggest that the basis for the
tissue-specific expression of crystallin genes may have
originated very early in the evolution of the lens, and
that the different classes of crystallin genes may be reg-
ulated by common mechanisms in all species.
Expression of the a-CAT Fusion Gene
in Transgenic Mice
Insertion of the a-CAT fusion gene into the germ
line of a transgenic mouse provides a more stringent
test for identifying a region of the aA-crystallin pro-
moter which functions in a tissue-specific manner. In
collaboration with the group of Dr. Heiner Westphal
(see above), we obtained two transgenic mice contain-
ing the «-CAT fusion gene in their germ line after mi-
croinjection into the pronuclei of 99 fertilized eggs.
Hybridization tests indicated that there was probably
no more than one a-CAT fusion gene per haploid ge-
nome integrated into the DNA of these mice. Most
exciting was the finding that bacterial CAT activity was
present in the eyes of these two transgenic mice. No
CAT activity was found in the tails of the mice. More-
over, 50% of the Fl progeny of one of the transgenic
mice (number 7378) inherited CAT activity in their
eyes, indicating Mendelian transmission of the a-CAT
Figure 18 illustrates the results of CAT assays per-
formed on homogenates of numerous organs of a 6-
month-old transgenic mouse heterozygous for the a-
CAT gene. The data show clearly that CAT activity
was confined to the eye. We assume that the a-CAT
gene was present in all the organs. Southern-blot anal-
ysis showed that the a-CAT gene was integrated into
the DNA of the liver and brain, but the other organs
of the transgenic mice were not tested.
Further tests were performed on separated eye tissues
of the transgenic mice (derived from mouse 7378). The
results are shown in Figure 19. Bacterial CAT activity
was found both in the epithelia and fibers of the trans-
genic mouse lenses. By contrast, no CAT activity was
detected in the retina (R) or the remaining (X) eye
tissues. In other transgenic mice containing the CAT
gene fused to the long terminal repeat of Rous sarcoma
virus, CAT activity was preferentially directed to organs
rich in tendon, bone, and muscle.
This indicates
the importance of the aA-crystallin promoter sequence
in directing CAT activity to the lens.
o # o o
c o
JS o

Fig. 18. CAT activity in
various organs of a 6-month-
old heterozygous transgenic
mouse. Samples (10 ng of
protein) from the homoge-
nates of the indicated organs
were assayed for CAT activ-
ity. CM, CM-AQ, and CM-
as in Figure 16. Taken
from Overbeek and cowork-
Having established the striking tissue-specificity of
the murine aA-crystallin promoter, we investigated
whether the a-CAT fusion gene is co-regulated with
the endogenous aA-crystallin gene in the transgenic
mice. Earlier studies in the rat showed that a-crystallin
is the first crystallin synthesized during development
of the rodent lens.
Subsequently, immunofluores-
cence studies have shown that a-crystallin appears in
the developing lens near day 11 in the embryonic
We thus compared the time at which «A-

- CM
T *
Fig. 19. Localization of CAT activity within pooled eye tissues
from 10 F2 transgenic mice which were heterozygous for the a-CAT
gene. Homogenates were assayed as follows: lane R, retina; lane Ep,
lens epithelium; lane X, all eye tissues except lens and retina; lane
F, lens fiber cells. Chloramphenicol and acetylated derivatives ab-
breviated as in Figure 16. Taken from Overbeek and coworkers.
crystallin is detectable in an immunoblot assay with
that at which CAT activity becomes evident in the Fl
progeny from the a-CAT transgenic mouse 7378. In
these experiments, both a-crystallin and CAT activity
were first detected in the excised eyes of 12.5-day-old
CAT activity was not detected in homog-
enates of the 13.5-day-old embryonic bodies. There
was a larger increase in CAT activity than in a-crystallin
content between 12.5 and 13.5 days of development,
suggesting that the a-CAT and endogenous aA-crys-
tallin promoters may not be activated at precisely the
same time in the developing lens. This small time dif-
ference, however, may be due to mRNA or protein
stability, or other unknown variables in the experi-
mental procedures. In general, the results indicated that
the aA-crystallin promoter in the a-CAT fusion gene
is properly regulated in the transgenic mice.
It is interesting that lens clarity and crystallin content
were not affected in the a-CAT transgenic mice. Thus,
the addition of a second aA-crystallin promoter in the
mouse genome does not appear to reduce the expres-
sion of the authentic aA-crystallin gene. This is con-
sistent with the observation that progeny of our trans-
genic mice which were homozygous for the a-CAT gene
expressed twice as much CAT activity as progeny which
were heterozygous for this fusion gene. Together, these
findings suggest that the regulatory factors controlling
the aA-crystallin promoter are not rate-limiting in the
embryonic mouse lens.
The stringent specificity of the aA-crystallin pro-
moter does not appear to be shared by all crystallin
genes. Initial experiments using a mixture of cDNAs
containing crystallin sequences indicated that embry-
onic chicken retina express crystallin genes.
sequent experiments using cloned cDNAs identified 5-
crystallin mRNAs in a number of tissues from chicken
More recently, 5-crystallin genes have
been integrated into PCC3 mouse teratocarcinoma
stem cells.
5-Crystallin was found in three lines of
these cells after they differentiated into skeletal muscle,
columnar epithelia, and unidentified spindle-shaped
cells. Experiments in which deletion mutants of the
No. 1 LENS GENE EXPRESSION / Piorigorsky
51-crystallin gene were microinjected into mouse fi-
broblasts have provided evidence that the sequence be-
tween nucleotide positions —93 and —80 are respon-
sible for the low level activity of this promoter in the
non-lens cells.
Identifying the different factors in-
volved in regulating the crystallin genes and under-
standing how different elements of the crystallin pro-
moters interact with one another to control the expres-
sion of their respective genes are among the exciting
challenges for future research.
There are a number of examples showing tissue-
preference for the expression of foreign genes in trans-
genic mice (reviewed elsewhere
). Many of these
examples include the use of extensive 5' flanking se-
quences, internal gene sequences, and even 3' flanking
sequences, leaving the precise regions which are re-
sponsible for regulating the expression of the intro-
duced gene unidentified. The most defined sequences
for tissue-specific expression in transgenic mice are
from the rat elastase I promoter, where 213 base pairs
(-205 to +8) can direct gene expression in the acinar
cells of the pancreas,
and our experiments, where
412 base pairs (—366 to +46) of the murine aA-crys-
tallin promoter can direct gene expression in the lens.
In addition to the present experiments showing an ap-
propriate developmental control of the a-CAT fusion
gene, proper developmental regulation of foreign adult
0-globin genes has been demonstrated in transgenic
It has even been possible to alter the de-
velopmental timing of the
7-globin gene from its fetal
expression in humans to an ancestral embryonic
expression by introducing it into the germ line of
These data indicate that there are at least some
genes, including those expressed in the eye, that can
be displaced from their normal chromosomal position,
and yet be properly regulated by relatively few se-
quences at their 5' end. This does not mean that there
are not also control sequences located within the
or even in their 3' flanking region.
Transformation of Lens Cells by an
aA-Crystallin-SV40 T Antigen
Fusion Gene in Transgenic Mice
The results with the a-CAT fusion gene suggested
to us that it might be possible to modify the behavior
of lens cells by using the aA-crystallin promoter to
drive other genes in genetically engineered mice. As
an initial step in this direction, we attempted, in col-
laboration with Dr. Heiner Westphal and his col-
leagues, to neoplastically transform lens cells in the eye
in transgenic mice by placing the SV40 T antigen gene
under the control of the aA-crystallin promoter. This
would serve to test further the ability of the aA-crys-
tallin promoter to direct the expression of foreign genes
in the lens and to examine the possibility that the lens
is resistant to malignant transformation.
There are no published reports of lens tumors, sug-
gesting an apparent resistance of the lens to malignancy.
Already, in 1948, Sachs and Larsen
speculated on
possible anticarcinogenic factors concerning the lens.
These included physical, metabolic, chemical, and
physiological considerations. Numerous reports, how-
ever, involving chemical, spontaneous, and viral trans-
formation indicate that lens cells are not intrinsically
refractive to uncontrolled proliferation or tumor for-
mation (see Piatigorsky
for review). Moreover, cells
in the lens epithelium of two strains of chicken
and in certain cataracts display multilayering and ab-
normal growth regulation (see von Sallman and
for further references). It is particularly
interesting that thioacetamide-induced cataracts in
rainbow trout showed marked proliferation of the lens
epithelium and contained a tumor-like cell mass.
We attempted to generate lens tumors in transgenic
mice using the T antigen gene of SV40 fused to the
aA-crystallin promoter, since it has already been dem-
onstrated that cultured lens epithelial cells from ham-
sters could be transformed with SV40 virus.
has also been shown in transgenic mice that the SV40
early region genes (coding for large and small T antigen)
produce tumors in the choroid plexus of the cerebel-
The choroid plexus tumors were found to be
due to the preferential expression of the large T antigen
gene in this region of the brain directed by the 72 base
pair repeat SV40 enhancer sequence.
Removal of
the SV40 enhancer caused peripheral neuropathies,
hepatocellular carcinomas, and pancreatic islet aden-
omas in transgenic mice, implying a change in the tis-
sue preference for T antigen gene expression.
In other
experiments, pancreatic tumors have been induced in
transgenic mice containing the SV40 T antigen gene
driven by the rat insulin II
or the rat elastase I
Initially, we obtained seven transgenic mice con-
taining one or more copies of the a-T antigen fusion
Figure 20 compares the appearance of a normal
eye with that of an a-T antigen transgenic mouse. The
lens of the transgenic mouse is opaque and whitish-
yellow. Transgenic mice with the a-T antigen fusion
gene die between 3 and 4 months of age; however, we
have been able to propagate two strains. We do not
know why these mice die prematurely.
Analysis of the a-T antigen transgenic mice is in
progress, and the details will be reported elsewhere
(Mahon and coworkers, in preparation). Histological
examination revealed that the eyes of 3'/2-month-old
transgenic mice bearing the a-T antigen gene were filled
Vol. 28
Fig. 20. The mouse on the left is a normal NIH inbred strain FVB/
N. The transgenic mouse on the right {from the same strain) with
whitish-yellow eyes carries the a-T antigen fusion gene in its chro-
mosomes. Taken from Westphal and coworkers.
with dividing cells, which appeared to have been de-
rived from the ruptured lens. This tumor-like mass was
vascularized, and contained cells with aberrant shapes.
Immunofluorescence studies indicated that the cells
filling the eye contained crystallins in their cytoplasm
in 18-day-old a-T antigen transgenic mice. Many cells
had a- and /3-crystallins, while only a few cells had y-
crystallin. The nuclei of these cells, both in the eye and
after cultivation, showed positive immunofluorescence
for SV40 T antigen (Fig. 21). This is consistent with
the interpretation that T antigen expression caused
transformation of the lens cells. Current investigations
indicate that lens cell transformation by SV40 T an-
tigen begins early in the embryonic eye. These exper-
iments demonstrate that the aA-crystallin promoter
can be used to alter lens phenotype by genetic engi-
Summary and a Look Ahead
In a relatively short period of time, the lens crystallins
have been redefined in terms of their genes. Each crys-
tallin gene family has a characteristic pattern of introns
reflecting its evolutionary history. Sequencing of genes
and cDNAs has accelerated our knowledge of the pri-
mary structures of the different crystallins. An impor-
tant advance coming from sequence comparisons is
the newly discovered ancestral relationship linking the
/3- and 7-crystallins into a /ify-superfamily of proteins.
Introns have a structural role by dividing exons which
encode individual structural motifs in the jS-crystallins
and domains in the 7-crystallins. Introns may also have
a functional role, as indicated by the generation of aA
and orA
polypeptides from the single murine aA-
crystallin gene by alternative RNA splicing.
5-Crystallin never ceases to amaze. Only one (51) of
two almost identical, linked genes produces two similar
polypeptides from the same mRNA by a still unknown
mechanism. The crystallin genes are differentially ex-
pressed in a time- and space-dependent manner during
lens development. Regulation exists both between and
within crystallin gene families. Crystallin promoters
function with tissue-specificity in foreign species.
Relatively few 5
flanking sequences composing the
murine aA-crystallin promoter can be used to express
foreign genes selectively in the lens. The developmental
regulation of a foreign gene under the control of an
aA-crystallin promoter is similar to that of the endog-
enous aA-crystallin gene in transgenic mice. Use of
crystallin promoters has thus opened new possibilities
for genetic engineering in the visual system.
As we look to the future and pursue practical solu-
tions to the problems related to genetic engineering,
we should not forget that it was curiosity and basic
studies that brought us to this point. We are only be-
ginning to understand the mysteries of genes and their
expression. Although genetic engineering and gene
therapy are very much on the forefront of current
thought, germ line applications to humans seem distant
(reviewed elsewhere
). We must remember that
relatively few microinjections of foreign genes into
mouse eggs are successful. It is not yet possible to target
a foreign gene into a specific, benign site in the host
genome or to eliminate a mutant gene from the chro-
Consequently, gene additions may be very
damaging to the individual, depending on the integra-
tion site,
although recent studies are making pro-
gress in this area.
Even a successful gene implant
results in a heterozygous individual, creating uncer-
tainty as to which offspring would inherit the new gene,
Fig. 21. Indirect immunofluorescent staining of cultured lens cells
from the eye of an a-T antigen transgenic mouse using a monoclonal
antibody against SV40 T antigen. The nuclear locations of the positive
immunofluorescence is consistent with the presence of SV40 T an-
tigen. Taken from Westphal and coworkers.
No. 1
Clearly, much work needs to be accomplished before
germ line genetic engineering reaches the clinic. So-
matic gene therapy for humans appears closer.
should be noted, however, that significant advances
have been accomplished in the ability to genetically
engineer the eye of Drosophilia.
On a more positive note, recombinant DNA tech-
nology has initiated a cascade of experimentally ap-
proachable questions which probe genomic structure
and function more deeply than ever imaginable. We
can now investigate not only how genes are expressed,
but what genes control their expression. In addition to
studying gene function per se, recombinant DNA
methodology allows us to examine structural and
functional components of ocular proteins at the mo-
lecular level. The production of transgenic mice (and
possibly other species
) can be used to develop animal
models for metabolic disorders and to explore the basis
for hereditary and metabolic diseases involving single
gene defects in vision. The Philly mouse cataract is an
example of a model hereditary eye disorder that may
be amenable to experimentation at the level of gene
It is also possible to begin experiments
attempting to dissect genetic events during develop-
ment, such as, for example, the creation of transgenic
mice harboring anti-sense gene sequences.
A useful
by-product of transgenic experiments is the possibility
of obtaining insertional mutants. Examples of inser-
tional mutants which identify genetic loci for important
biological processes include the recently reported limb
deformities in transgenic mice.
The possibilities
for leaps in our understanding of gene function seem
to be limited principally by imagination, resources, and
available time. Thus, we should use our new tools to
investigate the richness of the genome and visual system
in the spirit of Jonas Friedenwald's acceptance remarks
for the first Proctor Medal Award:
The slogan of our association should not be
that basic research is recondite, solemn, austere,
but that it is easy, joyous, and exciting. The or-
chard is full of golden fruit. One can hardly take
a step without discovering something new and il-
Let us listen to his words. Thank you for listening
to me and thank you for this wonderful award.
I am greatly indebted to my past and present laboratory
associates, without whom this work would not have been
accomplished. In particular I acknowledge the following:
Peggy S. Zelenka, Leonard M. Milstone, Toshimichi Shi-
nohara, David C. Beebe, Rosalie Reszelbach, Suraj P. Bhat,
Raymond E. Jones, Harry Ostrer, George Inana, J. Fielding
Hejtmancik, John M. Nickerson, James W. Hawkins, Jacques
A. Treton, Mark A. Thompson, Graeme J. Wistow, Eric F.
Wawrousek, George Thomas, Charlotte A. Peterson, and
John F. Klement (post-doctoral fellows); Sydney P. Craig,
James C. Liang, Deborah DeFeo-Jones, Charles R. King,
Cynthia J. Jaworski, Bernd Sommer, and Diana S. Parker
(graduate students); Teresa Borras, Ana B. Chepelinsky, and
Gokul C. Das (senior associates); Sonia S. Rothschild, Miriam
Wollberg, Aida S. Wakil, and Barbara Norman (research as-
sistants); and Leah A. Williams and David S. McDevitt (sab-
batical professors). In addition to their scientific contributions,
they have made my life in the laboratory a joyful and learning
I am grateful to Drs. Ana B. Chepelinsky, Graeme J. Wis-
tow, Joseph Horwitz, Suraj P. Bhat, and Robert Nussenblatt
for critically reading this manuscript. I also thank Dawn
Chicchirichi for remarkably able and good-humored secre-
tarial support.
1. Coulombre AJ: The eye. In Organogenesis, DeHaan RL and
Ursprung H, editors. Holt, Rinehart and Winston, Inc., 1965,
pp. 219-251.
2. Papaconstantinou J: Molecular aspects of lens cell differentia-
tion. Science 156:338, 1967.
3. Piatigorsky J: Lens differentiation in vertebrates. A review of
cellular and molecular features. Differentiation 19:134, 1981.
4. Verhoeff FH and Friedenwald JS: Injury to cornea and con-
junctiva due to fish bile. Am J Ophthalmol 5:857, 1922.
5. Rabaey M: Electrophoretic and immunoelectrophoretic studies
on the soluble proteins in the developing lens of birds. Exp Eye
Res 1:310, 1962.
6. Zwaan J and Ikeda A: An immunochemical study on the beta-
crystallins of the chicken lens: Ontogenetic and phylogenetic
aspects. In Eye Structure, II Symp., Rohen JW, editor. Stuttgart,
Schattauer-Verlag, 1965, pp. 419-429.
7. Zwaan J and Ikeda A: Macromolecular events during differ-
entiation of the chicken lens: Exp Eye Res 7:301, 1968.
8. Genis-Galvez JM, Castro JM, and Battaner E: Lens soluble
proteins: Correlation with the cytological differentiation in the
young adult organ of the chick. Nature 217:652, 1968.
9. Maisel H and Langman J: An immuno-embryological study
on the chick lens. J Embryol Exp Morphol 9:191, 1961.
10. Rabaey M: Lens proteins during embryonic development of
different vertebrates. Invest Ophthalmol 4:560, 1965.
11. Genis-Galvez JM, Maisel H, and Castro J: Changes in chick
lens proteins with aging. Exp Eye Res 7:593, 1968.
12. Zwaan J: Lens-specific antigens and cytodifferentiation in the
developing lens. J Cell Physiol 72:47, 1968.
13. Clayton RM: Problems of differentiation in the vertebrate lens.
In Current Topics in Developmental Biology, Moscona AA
and Monroy A, editors. Vol. 5, New York, Academic Press,
1970, pp. 115-180.
14. Maisel H and Goodman M: Analysis of mammalian lens pro-
teins by electrophoresis. Arch Ophthalmol 71:671, 1964.
15. Spector A: The soluble proteins of the lens. Invest Ophthalmol
4:579, 1965.
16. Clayton RM and Truman DES: Molecular structure and an-
tigenicity of lens proteins. Nature 214:1201, 1967.
17. Schoenmakers JGG and Bloemendal H: Subunits of alpha-
crystallin from adult and embryonic cattle lens. Nature 220:
790, 1968.
18. Zwaan J: Electrophoretic studies on the heterogeneity of the
chicken lens crystallins. Exp Eye Res 7:461, 1968.
19. Zigler JS Jr and Goosey J: Aging of protein molecules: lens
crystallins as a model system. TIBS 6:133, 1981.
20. Harding JJ and Crabbe MJC: The lens: Development, proteins,
metabolism and cataract. In The Eye, 3rd edition, Davson H,
editor. New York, Academic Press, 1984, pp. 207-492.
21. Morner CT: Unteruchungen der protein substanzen in den
lichtbrechenden medien des Auges. Hoppe Seyler Z Physiol
Chem 18:61, 1894.
22. Bloemendal H: The lens proteins. In Molecular and Cellular
Biology of the Eye Lens, Bloemendal H, editor. New York,
John Wiley and Sons, 1981, pp. 1-47.
23. Piatigorsky J: Delta crystallins and their nucleic acids. Mol Cell
Biochem 59:33, 1984.
24. Stapel SO, Zweers A, Versteeg M, Dodemont H, and de Jong
WW: f-Crystallin, a novel avian and reptilian eye lens protein.
Proc Int Soc Eye Res. Vol II, Eindhoven, The Netherlands,
1982, p. 43.
25. Tomarev SI, Zinovieva RD, Dolgilevich SM, Luchin SV, Krayev
AS, Skryabin KG, and Gause GG Jr: A novel type of crystallin
in the frog eye lens. 35-kDa polypeptide is not homologous to
any of the major classes of lens crystallins. FEBS Lett 171:297,
26. Brahma SK and van der Starre H: Ontogeny and localization
of the a-, |8- and 5-crystallin antigens during lens development
in mallard embryos. Curr Eye Res 2:663, 1983.
27. Williams LA, Piatigorsky J, and Horwitz J: Structural features
of 6-crystallin of turtle lens. Biochim Biophys Acta 708:49, 1982.
28. Stapel SO and de Jong WW: Lamprey 48-kDa lens protein
represents a novel class of crystallins. FEBS Lett 162:305, 1983.
29. de Pomerai D, Ellis DK, and Carr A: A lamprey lens protein
related to avian and reptilian 5-crystallin. Curr Eye Res 3:729,
30. Williams LA, Ding L, Horwitz J, and Piatigorsky J: r-Crystallin
from the turtle lens: Purification and partial characterization.
Exp Eye Res 40:741, 1985.
31. de Jong WW: Evolution of lens and crystallins. In Molecular
and Cellular Biology of the Eye Lens, Bloemendal H, editor.
New York, John Wiley and Sons, 1981, pp. 221-278.
32. Slingsby C: The structural variations in lens crystallins. TIBS
10:281, 1985.
33. Bloemendal H: Lens proteins. CRC Critical Rev Biochem 12:
1, 1982.
34. Bours J: Isoelectric focusing and immunochemistry of lens
crystallins. Doc Ophthalmol 37:1, 1974.
35. Shinohara T, Robinson EA, Appella E, and Piatigorsky J: Mul-
tiple 7-crystallins of the mouse lens: Fractionation of mRNAs
by cDNA cloning. Proc Natl Acad Sci USA 79:2783, 1982.
36. Hoenders HJ and Bloemendal H: Lens proteins and aging. J
Gerontology 38:278, 1983.
37. Harding JJ: Changes in lens proteins in cataract. In Molecular
and Cellular Biology of the Eye Lens, Bloemendal H, editor.
New York, John Wiley and Sons, 1981, pp. 327-365.
38. Bloemendal H: Biosynthesis of lens crystallins. In Molecular
and Cellular Biology of the Eye Lens, Bloemendal H, editor.
New York, John Wiley and Sons, 1981, pp. 189-220.
39. Reszelbach R, Shinohara T, and Piatigorsky J: Resolution of
two distinct embryonic chick 6-crystallin bands by polyacryl-
amide gel electrophoresis in the presence of sodium dodecyl
sulfate and urea. Exp Eye Res 25:583, 1977.
40. Bhat SP and Piatigorsky J: Molecular cloning and partial char-
acterization of 5-crystallin cDNA sequences in a bacterial plas-
mid. Proc Natl Acad Sci USA 76:3299, 1979.
41. Bhat SP, Jones RE, Sullivan MA, and Piatigorsky J: Chicken
lens crystallin DNA sequences show at least two 5-crystallin
genes. Nature 284:234, 1980.
42. Zelenka P and Piatigorsky J: Reiteration frequency of 5-crystallin
DNA in lens and non-lens tissues of chick embryos. 5-Crystallin
gene is not amplified during lens cell differentiation. J Biol Chem
251:4294, 1976.
43. Jones RE, Bhat SP, Sullivan MA, and Piatigorsky J: Comparison
of two 5-crystallin genes in the chicken. Proc Natl Acad Sci
USA 77:5879, 1980.
44. Yasuda K, Kondoh H, Okada TS, Nakajima N, and Shimura
Y: Organization of 5-crystallin genes in the chicken. Nuc Acids
Res 10:2879, 1982.
45. Hawkins JW, Nickerson JM, Sullivan MA, and Piatigorsky J:
The chicken 6-crystallin gene family. Two genes of similar
structure in close chromosomal approximation. J Biol Chem
259:9821, 1984.
46. King CR, Shinohara T, and Piatigorsky J: «A-crystallin mes-
senger RNA of the mouse lens: More non-coding than coding
sequences. Science 215:985, 1982.
47. Inana G, Shinohara T, Maizel JV Jr, and Piatigorsky J: Evo-
lution and diversity of the crystallins. Nucleotide sequence of
a 0-crystallin mRNA from the mouse lens. J Biol Chem 257:
9064, 1982.
48. Peterson CA and Piatigorsky J: Preferential conservation of the
globular domains of the /SA3/A1 -crystallin polypeptide of the
chicken eye lens. Gene 45:139, 1986.
49. Breitman ML, Lok S, Wistow G, Piatigorsky J, Treton JA,
Gold RJM, and Tsui L-C: 7-Crystallin family of the mouse
lens: Structural and evolutionary relationships. Proc Natl Acad
Sci USA 81:7762, 1984.
50. Nickerson JM and Piatigorsky J: Sequence of a complete chicken
5-crystallin cDNA. Proc Natl Acad Sci USA 81:2611, 1984.
51. Yasuda K, Nakajima N, Isobe T, Okada TS, and Shimura Y:
The nucleotide sequence of a complete chicken 5-crystallin
cDNA. EMBO J 3:1397, 1984.
52. Dodemont HJ, Andreoli PM, Moormann RJM, Ramaekers
FCS, Schoenmakers JGG, and Bloemendal H: Molecular clon-
ing of mRNA sequences encoding rat lens crystallins. Proc Natl
Acad Sci USA 78:5320, 1981.
53. Moormann RJM, van der Velden HMW, Dodemont HJ, An-
dreoli PM, Bloemendal H, and Schoenmakers JGG: An un-
usually long non-coding region in rat lens a-crystallin messenger
RNA. Nuc Acids Res 9:4813, 1981.
54. Bower DJ, Errington LH, Wainwright NR, Sime C, Morris S,
and Clayton RM: Cytoplasmic RNA sequences complementary
to cloned chick 5-crystallin cDNA show size heterogeneity.
Biochem J 201:339, 1981.
55. Moormann RJM, den Dunnen JT, Bloemendal H, and
Schoenmakers JGG: Extensive intragenic sequence homology
in two distinct rat lens 7-crystallin cDNAs suggests duplications
of a primordial gene. Proc Natl Acad Sci USA 79:6876, 1982.
56. Tomarev SI, Dolgilevich SM, Kozlov KA, Zinovieva RD,
Dzhumagaliev EB, Kogan GL, Skobeleva NA, Milhailov AT,
Frolova LY, and Gause GG Jr: Molecular cloning of double-
stranded cDNA from the eye lens of the frog Rana temporaria:
Construction of the cDNA clonotheque and identification of a
clone containing the nucleotide sequences of the 7-crystallin
gene. Gene 17:131, 1982.
57. Tomarev SI, Zinovieva RD, Dolgilevich SM, Krayev AS, Skry-
abin KG, and Gause GG Jr: The absence of the long 3'-non-
translated region in mRNA coding for eye lens aA
of the frog (Rana temporaria). FEBS Lett 162:47, 1983.
58. Hejtmancik JF and Piatigorsky J: Diversity of ^-crystallin
mRNAs of the chicken lens: Hybridization analysis with cDNA
clones. J Biol Chem 258:3382, 1983.
59. Gorin MB and Horwitz J: Cloning and characterization of a
cow beta crystallin cDNA. Curr Eye Res 3:939, 1984.
60. Quax-Jeuken Y, Janssen C, Quax W, van den Heuvel R, and
Bloemendal H: Bovine /8-crystallin complementary DNA clones:
Alternating proline/alanine sequence of j8Bl subunit orginates
from a repetitive DNA sequence. J Mol Biol 180:457, 1984.
61. Bhat SP and Spector A: Complete nucleotide sequence of a
No. 1 LENS GENE EXPRESSION / Piorigorsky
cDNA derived form calf lens 7-crystallin mRNA: Presence of
Alu I-Iike DNA sequences. DNA 3:287, 1984.
62. Quax-Jeuken Y, Driessen H, Leunissen J, Quax W, de Jong
W, and Bloemendal H: j8s-Crystallin: structure and evolution
of a distinct member of the (87-superfamily. EMBO J 4:2597,
63. den Dunnen JT, Moormann RJM, and Schoenmakers JGG:
Rat lens /3-crystallins are internally duplicated and homologous
to 7-crystallins. Biochim Biophys Acta 824:295, 1985.
64. Hejtmancik JF, Thompson MA, Wistow G, and Piatigorsky J:
cDNA and deduced protein sequence for the /3B1-crystallin
polypeptide of the chicken lens. Conservation of the PAPA
sequence. J Biol Chem 261:982, 1986.
65. Inana G, Piatigorsky J, Norman B, Slingsby C, and Blundell
T: Gene and protein structure of a /3-crystallin polypeptide in
murine lens: relationship of exons and structural motifs. Nature
302:310, 1983.
66. King CR and Piatigorsky J: Alternative RNA splicing of the
murine aA-crystallin gene: Protein-coding information within
an intron. Cell 32:707, 1983.
67. Lok S, Tsui L-C, Shinohara T, Piatigorsky J, Gold R, and Breit-
man M: Analysis of the mouse 7-crystallin gene family: assign-
ment of multiple cDNAs to discrete genomic sequences and
characterization of a representative gene. Nuc Acids Res 12:
4517, 1984.
68. Moormann RJM, Jongbloed R, and Schoenmakers JGG: Iso-
lation and characterization of /3- and 7-crystallin genes from
rat genomic cosmid libraries. Gene 29:1, 1984.
69. Quax-Jeuken Y, Quax W, van Rens G, Khan PM, and Bloe-
mendal H: Complete structure of the aB-crystallin gene: Con-
servation of the exon-intron distribution in the two nonlinked
a-crystallin genes. Proc Natl Acad Sci USA 82:5819, 1985.
70. van den Heuvel R, Hendriks W, Quax W, and Bloemendal H:
Complete structure of the hamster a A crystallin gene. Reflection
of an evolutionary history by means of exon shuffling. J Mol
Biol 185:273, 1985.
71. Meakin SO, Breitman ML, and Tsui L-C: Structural and evo-
lutionary relationships among five members of the human 7-
crystallin gene family. Mol Cell Biol 5:1408, 1985.
72. Moormann RJM, den Dunnen JT, Heuyerjans J, Jongbloed
RJE, van Leen RW, Lubsen NH, and Schoenmakers JGG:
Characterization of the rat 7-crystallin gene family and its
expression in the eye lens. J Mol Biol 182:419, 1985.
73. Nickerson JM, Wawrousek EF, Hawkins JW, Wakil AS, Wistow
GJ, Thomas G, Norman BL, and Piatigorsky J: The complete
sequence of the chicken 61 crystallin gene and its 5' flanking
region. J Biol Chem 260:9100, 1985.
74. den Dunnen JT, Moormann RJM, Lubsen NH, and Schoen-
makers JGG: Intron insertions and deletions in the /37-crystallin
gene family: The rat /3B1 gene. Proc Natl Acad Sci USA 83:
2855, 1986.
75. Nickerson JM, Wawrousek EF, Borras T, Hawkins JW, Norman
BL, Filpula DR, Nagle JW, Ally AH, and Piatigorsky J: Se-
quence of the chicken 52 crystallin gene and its intergenic spacer.
Extreme homology with the 61 crystallin gene. J Biol Chem
261:552, 1986.
76. den Dunnen JT, Moormann RJM, Lubsen NH, and Schoen-
makers JGG: Concerted and divergent evolution within the rat
7-crystallin gene family. J Mol Biol 189:37, 1986.
77. McDevitt DS, Hawkins JW, Jaworski CJ, and Piatigorsky J:
Isolation and partial characterization of the human aA-crystallin
gene. Exp Eye Res 43:285, 1986.
78. Piatigorsky J: Lens crystallins and their gene families. Cell 38:
620, 1984.
79. Blundell T, Lindley P, Miller L, Moss D, Slingsby C, Tickle I,
Turnell B, and Wistow G: The molecular structure and stability
of the eye lens: X-ray analysis of 7-crystallin II. Nature 289:
771, 1981.
80. Wistow G, Turnell B, Summers L, Slingsby C, Moss D, Miller
L, Lindley P, and Blundell T: X-ray analysis of the eye lens
protein 7-II crystallin at 1.9 A resolution. J Mol Biol 170:175,
81. Driessen HPC, Herbrink P, Bloemendal H, and de Jong WW:
The /3-crystallin Bp chain is internally duplicated and homol-
ogous with 7-crystallin. Exp Eye Res 31:243, 1980.
82. Driessen HPC, Herbrink P, Bloemendal H, and de Jong WW:
Primary structure of the bovine /3-crystallin Bp chain. Internal
duplication and homology with 7-crystallin. Eur J Biochem
121:83, 1981.
83. Berbers GAM, Hoekman WA, Bloemendal H, de Jong WW,
KJeinschmidt T, and Braunitzer G: Homology between the pri-
mary structures of the major bovine /3-crystallin chains. Eur J
Biochem 139:467, 1984.
84. Schoenmakers JGG, den Dunnen JT, Moormann RJM, Jong-
bloed R, van Leen RW, and Lubsen NH: The crystallin gene
families. In Human Cataract Formation, CIBA Foundation
Symposium 106, Nugent J and Whelan J, editors. London,
Pitman, 1984, pp. 208-218.
85. Lindley PF, Narebor ME, Summers LJ, and Wistow GJ: The
structure of lens proteins. In The Ocular Lens, Maisel H, editor.
New York and Basel, Marcel Dekker, Inc, 1985, pp. 123-167.
86. Summers L, Slingsby C, White H, Narebor M, Moss D, Miller
L, Mahadevan D, Lindley P, Driessen H, and Blundell T: The
molecular structures and interactions of bovine and human 7-
crystallins. In Human Cataract Formation, CIBA Foundation
Symposium 106, Nugent J and Whelan J, editors. London,
Pitman, 1984, pp. 219-236.
87. Wistow G, Slingsby C, and Blundell T: Eye-lens proteins: The
three-dimensional structure of/3-crystallin predicted from mo-
nomeric 7-crystallin. FEBS Lett 133:9, 1981.
88. Berbers GAM, Hoekman WA, Bloemendal H, de Jong WW,
KJeinschmidt T, and Braunitzer G: Proline- and alanine-rich
N-terminal extension of the basic bovine /3-crystallin Bl chains.
FEBS Lett 161:225, 1983.
89. Bloemendal H, Berbers GAM, de Jong WW, Ramaekers FCS,
Vermorken AJM, Dunia I, and Benedetti EL: Interaction of
crystallins with the cytoskeletal-plasma membrane complex of
the bovine lens. In Human Cataract Formation, CIBA Foun-
dation Symposium 106, Nugent J and Whelan J, editors. Lon-
don, Pitman, 1984, pp. 177-190.
90. Piatigorsky J, Nickerson JM, King CR, Inana G, Hejtmancik
JF, Hawkins JW, Borras T, Shinohara T, Wistow G, and Nor-
man B: Crystallin genes: templates for lens transparency. In
Human Cataract Formation, CIBA Foundation Symposium
106, Nugent J and Whelan J, editors. London, Pitman, 1984,
pp. 191-207.
91. Hogg D, Tsui L-C, Gorin M, and Brietman ML: Characteriza-
tion of the human 0-crystallin gene Hu/3A3/Al: Transcription
of a single mRNA containing two potential initiation codons.
J Biol Chem, in press.
92. van der Ouderaa FJ, de Jong WW, Helderink A, and Bloemendal
H: The amino-acid sequence of the aB
chain of bovine a-
crystallin. Eur J Biochem 49:157, 1974.
93. Cohen LH, Smits DPEM, and Bloemendal H: Isolation and
characterization of rat-lens messenger RNAs. Comparison of
lens proteins, synthesized in lens culture and in homologous
and heterologous cell-free systems. Eur J Biochem 67:563, 1976.
94. Cohen LH, Westerhuis LW, Smits DP, and Bloemendal H:
Two structurally closely related polypeptides encoded by 14-S
mRNA isolated from rat lens. Eur J Biochem 89:251, 1978.
95. Cohen LH, Westerhuis LW, de Jong WW, and Bloemendal H:
Rat a-crystallin A chain with an insertion of 22 residues. Eur
J Biochem 89:259, 1978.
96. de Jong WW, Cohen LH, Leunissen JAM, and Zweers A: In-
ternally elongated rodent a-crystallin A chain: Resulting from
incomplete RNA splicing? Biochem Biophy Res Comm 96:
648, 1980.
97. King CR and Piatigorsky J: Alternative splicing of aA-crystallin
RNA. Structural and quantitative analyses of the mRNAs for
the aA
and aA
-crystallin polypeptides. J Biol Chem 259:
1822, 1984.
98. Gillies SD, Morrison SL, Oi VT, and Tonegawa S: A tissue-
specific transcription enhancer element is located in the major
intron of a rearranged immunoglobulin heavy chain gene. Cell
33:717, 1983.
99. Banerji J, Olson L, and Schaffner W: A lymphocyte-specific
cellular enhancer is located downstream of the joining region
in immunoglobulin heavy chain genes. Cell 33:729, 1983.
100. Queen C and Baltimore D: Immunoglobulin gene transcription
is activated by downstream sequence elements. Cell 33:741,
101. Gruss P: Magic enhancers? DNA 3:1, 1984.
102. Khoury G and Gruss P: Enhancer elements. Cell 33:313, 1983.
103. Wistow G: Domain structure and evolution in a-crystallins and
small heat-shock proteins. FEBS Lett 181:1, 1985.
104. Shinohara T, Reszelbach R, and Piatigorsky J: Two tryptic
peptide differences among the subunits of 5-crystallin of the
embryonic chick lens. Exp Eye Res 30:361, 1980.
105. Shinohara T and Piatigorsky J: Regulation of protein synthesis,
intracellular electrolytes and cataract formation in vitro. Nature
207:406, 1977.
106. Piatigorsky J: Intracellular ions, protein metabolism, and cat-
aract formation. Curr Top Eye Res 3:1, 1980.
107. Shinohara T and Piatigorsky J: Anion and cation effects on 5-
crystallin synthesis in the cultured embryonic chick lens and
in a reticulocyte lysate. Exp Eye Res 30:351, 1980.
108. Das GC and Piatigorsky J: Transcription of the two 5-crystallin
genes in a Hela cell extract. In Transcriptional Control Mech-
anisms, Granner DK, Rosenfeld G, and Chang S, editors. UCLA
Symposium, New York, Alan R. Liss Inc, 1986, in press.
109. Borras T, Nickerson JM, Chepelinsky AB, and Piatigorsky J:
Structural and functional evidence for differential promoter ac-
tivity of the two linked 5-crystallin genes in the chicken. EMBO
J 4:445, 1985.
110. Wawrousek EF, Nickerson JM, and Piatigorsky J: Two 5-crys-
tallin polypeptides are derived from a cloned 51-crystallin
cDNA. FEBS Lett 205:235, 1986.
111. Agata K, Yasuda K, and Okada TS: Gene coding for a lens-
specific protein, 5-crystallin, is transcribed in nonlens tissues of
chicken embryos. Dev Biol 100:222, 1983.
112. Bower DJ, Errington LH, Cooper DN, Morris S, and Clayton
RM: Chicken lens 5-crystallin gene expression and methylation
in several non-lens tissues. Nuc Acids Res 11:2513, 1983.
113. Shinohara T and Piatigorsky J: Quantitation of 5-crystallin
messenger RNA during lens induction in chick embryos. Proc
Natl Acad Sci USA 73:2808, 1976.
114. Piatigorsky J, Webster H DeF, and Craig SP: Protein synthesis
and ultrastructure during the formation of embryonic chick
lens in vivo and in vitro. Dev Biol 27:176, 1972.
115. Yoshida K and Katoh A: Crystallin synthesis by chick lens. II.
Changes in synthetic activities of epithelial and fiber cells during
embryonic development. Exp Eye Res 11:184, 1971.
116. Brahma SK and v.d. Starre H: Studies on biosynthesis of soluble
lens crystallin antigens in the chick by isoelectric focusing in
thin-layer polyacrylamide gel. Exp Cell Res 97:175, 1975.
117. Bagchi M, Maisel H, and Ellis M: Protein synthesis in organ
cultured chick lens. Exp Eye Res 35:325, 1982.
118. Ostrer H and Piatigorsky J: /3-CrystaHins of the adult chicken
lens: Relatedness of the polypeptides and their aggregates. Exp
Eye Res 30:679, 1980.
119. Thomson I, Wilkinson CE, Jackson JF, de Pomerai DI, Clayton
RM, Truman DES, and Williamson R: Isolation and cell-free
translation of chick lens crystallin mRNA during normal de-
velopment and transdifferentiation of neural retina. Dev Biol
65:372, 1978.
120. Zelenka P and Piatigorsky J: Isolation and in vitro translation
of 5-crystallin mRNA from embryonic chick lens fibers. Proc
Natl Acad Sci USA 71:1896, 1974.
121. Treton JA, Shinohara T, and Piatigorsky J: Degradation of 5-
crystallin mRNA in the lens fiber cells of the chicken. Dev Biol
92:60, 1982.
122. Hejtmancik JF, Beebe DC, Ostrer H, and Piatigorsky J: 5- and
/3-Crystallin mRNA levels in the embryonic and posthatched
chicken lens: Temporal and spatial changes during development.
Dev Biol 109:72, 1985.
123. Takata C, Albright JF, and Yamada T: Lens fiber differentiation
and gamma crystallins: Immunofluorescent study of Wolffian
regeneration. Science 147:1299, 1965.
124. McAvoy JW: Cell division, cell elongation and the co-ordination
of crystallin gene expression during lens morphogenesis in the
rat. J Embryol exp Morph 45:271, 1978.
125. McAvoy JW: Cell division, cell elongation and distribution of
a-, /3- and 7-crystallins in the rat lens. J Embryol exp Morph
44:149, 1978.
126. Ostrer H, Beebe DC, and Piatigorsky J: /3-Crystallin mRNAs:
Differential distribution in the developing chicken lens. Dev
Biol 86:403, 1981.
127. Bower DJ, Errington LH, Pollock BJ, Morris S, and Clayton
RM: The pattern of expression of chick 5-crystallin genes in
lens differentiation and in trans-differentiating cultured tissues.
EMBO J 2:333, 1983.
128. de Jong WW: Eye lens proteins and vertebrate phylogeny. In
Macromolecular Sequences in Systematic and Evolutionary
Biology. Goodman M, editor. New York, Plenum Publishing
Corp., 1982, pp. 75-114.
129. de Jong WW, Zweers A, Versteeg M, and Nuy-Terwindt C:
Primary structures of the a-crystallin A chains of twenty-eight
mammalian species, chicken and frog. Eur J Biochem 141:131,
130. Myers RM, Tilly K, and Maniatis T: Fine structure genetic
analysis of a /3-globin promoter. Science 232:613, 1986.
131. Gorman CM, Moffat LF, and Howard BH: Recombinant ge-
nomes which express chloramphenicol acetyltransferase in
mammalian cells. Mol Cell Biol 2:1044, 1982.
132. Chepelinsky AB, King CR, Zelenka P, and Piatigorsky J: Lens-
specific expression of the chloramphenicol acetyltransferase gene
promoted by 5' flanking sequences of the murine aA-crystallin
gene in explanted chicken lens epithelia. Proc Natl Acad Sci
USA 82:2334, 1985.
133. Gordon JW and Ruddle FH: DNA-mediated genetic transfor-
mation of mouse embryos and bone marrow—a review. Gene
33:121, 1985.
134. Phillpott GW and Coulombre AJ: Lens development. II. The
differentiation of embryonic chick lens epithelial cells in vitro
and in vivo. Exp Cell Res 38:635, 1965.
135. Piatigorsky J and Rothschild SS: Loss during development of
the ability of chick embryonic lens cells to elongate in culture:
Inverse relationship between cell division and elongation. Dev
Biol 28:382, 1972.
136. Piatigorsky J, Rothschild SS, and Milstone LM: Differentiation
of lens fibers in explanted embryonic chick lens epithelia. Dev
Biol 34:334, 1973.
137. Piatigorsky J: Lens cell elongation in vitro and microtubules.
Ann Acad Sci 253:333, 1975.
No. 1 LENS GENE EXPRESSION / Piorigorsky
138. Graham FL and van der Eb AJ: A new technique for the assay
of infectivity of human adenovirus 5 DNA. Virology 52:456,
139. Stow ND and Wilkie NM: An improved technique for obtaining
enhanced infectivity with herpes simplex virus type I DNA. J
Gen Virol 33:447, 1976.
140. Overbeek PA, Chepelinsky AB, Khillan JS, Piatigorsky J, and
Westphal H: Lens-specific expression and developmental reg-
ulation of the bacterial chloramphenicol acetyltransferase gene
driven by the murine aA-crystallin promoter in transgenic mice.
Proc Natl Acad Sci USA 82:7815, 1985.
141. Chepelinsky AB and Piatigorsky J: Activator sequences of the
murine aA-crystallin gene. J Cell Biochem (Suppl)10D:182,
142. Okazaki K, Yasuda K, Kondoh H, and Okada TS: DNA se-
quences responsible for tissue-specific expression of a chicken
a-crystallin gene in mouse lens cells. EMBO J 4:2589, 1985.
143. Kondoh H, Yasuda K, and Okada TS: Tissue-specific expression
of a cloned chick 5-crystallin gene in mouse cells. Nature 301:
440, 1983.
144. Hayashi S, Kondoh H, Yasuda K, Soma G-I, Ikawa Y, and
Okada TS: Tissue-specific regulation of a chicken 6-crystallin
gene in mouse cells: involvement of the 5' end region. EMBO
J 4:2201, 1985.
145. Williams LA and Piatigorsky J: Comparative and evolutionary
aspects of 5-crystallin in the vetebrate lens. Eur J Biochem 100:
349, 1979.
146. Clayton RM: Comparative aspects of lens proteins. In The Eye,
Vol. 5, Davson H and Graham LT, editors. New York, London,
Academic Press, 1974, pp. 399-494.
147. Lok S, Breitman ML, Chepelinsky AB, Piatigorsky J, Gold RJM,
and Tsui L-C: Lens-specific promoter activity of a mouse y-
crystallin gene. Mol Cell Biol 5:2221, 1985.
148. Lok S, Stevens W, Breitman M, Gold R, and Tsui L-C: Iden-
tification of a lens-specific enhancer element in the mouse
gamma-2 crystallin gene. J Cell Biochem (Suppl)lOD: 121, 1986.
149. Treton JA, Jones RE, King CR, and Piatigorsky J: Evidence
against 7-crystallin DNA or RNA sequences in the chicken.
Exp Eye Res 39:513, 1984.
150. Westphal H, Overbeek PA, Killan JS, Chepelinsky AB, Schmidt
A, Mahon KA, Bernstein KE, Piatigorsky J, and de Crom-
brugghe B: Promoter sequences of murine ak crystallin, murine
a2(I) collagen or of avian sarcoma virus genes linked to the
bacterial chloramphenicol acetyl transferase gene direct tissue-
specific patterns of chloramphenicol acetyl transferase expres-
sion in transgenic mice. Cold Spring Harbor Symposia on
Quantitative Biology, Vol. L. New York, Cold Spring Harbor
Laboratory, 1985, pp. 411-416.
151. Overbeek PA, Lai S-P, Van Quill KR, and Westphal H: Tissue-
specific expression in transgenic mice of a fused gene containing
RSV terminal sequences. Science 231:1574, 1986.
152. Zwaan J: The appearance of a-crystallin in relation to cell cycle
phase in the embryonic mouse lens. Dev Biol 96:173, 1983.
153. Clayton RM, Thomson I, and de Pomerai DI: Relationship
between crystallin mRNA expression in retina cells and their
capacity to re-differentiate into lens cells. Nature 282:628, 1979.
154. Jackson JF, Clayton RM, Williamson R, Thomson I, Truman
DES, and de Pomerai DI: Sequence complexity and tissue dis-
tribution of chick lens crystallin mRNAs. Dev Biol 65:383,
155. Kondoh H, Takahashi Y, and Okada TS: Differentiation-de-
pendent expression of the chicken 5-crystallin gene introduced
into mouse teratocarcinoma stem cells. EMBO J 3:2009, 1984.
156. Palmiter RD and Brinster RL: Transgenic mice. Cell 41:343,
157. Molecular biology of development, Cold Spring Harbor Sym-
posia on Quantitative Biology, Vol. L. New York, Cold Spring
Harbor Laboratory, 1985, pp. 361-420.
158. Ornitz DM, Palmiter RD, Hammer RE, Brinster RL, Swift
GH, and MacDonald RJ: Specific expression of an elastase-
human growth hormone fusion gene in pancreatic acinar cells
of transgenic mice. Nature 313:600, 1985.
159. Chada K, Magram J, Raphael K, Radice G, Lacy E, and Cos-
tantini F: Specific expression of a foreign /3-globin gene in ery-
throid cells of transgenic mice. Nature 314:377, 1985.
160. Magram J, Chada K, and Costantini F: Developmental regu-
lation of a cloned adult /8-globin gene in transgenic mice. Nature
315:338, 1985.
161. Townes TM, Lingrel JB, Chen HY, Brinster RL, and Palmiter
RD: Erythroid-specific expression of human /3-globin genes in
transgenic mice. EMBO J 4:1715, 1985.
162. Chada K, Magram J, and Costantini F: An embryonic pattern
of expression of a human fetal globin gene in transgenic mice.
Nature 319:685, 1986.
163. Hesse JE, Nickol JM, Lieber MR, and Felsenfeld G: Regulated
gene expression in transfected primary chicken erythrocytes.
Proc Natl Acad Sci USA 83:4312, 1986.
164. Sachs E and Larsen RL: Cancer and the lens. Am J Ophthalmol
31:561, 1948.
165. Clayton RM: Failure of growth regulation of the lens epithelium
in strains of fast-growing chicks. Genet Res Camb 25:79, 1975.
166. Clayton RM, Eguchi G, Truman DES, Perry MM, Jacob J,
and Flint OP: Abnormalities in the differentiation and cellular
properties of hyperlastic lens epithelium from strains of chickens
selected for high growth rate. J Embryol exp Morph 35:1, 1976.
167. Odeigah PGC, Clayton RM, and Truman DES: Ultrastructural
and biochemical abnormalities of lens cell membranes from
two strains of chicks with epithelial hyperplasia. Exp Eye Res
28:311, 1979.
168. von Sallmann L, Halver JE, Collins E, and Grimes P: Thioac-
etamide-induced cataract with invasive proliferation of the lens
epithelium in rainbow trout. Cancer Res 26:1819, 1966.
169. Albert DM, Rabson AS, Grimes PA, and von Sallmann L:
Neoplastic transformation in vitro of hamster lens epithelium
by simian virus 40. Science 164:1077, 1969.
170. Bloemendal H, Lenstra JA, Dodemont H, Ramaekers FCS,
Groeneveld AA, Dunia I, and Benedetti EL: SV40-transformed
hamster lens epithelial cells: a novel system for the isolation of
cytoskeletal messenger RNAs and their translation products.
Exp Eye Res 31:513, 1980.
171. Bloemendal H, Ramaekers FCS, Lenstra JA, Dodemont H,
Dunia I, and Benedetti EL: Neoplastic transformation of lens
cells in vitro. In Applied Methods in Oncology, Vol. 3, Williams
GM, Kroes R, Waaljers HW, and van de Poll KW, editors.
Amsterdam, Elsevier/North Holland Biomedical Press, 1980,
pp. 199-212.
172. Brinster RL, Chen HY, Messing A, van Dyke T, Levine AJ,
and Palmiter RD: Transgenic mice harboring SV40 T-antigen
genes develop characteristic brain tumors. Cell 37:367, 1984.
173. Palmiter RD, Chen HY, Messing A, and Brinster RL: SV40
enhancer and large-T antigen are instrumental in development
of choroid plexus tumours in transgenic mice. Nature 316:457,
174. Messing A, Chen HY, Palmiter RD, and Brinster RL: Peripheral
neuropathies, hepatocellular carcinomas and islet cell adenomas
in transgenic mice. Nature 316:461, 1985.
175. Hanahan D: Heritable formation of pancreatic /J-cell tumours
in transgenic mice expressing recombinant insulin/simian virus
40 oncogenes. Nature 315:115, 1985.
176. Ornitz DM, Palmiter RD, Messing A, Hammer RE, Pinkert
CA, and Brinster RL: Elastase I promoter directs expression of
human growth hormone and SV40 T antigen genes to pancreatic
acinar cells in transgenic mice. In Molecular Biology of De-
velopment, Symposia on Quantitative Biology, Vol. L, Cold
Spring Harbor Laboratory, 1985, pp. 399-409.
177. Westphal H, Khillan JS, Mahon KA, Overbeek PA, Chepelinsky
AB, Piatigorsky J, Schmidt A, and de Crombrugghe B: Speci-
ficity of gene expression and insertional mutagenesis in trans-
genic mice. In New Frontiers in the Study of Gene Functions,
New York, Plenum Publishing Co., in press.
178. Marx JL: Gene therapy—so near and yet so far away. Science
232:824, 1986.
179. Walters L: The ethics of human gene therapy. Nature 320:225,
180. Maniatis T: Targeting in mammalian cells. Nature 317:205,
181. Wagner EF, Covarrubias L, Stewart TA, and Mintz B: Prenatal
lethalities in mice homozygous for human growth hormone
gene sequences integrated in the germ line. Cell 35:647, 1983.
182. Lacy E, Roberts S, Evans EP, Burtenshaw MD, and Costantini
FD: A foreign /3-globin gene in transgenic mice: Integration at
abnormal chromosomal positions and expression in inappro-
priate tissues. Cell 34:343, 1983.
183. Harbers K, Kuehn M, Delius H, and Jaenisch R: Insertion of
retrovirus into the first intron of a 1(1) collagen gene leads to
embryonic lethal mutation in mice. Proc Natl Acad Sci USA
81:1504, 1984.
184. Breindl M, Harbers K, and Jaenisch R: Retrovirus-induced le-
thal mutation in collagen 1 gene of mice is associated with an
altered chromatin structure. Cell 38:9, 1984.
185. Lohler J, Timpl R, and Jaenisch R: Embryonic lethal mutation
in mouse collagen 1 gene causes rupture of blood vessels and
is associated with erythropoietic and mesenchymal cell death.
Cell 38:597, 1984.
186. Hartung S, Jaenisch R, and Breindl M: Retrovirus insertion
inactivates mouse a 1(1) collagen gene by blocking initiation of
transcription. Nature 320:365, 1986.
187. Smithies O, Gregg RG, Boggs SS, Koralewski MA, and Kuch-
erlapati RS: Insertion of DNA sequences into the human chro-
mosomal 0-globin locus by homologous recombination. Nature
317:230, 1985.
188. Thomas KR, Folger KR, and Capecchi MR: High frequency
targeting of genes to specific sites in the mammalian genome.
Cell 44:419, 1986.
189. Parkman R: The application of bone marrow transplantation
to the treatment of genetic diseases. Science 232:1373, 1986.
190. Montell C, Jones K, Hafen E, and Rubin G: Rescue of the
Drosophila phototransduction mutation trp by germline trans-
formation. Science 230:1040, 1985.
191. Hammer RE, Pursel VG, Rexroad Jr CE, Wall RJ, Bolt DJ,
Ebert KM, Palmiter RD, and Brinster RL: Production of trans-
genic rabbits, sheep and pigs by microinjection. Nature 315:
680, 1985.
192. Kador PF, Fukui HN, Fukushi S, Jernigan HM, and Kinoshita
JH: Philly mouse: A new model of hereditary cataract. Exp Eye
Res 30:59, 1980.
193. Carper D, Shinohara T, Piatigorsky J, and Kinoshita JH: De-
ficiency of functional messenger RNA for a developmentally
regulated /3-crystallin polypeptide in a hereditary cataract. Sci-
ence 217:463, 1982.
194. Izant JG and Weintraub H: Constitutive and conditional
suppression of exogenous and endogenous genes by anti-sense
RNA. Science 229:345, 1985.
195. Woychik RP, Stewart TA, Davis LG, D'Eustachio P, and Leder
P: An inherited limb deformity created by insertional muta-
genesis in a transgenic mouse. Nature 318:36, 1985.
196. Friedenwald JS: Remarks on the acceptance of the Proctor
Award. Am J Ophthalmol 32:4, 1949.
197. Ingolia TD and Craig EA: Four small Drosophila heat shock
proteins are related to each other and to mammalian a-crys-
tallin. Proc Natl Acad Sci USA 79:2360, 1982.
198. Wistow G, Summers L, and Blundell T: Myxococcus xanthus
spore coat protein S may have a similar structure to vertebrate
lens 187-crystallins. Nature 316:771, 1985.
199. Crabbe MJC: Partial sequence homology of human myc on-
cogene protein to beta and gamma crystallins. FEBS Lett 181:
157, 1985.
200. Law ML, Cai G-Y, Hartz J, Kao F-T, Hogg D, Breitman ML,
and Tsui L-C: Localization of a /3-crystallin gene, Hu/3A3/Al
(gene symbol: CRYB1), to the long arm of human chromosome
17. Hum Genet, in press.
201. Sparkes RS, Mohandes T, Heinzmann C, Gorin MB, Zollman
S, and Horwitz J: Assignment of a human beta crystallin gene
to 17 cen-q23*. Hum Genet, in press.
202. den Dunnen JT, Jongbloed RJE, van Kessell AHMG, and
Schoenmakers JGG: Human lens 7-crystallin sequences are
located in the pl2-qter region of chromosome 2. Hum Genet
70:217, 1985.
203. Willard HF, Meakin SO, Tsui L-C, and Breitman ML: Assign-
ment of human gamma crystallin multigene family to chro-
mosome 2. Somatic Cell and Mol Genetics 11:511, 1985.
204. Gilbert W: Why genes in pieces? Nature 271:501, 1978.
205. Blake CCF: Do genes-in-pieces imply proteins-in-pieces? Nature
273:267, 1978.
206. Gluzman Y: SV40 transformed simian cells support the rep-
lication of early SV40 mutants. Cell 23:175, 1981.