, A
(V

l-'hl-L-_-‘r

" ““ -—- —--K '

Michigan“: -2253’
basally l

 

 

This is to certify that the

thesis entitled

THE NUCLEOTIDE SEQUENCE
OF THE CHICKEN

CARBONIC ANHYDRASE II GENE

presented by

Corinne Misae Yoshihara

has been accepted towards fulfillment
of the requirements for

Doctor of Philosophy degree in Genetics

 

1' '
ajor Q[{A‘.¥professor/\7

MS U i: an Affirmative Action/Equal Opportunity Institution

Date 2/21/86

 

04639

THE NUCLEOTIDE SEQUENCE
OF THE CHICKEN

CARBONIC ANHYDRASE II GENE

By

Corinne Misae Yoshihara

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Genetics Interdisciplinary Program

1986

Copyright by
CORINNE MISAE YOSHIHARA
1986

 

 

ABSTRACT

THE NUCLEOTIDE SEQUENCE OF THE CHICKEN
CARBONIC ANHYDRASE II GENE

By

Corinne Misae Yoshihara

The complete nucleotide sequence of the coding region of the
chicken carbonic anhydrase II (CA II) gene has been determined from
clones isolated from a ACharon 4A chicken genomic library: The gene is
approximately 17 kilobase pairs (kb) in size and codes for a protein
that is comprised of 259 amino acid residues. Six introns ranging in
size from 0.3-lO.3 kb interrupt the gene. The S'-flanking region
contains the ATA and.CCAAT consensus sequences commonly associated with
eucaryotic genes transcribed by RNA Polymerase II. The S'-flanking
region farther upstream is extremely GC-rich and characteristic short
stretches of GC sequences are found around the CCAAT sequence. The 3'-
flanking region contains long stretches of the nucleotides A and T.
There is a relatively long untranslated region with 5 possible sites of
polyadenylation, the most probable site being the one closest to the
stop codon.

A comparison of the derived amino acid sequence of chicken CA II
to that of mouse CA II shows a 65% amino acid sequence homology. The
number of introns as well as 5 of the 6 intron locations are conserved

between the two homologous genes. The site of the fourth intron is

Corinne Misae Yoshihara
shifted by 4 2/3 codons further 3' in the chicken and now falls
between the codons 147/148 rather than within the codon 143 as in the
mouse gene. A comparison of the 5'-flanking regions of mouse and
chicken CA II genes shows that the genes are quite homologous in a
region up to 80 bp 5' to the first nucleotide coding for mRNA (Cap
site), although they are less similar than the mouse and human CA II

genes in this region.

To My Mother and Father

ii

ACKNOWLEDGEMENTS

I would like to thank several people for their support during my
career as a graduate student.

First I would like to thank Dr. Jerry Dodgson for the Opportunity
to be in science and for the guidance and training he provided.

I would next like to thank Dr. Richard Tashian and Dr. Pat Venta
for the informative conversations, materials, and helpful suggestions
they offered from afar.

I would also like to thank my committee members for their patience
and suggestions through it all.

The lab. would not be the same without its members. I would like
to acknowledge the people in the lab. for the help and friendship they
offered.

I would like to thank my friend S. Decker for his patience and
understanding through the difficult times.

Last of all I would especially like to thank my parents and
brother for their long distance support and for their belief in my

abilities.

iii

TABLE OF

List of Tables

List of Figures . . . . . . . .

INTRODUCTION

CHAPTER 1. LITERATURE REVIEH .

The Carbonic Anhydrase Protein

General Properties

Mammalian Isozyme Comparison

CA Isozymes

CONTENTS

Comparison of CA I, CA II,

CA IV

and

Mitochondrial CA . . . -,- . .

Tissue Distribution and Function

CA I - CA V

Transport Epithelia . . .

Macromolecule-secreting

Nonepithelial Cells . . .

unique CA's

Expression During Development . .
Expression in Erythrocytes

variability in Expression of
Changes in Isozyme Levels During

Maturation . . . . . . . . . .

Quantitative Genetic variation

Qualitative Genetic Variation . . .

Hormonal Control

Evolution of Carbonic Anyhdrases

The Carbonic Anhydrase Gene
Mouse CA II Gene

Isolation of Mouse CA II Gene

Rabbit CA I cDNA

iv

Isozymes

Red

Cell

Page
vi
vii

COMM» u: N N N

Page
CHAPTER 1. (CONTINUED)

Comparison of Mouse and Human CA II Genes . . . . . . . 27
References I I I I I I I I I I I I I I I I I I I I I I I I I 29
CHAPTER 2. ISOLATION OF THE CHICKEN CARBONIC ANYHDRASE II GENE . 36
Acknowledgments I I I I I I I I I I I I I I I I I I I I I I 39
References I I I I I I I I I I I I I I I I I I I I I I I I I 3 9

CHAPTER 3. THE NUCLEOTIDE SEQUENCE OF THE CHICKEN CARBONIC

ANHYDRASE II GENE . . . . . . . . . . . . . . . . . . no
Introduction . . . . . . . . . . . . . . . . . . . . . . . . no
Experimental procedures . . . . . . . . . . . . . . . . . . HZ

Isolation of cDNA . . . . . . . . . . . . . . . . . . . AZ

Isolation of ACA III and ACA XVI . . . . . . . . . . . “2
Preparation of Subclones Containing Fragments of

the CA II Gene . . . . . . . . . . . . . . . . . . . AB

Restriction Enzyme Digestion and Southern Blotting . . ”3

DNA Sequence Analysis . . . . . . . . . . . . . . . . . “3

Nuclease Protection/End Analysis . . . . . . . . . . . nu

Materials . . . . . . . . . . . . . . . . . . . . . . . #5

Results . . . . . . . . . . . . . . . . . . . . . . . . . . “5

Isolation of Larger Chicken CA II cDNA Clones . . . . . 45

Isolation of the Chicken CA II Gene . . . . . . . . . . “5

DNA Sequencing Strategy of the Chicken CA II Gene . . . A9

Sequence of the Chicken CA II Gene . . . . . . . . . . 52

Chicken CA II Gene Intron/Exon Organization . . . . . . 63

5' and 3' Ends of the Chicken CA II mRNA . . . . . . . 66

5' and 3' Flanking Sequences of the Chicken CA II Gene. 69

Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

References . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

LIST OF TABLES

TABLE PAGE
CHAPTER 1
I Residues located, or postulated to occur, in the active

site regions of the carbonic anhydrase isozymes, CA I,
CA II and CA III I I I I I I I I I I I I I I I I I I I I I 8

CHAPTER 3
I Nucleotide and amino acid divergence of chicken CA II gene 61
11 Unique and invariant residues for the mammalian CA

isozymes and the homologous residues in the chicken CA II
isozyme I I I I I I I I I I I I I I I I I I I I I I I I I 62

111 DNA sequence of intron donor and acceptor sites . . . . . 64

IV Intron size comparison of the chicken and mouse CA II
genes I I I I I I I I I I I I I I I I I I I I I I I I I I 65

vi

FIGURE

LIST OF FIGURES

CHAPTER 1
Location of amino acid substitutions [o] for genetic
variants of the human CA I, CA II and CA III isozymes on a
stylized version of the alpha-carbon chain of human CA I
and CA II molecules, which have very similar three-
dimensional structures. . . .. . .. . .. . .. . .. .
Restriction endonuclease map and exon positions of the
mouse carbonic anhydrase II gene in cosmid cosMCAII-Z...
CHAPTER.2

cDNA clone of carbonic anhydrase II . . . . . . . . . . .

NUOIIODIGB sequence or pPES-OoB o o e o o a o o o o o o 0

CHAPTER 3
Restriction map of the chicken CA II gene locus . . . . .

Restriction maps of the subclcned DNA fragments of the
chicken CA II gene. . . . . . . . . . . . . . . . . . . .

DNA sequence of the chicken CA II gene . . . . . . . . . .

Amino acid sequence comparison of chicken and mouse CA II

genes I I I I I I I I I I I I I I I I I I I I I I I I I
5' end analysis of chicken CA II cytoplasmic mRNA . . .

DNA sequence of putative signal sequences flanking the CA
II gene I I I I I I I I I I I I I I I I I I I I I I I I I I

Comparison of the 5' flanking region of the chicken,
mouse, and human CA II genes. .. . .. . .. . .. . ..

Comparison of intron A location in chicken and mouse CA II

genes I I I I I I I I I I I I I I I I I I I I I I I I I I I

Nucleotide sequence of Sharlanking and intron regions of
the chicken CA II gene . . . . . . . . . . . . . . . . . .

vii

PAGE

25

37
38

48

51

54

59
68

71

7H

79

86

INTRODUCTION

Carbonic anhydrase (CA) is a zinc metalloenzyme that functions as
a general acid-base catalyst. It is because it has such a basic role
that CA is fundamental to nearly all organisms.

CA was first found in equine blood as the enzyme responsible for
catalyzing C02 hydration. It was not until chromatographic separation
allowed the purification of individual proteins that red blood cell CA
was found to be two isozymes, CA I and CA II. Since then, there have
been several other isozymes of CA that have been identified from cells
other than erythrocytes. Genetic studies as well as protein structure
data provided evidence that the isozymes were coded for by separate
genes.

This review will first discuss what is known about CA.protein, its
structure and function. In the following section the review will cover
the more recent findings on CA gene structure and its role as a model

system in which to study eucaryotic gene expression.

CHAPTER 1

LITERATURE REVIEW

THE CARBONIC ANHYDRASE PROTEIN

 

General Properties

 

Since its initial identification in equine blood CA has been found
in a wide variety of organisms, including bacteria, algae, yeast,
mosses, plants and numerous vertebrate and invertebrate species. CA is
thought to have such diverse physiological roles as C02 fixation
(photosynthesis); calcification (shell formation); respiration, where
the specific catalytic role of the enzyme is the interconversion of 002
and HCOé (the hydration of C02 released in body tissues and the
dehydration of HCOé in the lungs); and the transport and accumulation
of H+ or H005 in organs of secretion such as stomach and kidney (for a
review see 1,2)

The primary reaction catalyzed by CA is the reversible hydration
of C02; i.e., C02 + H20 ‘-9 H605 4- 11+. However, CA also utilizes
substrates other than C02 and HCOé. It is able to hydrolyze certain
esters of carboxylic acids (3JHS,6), carbonic acids (7), and cyclic
sulfonates (8), and hydrate a variety of aldehydes (9,10) and pyruvic

acid (11).

Mammalian Isozyme Comparison

 

CA Isozymes

Five different CA isozymes have been identified in mammalian
species. Three soluble isozymes of CA; CA I, CA II, and CA III, are
known to exist in amniotes (birds, reptiles, and mammals) (12). A
fourth isozyme, a membrane-bound CA, has been found in mammalian brain
(13), lung (14,15), and kidney (16,17,18). A fifth isozyme, a soluble
mammalian CA in mitochondria (19,20) that has been implicated in the
urea cycle, gluconeogenesis and fatty acid synthesis (21), has recently
been characterized. A form that is synthesized in the parotid accinar

cells and exported to saliva (22) is possibly another isozyme.

Comparison of CA I, CA II, and CA III

The isozymes that have been well characterized.(I,II,III) vary in
terms of catalytic efficiency, amino acid composition, appearance
during development, sulfonamide inhibition, tissue distribution, and
sometimes in molecular weight.

The amino acid sequence of human CA was first determined in 1972
(23). Since then the amino acid sequences of several CA I, II, and III
isozymes from different species have been determined (33). IHowever,
since CA I and CA 11 are the best characterized isozymes, most of the
isozyme comparisons are made between these twat Comparisons made with
the newer isozymes are limited.

CA I and CA II are similar in amino acid composition. Both
isozymes contain a high proportion of proline residues, acidic and
basic residues, and few sulfur-containing residues. The high activity

enzyme, CA II, contains more basic residues (histidine, lysine, and

4
arginine) than CA I. CA I, on the other hand, has a high serine
content. In mammals, CA I possesses one and a half times as many
serine residues as CA II (24). Despite these differences there is 60%
homology between the amino acid sequences of CA I and CA II in humans
(25).

In contrast, the amino acid composition of CA III is distinct from
that of I and II. CA III has been found to have a higher basic amino
acid content (26).

CA I and CA 11 isozymes also differ in their immunological
properties. The antigenic sites of CA are independent of the active
sites. The enzyme-antibody complex thus retains some CA activity and
the presence of sulfonamide inhibitor (which binds to the active site)
does not affect enzyme-antibody complex.formation. The antigenic sites
differ for CA I and CA II. Antisera specific for the’ human CA I enzyme
will not cross-react with CA II and vice verse" In fact, there is more
homology between the same isozymes between species than.between.CA I
and CA II of the same species. CA I and CA II from species closely
related to man cross-react with antisera prepared against the
homologous human isoenzyme (2%

CA I and CA 11 isozymes differ in their sulfonamide binding
affinities. Sulfonamides inhibit CA activity by binding to the zinc
ion in the active site of the CA molecule. CA II isozymes bind most
sulfonamides more readily than the other CA isozymes (27) and this
difference in binding reflects the difference among the isozymes at the
active site. For example, the isozyme with the least homology among

the three isozymes, CA III, is the least sensitive to inhibition (26)

in chicken.

This type of variation in isozyme inhibition also holds
true for anions that bind to the active site. The C02 hydrase activity
of human CA I is strongly inhibited by physiological levels of chloride
and bicarbonate ions found in human erythrocytes whereas the activity
of CA II is not (28).

Although all the CA's catalyze the reversible reaction C02 + H206
H003 + H+, the three isozymes vary considerably in their catalytic
efficiencies. CA II has a higher specific 002 hydrase activity than CA
I, and CA III has the lowest activity. The specific activities of
chicken CA isozymes toward COz/H2C03 differ according to the ratio
III:I:II - 1:6:47 (26).

The x-ray diffraction study of crystals of human erythrocyte CA II
was initiated by Strandberg in 1962 and by 1975 the three-dimensional
structures of both human CA I (29,30) and CA II (31,32)'had been
determined. The tertiary structures of CA I and CA II are very
similar. Human CA I and CA II are ellipsoidal in shape and measure 41
x 42 x 55 A (Figure 1). Approximately 20% of the CA protein exists as
helical structures and another 35% of the protein comprises a large
beta-structure which is made up of 10 segments. The beta-structure is
the dominating feature of the protein and.passes through the whole
molecule dividing the molecule in halffi Each half contains 6 pieces of
polypeptide chain that run perpendicular to the beta-structure. The
active site cavity is conical, approximately 12 A deep and contains a

single zinc ion at the bottom. The zinc is liganded by His 94, His 96,

237

 

Ann

 

 

 

 

 

 

I Anpomy I Ly: I BIu-LW
u cw u Lnocm u 6»
III All III GIu ‘ III Alp

 

 

 

 

 

II ProoI-Iis
III Pro

 

 

IN

 

I TI»- Ly!
II Lou
III Sec

 

 

 

I AspoVII
II Clo
III GIu

 

 

 

I GhoLydArg
II Ly:
III Ly:

 

 

Figure l.

  
  
 

. 31
I Vol
II VaI
III Ila-Val

 

253

._
I Sly—Arg

II AmoAu:
III Mn

 

255
.,______—
I ThtoArg

, ::.

 

 

 

 

 

I AnoGIn I AlpoGIy
II Ly: II GIy
III Am III GIy

 

 

 

 

 

 

Location of amino acid substitutions [o] for genetic
variants of the human CA 1, CA II and CA III isozymes on a
stylized version of the alpha-carbon chain of human CA I and
CA II molecules, which.have very similar three-dimensional
structures. Boxes include the original substitution and
homologous residues at this position for the other human
isozymes. Beta-Structure segments are unshaded and helical
structures are indicated by cylinders. The zinc ion is
shown liganded to histidine residues 94, 96, and 119.
Active site residues unique to human CA III [0] are Lys-64,
Thr-65, Arg-67, Val-69, and Arg 91. (Source: Biology and
Chemistry of the Carbonic Anhydrases. 1984. R.E. Tashian
and D. Hewett-Emmet, eds.)

7
and His 119. One-half of the active site is composed of hydrophobic
side chains and the other half is composed of hydrophilic residues
which are exposed at the surface of the site (2)

Five types of residues of CA I and CA II determined from the
tertiary structures appear to be important in maintaining stability and
activity of the molecules (27). These residues include 3 aromatic
clusters, 23 active site residues, 30 hydrophobic core residues, and 8
residues hydrogen-bonded to the zinc ligands, His 94 and His 119, or to
the zinc-bound solvent molecule.

The amino acid residues that are linked to the metal ligands by a
hydrogen-bond network are invariant in all sequenced CA's. However,
there are certain characteristic differences between the isozymes with
respect to polar amino acid residues in their active sites (Table 1)
(34). Residues 64 and 200 appear to critically modify catalytic
behavior. His is present at position 64 and Thr or Asn at position 200
in all sequenced CA II's. Histidine residues appear in both positions
in all sequenced CA Ifs. The presence of His at position 200 is the
only feature that distinguishes the active sites of CA I from those of
CA II. CA III appears to have Thr at position 200, Lys at position 64
and additional basic arginine residues at positions 67 and 91 in their
active sites. The active site residues that are constant for each
isozyme type but differ between isozymes (67 His/Asn, 69 Asn/Glu, 131
Leu/Phe, 204 Tyr/Leu, and 211 Ile/Val) may be responsible for the
differences in the activities of the two isozymes, and in their
inhibition patterns by sulfonamides and anions (27). The active sites

also can be indirectly influenced by variable residues not in the

. mvo . uumgmnuumaom . a
use 2..:mme .2.2 .2... .mmmmue.::<
caconumo 93 mo muumaaosu can zwoaowm .muusom .: <0 .2 8.2: .o .: <0 .2 6...; 4...... 22.8.2
3.22:3 a c. 23:22.. .2... o. 2.3.... o... .c 36.33. 3:: =2 e. .5223... 65.9 o... 3:23. 3.5: ._ (U c. 9.663.. 3.2.6.... 9... 622.8. .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3.. 2. .6 .6 2.. 2.. 2.... 2:. .3 .3 .3 ...> .= 8...:
.3 2. a... .6 2.. 2.. .3 .3 .3 .3 .6 ...> 2. 2. .2. .= .5
2.. 2.. 2:. S: .3 m... 2... ...> =. 52...:
w> ...> 6 a... 2.. 2.. 2:. .3 .3 .3 .6 ...> 3.. 2... ...> = 8.5:.
.a> ...> .6 3.. 2.. 2.. .3 2... .3 .E. .6 _..> 8.. 2... ..> = 22......
.a> _..> .6 3.. 2.. 2.. .3 .3 .3 .5 .6 ...> 8.. 2... ...> .. 8...:
s> _..> .8 a... 2.. 2.. =3 .3 .3 .3 .6 _..> 3.. 2... ...> = .6
_..> ...> .6 a... 2.. 2.. 23 .3 .3 .E. 6 ._.> 8.. 2... ..> _. 5.5::
2. ...> .9... ta 2.. 2.. 2: .3 .3 3.. 3.. 2. ...> . 2......
..> =5 .8 .3 2.. 2.. 2: 2... .3 a... .6 2. 3.. 2... ...> . 8...:
2. ..> .8 3.. A 2.. 2.. 2: .3 .3 a3 .6 3.. 3.. 2... ..> . .6
2. ...> t... .3 2.. 2.. 2: 2... .3 .3 6 ...> 3.. 3.. _..> . 3.2...
2. ...> tn .3 2.. 2.. 2: .3 .3 .E. 6 ..> 2..... 3.. 2< . 52...:
:N B. 8. x: 8. a. 8. 8. 2.. a. .2 9.. .z .2 .2
2: 2: 6 x ...> x .0. E 5.26
2: .6 2: .6 2: 2: 6 ....> ...< .2. .3 E 3...:
2: .6. 2: .6 2: 2: .6 a... ...> a... .3 .3 =2. .02 .3 .= .6
2: .6 6 2: 2: 6 .2 a> a... .3 m... S... .0. .3 z. 52...:
.6 5.. .0. 2: :2. .om .. 22.2.6
2: .6 2: .6 2: 2: 6 2. 6 =2. .2. 2: 5.. .02 .3 = 882
2: .6 2: .6 2: 2: 6 2. 6 a... .2. 2: =2 .2. .3 = 2......
2: .6 2: .6 2: 2: 6 2. 6 .2 .om 2: 5.. .um .3 = 2...:
2: .6 2: .6 2: 2: 6 .2. 6 =2. .8 2: 5.. tn .3 .. .6
2: .6 2: .6 2: 2: .6 2. 6 22.. 2... 2: 5< t... .3 = 22.5:
2: .6 2: .6 2: 2: 6 2: .2. 2: .2. 2: a... t... .3 . 2......
2: .6 2: .6 2: 2: 6 .2. 2.. 6 .2. 2: .2. .8 .3 . 92o:
2: .6 2: 6 2: 2: 6 2... ...< 2: .3 2: 5.. .92 .3 . .5
2: .6 2: .6 2: 2: 6 2... =2. 2: .0... 2: S... .8 .3 . 2:2...
2: .6 2: .6 .2: 2: 6 2... ...< 2: t. 2: 5.. .8 .3 . 52...:

 

 

 

 

 

 

 

o: 2. B. 8. 8 3 a a s B a. 3. 3 a . 82.32
62:52 2.23:

 

 

.=..n.es.2.¢n.
.. <0 55.82 8.2.2:... 2.526 2.. .o 22...... 2.2 22.3. 2.. 2 .38.. o. 8.238.. 3 62.8.. .22.... . H 3......

active site regions.

CA IV

There is a membrane-bound form of CA designated CA IV (14).
Histochemical and immunohistochemical techniques have shown that
membrane-bound CA is present in most electrolyte-secreting epithelia,
in skeletal muscle (35,36), and in the endothelium of lung capillaries
(37). CA activity is found in plasma membranes, luminal and/or anti-
luminal, and in cell organelles. Membrane-bound CA has been best
studied in the gastrointestinal tract (38,39) and epithelia of the
kidney (40,41).

The amino acid composition of human CA IV differs from that of the
cytoplasmic enzymes in'that CA IV has more glutamic acid, methionine,
and arginine, and less proline and lysine than CA I and CA II (18).
The membrane-bound enzymes of bovine lung (14), rat saliva (42), and
rat brain myelin (43) show the same pattern when compared to the
cytoplasmic forms of the respective species.

The specific activity of purified human CA IV is lower than that
of CA II (18). The inhibition by sulfanilamide is similar to that of
CA II. CA IV of bovine lung (14) and rat brain myelin (13) have
inhibitor and substrate kinetics similar to those of CA II.

CA IV is antigenically different from the cytoplasmic enzymes
(18). This difference is seen for CA IV of human kidney (44) as well

as for the CA IV of bovine lung (14).

10

Mitochondrial CA

Mitochondrial CA (CAmit) has been reported in the liver of guinea
pig (45). The CAmit is in the matrix, it is soluble, and its activity
is very dependent on pH. It is believed to be involved in citrulline
synthesis in the mitochondrial matrix.

CAmit is different from CA I and CA II in guinea pig in being less
negatively charged but it is the same as CA I and CA II in molecular
weight. CAmit has approximately one third the catalytic efficiency of
erythrocyte CA II, is more dependent on pH, more prone to inhibition by
$0; than by 01' and is unaffected up to .OSM by the anions phosphate,
citrate, and succinate at pH 7!. The only resemblance it has to CA II
is its insensitivity to chloride. CAmit makes up 0.14% of the
mitochondrial protein. CAmit appears to be a different isozyme that

may be coded for by a separate genetic locus (46).

Tissue Distribution and Function
CA I - CA V

Carbonic anhydrases have considerable variability in their
patterns of tissue distribution and functional diversity in different
organisms. By immunohistochemical techniques, human CA I and CA II
were first found in erythrocytes (47-50). Human CA II has since been
localized in normal or pathological organs of submandibular gland (51),
stomach (49,51,52), appendix (49,52), liver (51), gall bladder (51),
pancreas (51,53), kidney (51,52,54), trachea and bronchi (51), brain
(48,55,56), retina (56,57), ciliary process (58), bone and cartilage
(90), salivary gland (69), and sweat glands (51,52). CA I is found at

higher levels and is more limited in its distribution. ll:has been

11
localized to appendix (51) and colon (51) in addition to erythrocytes.
CA III is primarily found in red skeletal muscle of sheep, cat and
chicken (26,59), rabbit (60,61), human (62,63) and ox (27,64) but has
also been found in rodent liver (65), rabbit liver and sheep lung (59),

and human smooth cardiac muscle, liver, and lung (66)

Transport Epithelia. CA is found in the epithelial cells of

 

ducts, tubules or surfaces which specialize in ion transport and lack
the capacity for secretion of macromolecules. An example of such a
cell is the gastric parietal cell. Immunocytochemical staining shows
CA to be present in the cytoplasm (51,67) and histochemical staining
shows CA to be present in the membrane (38)Iof the parietal cell. CA
is thought to generate protons which the parietal cell secretes across
the apical plasmalemma into the lumen. The cell releases the
bicarbonate ions, also generated by CA, across the basolateral
plasmalemma alkalinizing the interstitial fluid at the surface of the

epithelhnn(68)

Macromolecule-Secreting Epithelia. CA has been
immunocytochemically located in epithelial cells which contain granules
for storing macromolecules which the cells secrete by a merocrine
secretory mechanism (68). Such CA rich merocrine cells include cells
of the tracheobronchial and submandibular glands in man (51), parotid,
submandibular, and exorbital lachrimal glands in mouse and rat (69).
Since CA has actually been found in the saliva of certain species such
as sheep (70), there is a possibility that CA may be located in the

secretory granule and secreted along with the other molecules.

12

Moreover, a glycosylated CA has been reported in rat salivary gland;
the carbohydrate association expected of a secretory product.

CA has also been found in glycoprotein-secreting cells such as the
goblet cells of the guinea pig intestine (68). In the secretory cells

CA may have a role in dehydrating the secretory product.

Nonepithelial Cells. Immunocytochemistry shows that CA is

 

distributed throughout the cytoplasm in erythrocytes, adipocytes,
oligodendroglia, and skeletal muscle (48,67,7l,72). In erythrocytes CA
correlates with the specialization of the cell in respiratory gas
exchange. Its role in adipocytes is uncertain; it may influence
intracellular pH (68).

There are high levels of CA in oligodendrocytes and myelin sheath
of brain. (3V5 role may involve ion fluxes which are fundamental to
neural reactivity. Since impulse conduction by axons is dependent on
depolarization mediated by Na+ influx, CA in myelin and oligodendroglia
sheathing the axon may be involved in maintaining ion concentration at
the outer surface of the axolemma. CA has been found both in bound and
soluble forms in the central nervous system.(73,74).

Cobalt-trapping experiments show that CA is in the sarcolemma of
all and in the sarcoplasm of some muscle fibers (35). Immunostaining
shows CA II to be present in slow twitch fibers (75). CA is thought to

provide buffering action in the site of rapid 002 production.

13

Unique CA's

CA's that are distinct from the isozymes CA 1, CA II, CA III,
membrane-bound CA, and mitochondrial CA have greater molecular weights.
These newly identified CA's include ovine parotid gland CA, rat saliva
CA, rabbit erythrocyte CA and plant CA.
a. As mentioned in the previous section, the ovine parotid gland which
secretes bicarbonate in saliva has been shown to contain a cytoplasmic
high molecular weight CA (49) in addition to CA II. The native
molecular weight of the new enzyme is 540,000 and its subunit molecular
weight is 45,000. The enzyme is similar to CA II in that it has zinc
and is inhibited by sulfonamide but its amino acid composition shows no
homology to CA II; it has 3 cysteine residues per subunit whereas
cysteine is usually absent from CA's. It is also immunologically
distinct.
b. Rat saliva CA has a higher molecular weight than rat erythrocyte
CA. Up to 35% of the saliva CA is composed of oligosaccharide
components. Its amino acid sequence shows more homology to bovine lung
membrane-bound CA than to erythrocyte CA (22).
c. A 54,000 molecular weight CA has been found in rabbit erythrocytes
(77). The CA appears to be composed of the 30,000 molecular weight CA
I (soluble CA I being at a reduced level) and a 24,000 molecular weight
binding protein. The bound CA appears to be associated with the
erythrocyte membrane where it may be involved in increasing the
diffusion rate of C02 or HC0'3 across the membrane.
d. Plant CA's differ in a number of ways from animal CA's (78). Plant

CA's have a wide range of molecular weights from 42K-250K, usually 180K

14

(78). The plant CA's have a similar amino acid content, subunit size,
zinc content, to the CA's in animals (78). However, the plant CA's
have no esterase activity and as a whole are not strongly inhibited by
sulfonamides. The CAAs can be separated by gradient polyacrylamide gel
electrophoresis into two separate isozymes, a chloroplast and a cytosol
form. The roles of CA in plants may be 1) to facilitate diffusion of
C02 to the site of photosynthetic fixation, 2) regulate pH, 3) provide
bicarbonate in underwater plants, or 4) concentrate inorganic carbon

within the chloroplast (78).

Expression During Development

 

The levels of the CA isozymes change during development. cm.I and
CA II are both present in human fetal blood cells as early as the 22nd
week of gestation.(79). CA I reaches adult levels more slowly when
compared to the rise of hemoglobin (Hb) A levels after birth. Hb A
reaches adult levels at age 6 months whereas CA I has 30% the activity
of adult levels at that time. CA I activity is 60% of adult levels
after 1 year and reaches adult levels after 5-6 years of life.

CA III appears in muscle extracts 8 days after birth (59). CA III
activity is 50% of adult levels 21 days after birth.

In cats, CA II is the only isozyme present in a mid-term fetus
(26). CA III appears in muscle in the late stages and CA I appears in
cecum within 2 days of birth.

The location of CA II in cell types changes during the development
of the vertebrate retina (80). By immunocytochemical staining, CA 11
is found very early in all retinoblast cells. Later, CA becomes

restricted to certain cell types with the cell type depending on the

15

vertebrate class. In adult mammals CA II is found in a specific
neuronal cell type as well as in a type of satellite glial cell. As
one goes down the vertebrate classes CA becomes confined to the glial
cells and then to the primitive horizontal cells. The variability of
CA II localization during development may indicate a.varied role of CA
in tissue functions. The early generalized CA II expression in chick
and mouse retinas may indicate a CA role in eye morphogenesis by
influencing intraocular pressure.

CA has been foundto be involved in formation of avian eggshell
(81). During shell formation blood bicarbonate is transformed into
insoluble calcium carbonate and carbonic acid. The carbonic acid is
rapidly broken down by CA to H20 and C02 to prevent the eggshell from
redissolving by the reversible reaction. CA has been found to be
involved in calcification reactions in other species such as the
mollusks (82), fresh water snails (83), barnacles (84) and reef corals
(85).

CA has also been found to be involved in calcium transport and
dissolution of avian eggshell (86). During chick embryonic development
calcium is supplied to the embryo by two sources. During the first
half of gestation (up to day 9 or 10) the yolk supplies calcium.
During the second half of gestation (days 11-12) shell calcium (CaC03)
is mobilized and embryonic calcium content increases. CA activity has
been found to parallel calcium uptake by the choriallantoic membrane
(CAM). Moreover, sulfonamide inhibits calcium uptake as well as CA
activity in CAM (87,88,89). Chick erythrocyte CA and CAM CA have the

same molecular weight but are antigenically different and are different

16

in C02 hydration. CAM CA is thought to be CA 1.

CA also appears to be involved in bone resorption (90). The
carbonic acid produced by CA may be the acid secreted by osteoclasts to
dissolve mineral. Bone resorption by osteoclasts is accompanied by H+
release and greater CA activity. Further support of CA involvement is
a deficiency of CA 11 which is associated with osteopetrosis where bone
resorption is impaired (91)

CA II has been identified in the mouse brain (glia) by
immunological, electrophoretic, and kinetic studies (73,92,93). Over
50% of the CA in the central nervous system (CNS) glia is:membrane-
bound (94,95) and may be an intrinsic membrane protein. Both soluble
and membrane-bound CA were purified from mouse brain.(13L Both forms
had the same specific activity, esterase activity, the same molecular
weight and isoelectric point which was that of a CA 11 isozyme.
However, pulse chase experiments of CA biosynthesis showed that one
form was not a modification of the other.

CA activity has been found in myelin (73,74,98) in several species
and at several developmental stages. The highest CA activities in
myelin occur in the mouse. In rat, myelin from spinal cord has lower
CA activity than brain myelin (95,97). Immunocytochemical staining is
consistent with the biochemical data that the primary sites of CA in

the CNS is in oligodendroglia and myelin.

Expression in Erythrocytes

 

Variability in Expression of Isozymes
CA is the second most abundant protein in the red blood cell (RBC)

next to hemoglobin (98). There are 1-2 grams of CA in a liter of

17

mammalian RBC (1,99,100).

In mammalian erythrocytes there are two isozymes of CA present--
the low activity isozyme CA I and the high activity isozyme CA II which
has 5-10 times the activity of CA-I (27). It has been shown that both
isozymes contribute substantially to the overall activity (101). The
physiological or metabolic significance of the presence of the two
isozymes in erythrocytes is not known since they appear to perform a
straightforward role.

There is considerable variation in the CA I and CA II protein
levels in mammalian red cells and each species has a characteristic
level. In humans and many other primates the CA I level is greater
than CA II, whereas in some inbred mouse strains the CA II level is
greater. In primates the ratio of CA I:CA II can range from 2.6:1 in
macaque monkeys to 27:1 in orangutan. The ratio in humans is 6:1 (27).

CA I is absent from the RBC of some mammalian species. Only CA II
is present in the red cell in two groups of mammals, ruminants (ox,
sheep, goat, elk, deer) and felids (cat, tiger, lion, jaguar, leopard)
(27). Dog, dolphin, chicken, and shark RBC also lack CA I (24).
Species whose erythrocytes contain both CA I and CA 11 are primates
(man, green monkey, pig-tailed macaque, baboon, spider monkey, rhesus
monkey), deer mouse, pig, rabbit, rat, guinea pig, and horse (24).
Although CA I is absent in blood it is not necessarily absent in other
tissues. CA I is absent from bovine blood but is present in very high
concentration in bovine rumen epithelium (24).

There are inherited deficiencies of red cell CA I in humans,

pigtail macaques, and chinchilla (98,102) where there are no detectable

18

defects associated with the absence of CA I in mature red cells.
Therefore, although CA I may have alternate functions in other tissues,
it does not appear to be essential for adequate rates of C02 exchange
(27). CA II appears to have been more strongly selected for in
mammalian red cells.

Changes in Isozyme Levels During

Red Cell Maturation

CA activity increases as cells mature from erythroblasts to
reticulocytes during erythroid development in rabbit marrow (103).
When lysates from the three major developing cell types (basophilic,
polychromatic, and orthochromatic erythroblasts) and reticulocytes in
rabbit marrow are measured for the levels of CA.I and CA II by
immunoassay, CA 11 levels remain.unchanged in erythroblasts but
increase in reticulocytes. CA I shows a progressive increase in the
developing cells. The increased activity of CA during erythroid
development cannot be accounted for on the basis of the isozyme levels
alone since CA II has a 20-fold higher specific hydrase activity than
CA I. Immunoreactive precursors of CA II may be inactive, for example,
because they lack zinc which is at lower levels in early erythroblasts
(104). CA accumulates earlier than hemoglobin in erythroid cells and
at a lower rate. The increase in CA activity continues after the late
polychromatic stage after termination of RNA synthesis so the half-life

of CA mRNA appears to be fairly long.

Quantitative Genetic Variation
There are two mutations in humans that result in the absence of CA

I or CA II in erythrocytes in individuals who are homozygous for the

19
defective gene. A deficiency gene of CA I has been found in a Greek
family (102). Individuals homozygous for the deficiency gene show no
clinical symptoms and have a normal level of CA II in the red cell
(105). The CA I/CA II ratio in the red cell of heterozygous
individuals are half those of normal individuals. Similarly, absence
of CA I in the red cells of pigtail macaques homozygous for a CA I
deficiency gene has no deleterious effect (98,106}

It has not yet been determined if CA I in a CA I-deficient
individual is deficient in the other tissues in which it is normally
present. Cultured lymphocytes from CA I-deficient humans show strong
fluorescence using a fluorescent antibody technique (27). This result
suggests that CA I-deficiency in red cells may be due to CA I protein
or mRNA being labile and rapidly degraded during erythropoiesis.
Degradation may not be as severe in different cells where the kinds and
or levels of proteolytic enzymes or ribonucleases may be different
(107).

Unlike the CA I deficiency, CA II deficiency individuals show an
abnormality of bone, kidney, and brain. Individuals who are homozygous
for a CA II-deficiency gene have the syndrome osteopetrosis with renal
tubular acidosis and cerebral calcification (91). There have been a
total of 11 families that have been diagnosed with this condition.
Osteopetrosis was present in all the homozygous individuals but there
was variation in the severity of renal tubular acidosis and in the
degree of cerebral calcification (108). The CA I/CA II ratios of the

heterozygotes (91,109) were twice that of normal individuals.

20

Interestingly, CA II is found in many other tissues yet only bone,
kidney, and brain are affected in the CA II deficiency. If each CA
isozyme is a product of a single, separate locus one would expect a
defect in the gene to be present in every cell in which the isozyme is
expressed. Furthermore, an organ such as the eye which contains mainly
the CA II isozyme appears to lack CA II in the CA II deficiency yet be
able to function without it (Sly, Whyte and Krupin, unpub. results)
The reasons for the limitation are not known but possible explanations
have been offered: 1) There may be alternate mechanisms that could
substitute for CA II function in other tissues; CA I takes over the
role of CA II in red cells. 2) If mutation caused the protein or mRNA
to be susceptible to degradation, then the CA II turnover levels in a
particular cell would depend on CA II turnover, or on the type of
proteolytic enzymes (107) or ribonucleases in the cell. 3) There may
be two or more CA II or CA II-like isozymes that are coded for by
separate genes that are differentially expressed in different tissues.
(There may be more than one CA II gene in the mouse (110))

The level of CA I is slightly elevated in erythrocytes of a CA II-
deficient individual and the CA III level is doubled (the relative
levels of CA 1, CA II and CA III isozymes in adult human red cells are
85:13zl, respectively)(lll). However, since CA I and CA III have low
specific activities and are thought to be inhibited by chloride ions in
the red cell (112,113), they contribute only 10% of the total CA
activity of the human red cell. Surprisingly, it appears that red
cells of CA II-deficient individuals function normally. There appears

to be a compensatory mechanism for respiratory function of CA under

21

unstressed conditions. (CA activity of homozygous individuals went
from 22% of control level at 25°C to 55% of control level at 37°C)
Chloride inhibition of CA I may be relaxed in the CA II-deficient

individual.

Qualitative Genetic Variation

 

There is considerable genetic variability for CA in mammals. In
'humans there are 25 CA I electrophoretic variants (105). The majority
of these variants do not show any difference in C02 hydrase or specific
esterase activity with the exception of one variant which is an
activity variant rather than a mobility one. There are seven
electrophoretic variants of human CA II (105). The amino acid
substitutions of some of these variants have been determined and the
substitutions as a whole are conservative in their effect in changing
secondary structure. Electrophoretic and quantitative variation occur
quite frequently at the CA locus in the higher apes other than humans
(9), with greater variability occurring at the CA I locus than at the

CA II locus (64).

Hormonal Control

 

A CA III isozyme which is immunochemically indistinguishable from
rat muscle CA III is present in mature rat liver (65) and appears to be
hormonally controlled (116). This isozyme is a major liver enzyme
comprising 8% of cytosol protein (117). The concentration of CA III in
homologous muscle from male and female rats was similar'but the
concentration of CA III in mature male rat liver was 30 times greater

than in mature female rat livers by radioimmunoassay.

22

Sex hormone levels were manipulated in gonadectomized rats and CA
III concentration was monitored by radioimmunoassay. Testosterone
administration resulted in an increase in CA III level in female rats
and estradiol administration resulted in a decrease in CA III levels in
male rats.

When the amount of mRNA in rat liver was quantified by
immunoprecipitation there was nine times more translatable CA III mRNA
present in male rat liver than in female rat liver (116). These
results indicated that the androgen-mediated difference in CA III
expression may be at the level of translatable CA III mRNA.

CA III levels may also vary during rat development. CA III levels
are low in prepuberty rats and increase 7-fold to become 1% of the
total liver protein in adult rat (116). However, different control
mechanisms may operate in mouse and human CA III levels since
radioimmunoassay of human male and female livers show similar levels of

CA III.

Evolution of Carbonic Anhydrases

 

The high degree of evolutionary homology between the structures of
CA I and CA II isozymes indicates that CA I and CA II originated from a
common ancestral gene (27). The greater enzymic, immunological, and
chemical variability among mammalian CA I isozymes and the absence of
CA I in some vertebrate blood suggests that CA II is an evolutionarily
older molecule (118). CA I may have arisen by gene duplication.

It appears that there was an ancestral CA with an active site
similar to that of CA II isozyme. After gene duplication both CA I and

CA II lineages fixed important substitutions in their active sites that

23
are presumably responsible for their different properties. Since the
mammalian radiation, CA I, and to a lesser extent CA III, have
continued to fix substitutions in their active sites. If the molecule
is considered as a whole, it appears that the CA III lineage evolved
rapidly after duplication. Recently, since the mammalian radiation, CA
III has been most conserved, and CA II the least conserved of the three

CA isozymes (33).

THE CARBONIC ANHYDRASE GENE

 

Mouse CA II Gene

 

The first structural work at the nucleic acid level on CA II was
done in mouse with the isolation of mouse CA II cDNA, pMCA II (119).
The cDNA was prepared to size fractionated anemic spleen poly(A)+RNA
and identified by hybridization selection. The predicted amino acid
sequence from the cDNA sequence showed homology to known amino acid
sequences of rabbit and human CA I and CA II. The cDNA clone, pMCA II,
was 1500 bp and contained sequence from the coding region as well as
700 bp of sequence at the 3' non-coding region of the mRNA. To obtain
the 5'-end of the cDNA, a library of mouse cDNA sequences derived from
poly(A+)RNA of anemic spleen was screened using the mouse cDNA (pMCA
II) as a probe (120). Two clones were selected that had more of the
5'-region but lacked sequence at the 3'-end.

The complete amino acid sequence of the mouse CA gene was
determined from the nucleotide sequence data (120). The amino acid
sequence was compared to residues which are invariant in a given

isozyme but differ at this position in the other two isozymes. The

24

mouse CA had one residue in common with the 21 unique and invariant
residues of the CA I isozyme, no residues in common with the 31 unique
residues of CA III, and 16 of the 23 unique residues of CA II. pMCA II
was thus confirmed to code for CA II.

Comparison of the predicted mouse amino acid sequence to that of
human, sheep, ox, rabbit, and horse showed that 33 residues were unique
to mouse (120). Based on the human CA II three-dimensional structure,
all but 5 of the 33 residues are located on the surface of the human CA

II molecule.

Isolation of Mouse CA II Gene

Cosmid and lambda libraries were screened using labeled pMCA II as
the probe and four cosmid clones, A5, A6, 54, 103, and four lambda
clones were isolated and studied (121). Cosmid clone A6 (cosMCAII) was
found to contain a complete mouse CA II gene by comparing hybridizing
bands of the cosmid.with those seen on genomic blots. A6 contains the
16 kb-long gene, 12 kb of 5'-flanking region, and 10 kb of 3'-flanking
region.

The mouse carbonic anhydrase II gene was found to have 7 exons and
6 introns (Figure 2) (121) The three introns A, B, and D interrupt
the codons for Gly-ll, Val-77, and Gly-l43, respectively. Introns C,
E, and F fall between the codons for Glu-116 and Leu-117, Lys-l68 and
Gly-l69, and Gln-220 and Met-221, respectively. The exons range in
size from 76 bp of exon 5 to 799 bp of exon 7. The introns range in
size from 0.3 kb of intron C to 7.2 kb of intron B. There is a TATA
box at position -92 bp and a putative "CAAT" box (CCACT) at position -

140 bp upstream from the ATG initiation codon, similar to the location

25

 

 

Eco R] I I I I I I I I I I I
Hmaxn I I I III I I I
BONIHI I I I n
O 5 IO I5 20 25 30 35 38
(kb)
Figure 2. Restriction endonuclease map and exon positions

of the mouse carbonic anhydrase II gene in cosmid cosMCAII-
2. The solid boxes represent the coding regions and the
open boxes are the untranslated regions of the exons. The

exons are designated by numbers and the introns by capital
letters. (From 121)

26

of consensus sequences common.to eukauyotic genes particularly the
chicken globin genes.

A comparison of the nucleotide sequence of the YBR mouse CA II
gene exons and the cDNA sequence for Balb/c mouse CA II cDNA revealed
strain differences represented by four point mutations (121). Two of
the point mutations were silent. AIthird point mutation was in the 3“-
untranslated region and a fourth at residue 38 resulted in an amino
acid change (Cln/His). This change in amino acid would result in a
change in electrophoretic mobility and may explain the electrophoretic

variation of CA II in many inbred mouse strains.

Rabbit CA I cDNA

 

Low-abundance CA I mRNA from rabbit reticulocyte polyKA)+RNA was
purified by immunoprecipitation of specific nascent polypeptides
attached to rabbit reticulocyte polyribosomes. The purified mRNA was
translated in a cell-free system and the protein product (3H-1abeled)
was fractionated on a polyacrylamide gel to confirm its identity by
molecular weight (122)

The purified mRNA was used to identify homologous clones from
size-selected cDNA libraries prepared from rabbit reticulocytes (122).
The cDNA clones that hybridized to radiolabeled CA I probes but not to
globin-specific probes were further studied and their identity verified
in hybrid selection experiments. Identification was also performed by
hybridizing nick translated CA I cDNA to RNA that had been gel
fractionated, transferred to filters, and identifying the size of the

RNA.

27

The cDNA insert was 300 bp and probably corresponds to the 3'-end
of the mRNA. Therefore, it probably contains a noncoding region, and
definitive proof of its identity could not be obtained by comparing
amino acid sequence to known amino acid sequence. 'The cDNA clone was
used to screen a new rabbit reticulocyte cDNA library in an attempt to
isolate larger cDNA inserts (122).

Comparison of Mouse and
Human CA II Gene

 

 

The mouse CA II gene has been examined at the DNA level (121). A
comparison was made of mouse and human CA II (llQL. Since 81% of the
amino acid residues of mouse and human CA II isozymes are identical at
homologous positions and the three-dimensional structures of human CA I
and CA II are very similar (60% amino acid sequence identity), it was
assumed that the mouse and human CA II tertiary protein structures were
identical (121). Also, mouse and human CA II genes have introns A and
B in identical positions. (Part of the human CA II gene has been
analyzed to date) It was presumed that active site residues of human
CA II occupied corresponding positions in the active site of mouse CA
II.

The human CA I and CA II molecules have a large portion of 6-
structure which runs through the entire molecule (2). One-half of the
active site cavity is formed from hydrophobic side chains. The other
half of the active site is composed of exposed residues which are
mainly hydrophilic.

The hydrophilic region is mostly coded by exons 2 and 3 and the

hydrophobic region is mostly coded by exons 4 and 6. The three

28

histidine residues in the active site region which bind the zinc ion
are encoded by exon 3 and exon 4. It appeared that there was no
correspondence between exons and domain structure for CA (121).

Also, the three dimensional structure of human CA I and II
proteins showed that the splice junction sites for the mouse CA II gene
were located relative to the protein amino acid sequence on both the
outside and inside of the molecule (121). Splice junctions for introns
A, B, E, and F are located on the outside of the molecule and splice
sites for introns C and D are located at the bottom of the active site
cavity. These results are contrary to the idea that intron-exon

junctions map to protein surfaces.

10.

ll.

12.

l3.

14.

15.

29

REFERENCES

Maren, T.B. 1967. Physiol. Rev. 47:595-781.

Lindskog, S., L.E. Henderson, K.K. Kannan, A. Liljas, P.0. Nyman
and B. Strandberg. 1971. pp. 587-665. In The Enzymes. P.D.
Boyer, ed., Academic Press, New York.

Tashian, R.E., C.C. Plato, T.B. Shows. 1963. Science. 140:53-
54.

Tashian, R.E., D.P. Douglas and Y.-S.L. Yu. 1964. Biochem.
Biophys. Res. Commun. 14:256—261.

Schneider, F. and M. Lieflander. 1963. Z. Physiol. Chem.
334:279-282.

Pocker, Y. and N. Watamori. 1973. Biochemistry. 12:2475-2482.
Pocker, Y. and L.J. Guilbert. 1974. Biochemistry. 13:70-78.
Kaiser, E.T. and K.W. Lo. 1969. J. Am. Chem. Soc. 91:4912-4918.

Pocker, Y. and J.E. Meany. 1965. J. Am. Chem. Soc. 87:1809-
1811.

Pocker, Y. and J.E. Meany. 1967. Biochemistry. 6:239-246.

Pocker, Y., J.E. Meany and B.C. Davis. 1974. Biochemistry
13:1411-1416.

Tashian, R.E., D. Hewethmmett and M. Goodman. 1983. pp. 79-
100. I_n Isozymes: Current Topics in Biological and Medical
Research. M.C. Rattazzi, J.G. Scandalios and G.S. Whitt, eds.,
Alan R. Liss, Inc., New York.

Sapirstein, V.S., P. Strocchi and J.M. Gilbert. 1984. Ann. N.Y.
Acad. Sci. 429:481-493.

Whitney, P.L. and T.V. Briggle. 1982. J. Biol. Chem. 257:12056-
12059.

Henry, R.P. and J.N. Cameron. 1983. J. Exp. Biol. 103:205-223.

l6.

17.

18.

19.

20.

21.

22.

23.

24

25.

26.

27.

28.

29.

30.

31.

32.

30

McKinley, D.N. and P.L. Whitney. 1976. Biochem. Biophys. Acta.
445:780-790.

Sanyal, G, N.I. Pessah and T.B. Maren. 1981. Biochem. Biophys.
Acta 657:128-137.

Wistrand, P.J. 1984. Ann. N.Y. Acad. Sci. 429:195-206.

Dodgson, S.J., R.E. Forster, II, B.T. Storey and L. Mela. 1980.
Proc. Natl. Acad. Sci. USA 77:5562-5566.

Vincent, S.H. and D.N. Silverman. 1982. J. Biol. Chem.
257:6850-6855.

Dodgson, S.J., J.M. Kamerling, S. Nioka and R.E. Forster, II.
1984. Fed. Proc. USA 43:300.

Feldstein, J.B., and D.N. Silverman. 1984. Ann. N.Y. Acad. Sci.
429:214-215.

Anderson, B., P.O. Nyman and L. Strid. 1972. Biochem. Biophys.
Res. Commun. 48:670-677.

Carter, N. 1972. Biol. Rev. 47:465-513.

Tashian, R.E., M. Goodman, R.J. Tanis, R.E. Ferrell and W.
Osborne. 1975. pp. 207-223. In Isozymes. C.L. Markert, ed.
Academic Press, New York.

Holmes, R.S. 1977. Eur. J. Biochem. 78:511-520.

Tashian, R.E. 1977. pp. 21-62. I_n Isozymes: Current Topics in
Biological and Medical Research. M.C. Rattazzi, J.G. Scandalios,
and G. S. White, eds. Alan R. Liss, Inc., New York.

Maren, T.H., G.S. Rayburn and M.E. Liddell. 1976. Science.
191:469-472.

Kannan, K.K., B. Notstrand, K. Fridborg, S. Lovgren, A. Ohlsson
and M. Petef. 1975. Proc. Natl. Acad. Sci. U.S.A. 72:51.

Notstrand, B., I. Vaara and K.K. Kannan. 1975. ll! Isozymes.
C.L. Markert, ed. 1:575-599. Academic Press, New York.

Kannan, K.K., A. Liljas, I. Waara, P.C. Bergsten, S. Lovgren, B.
Strandbert, U. Bengtsson, U. Carlbom, K. Fridborg, L. Jarup and
M. Petef. 1971. Cold Spring Harbor Symp. Quant. Biol. 36:221.

Liljas, A., K.K. Kannan, P.C. Bergten, K. Fridborg, B.
Strandberg, U. Carlbom, L. Jarup, S. Lovgren and M. Petef. 1972.
Nature (London) New Biol. 235:131-137.

 

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

31

Hewett-Emmett, D., P.J. Hopkins, R.E. Tashian and J. Czelusniak.
1984. Ann. N.Y. Acad. Sci. 429:338-358.

Lindskog, S.,P. Engberg, C. Forsman, S.A. Ibrahim, B.H. Jonsson,
I. Simonsson and L. Tibell. 1984. Ann. N.Y. Acad. Sci. 429:61-
75.

Lonnerholm, G. 1980. J. Histochem. Cytochem. 28:427-433.
Ridderstrale, Y. 1979. Acta Physiol. Scand. 106:239-240.

Ryan, U.S., P.L. Whitney and J.W. Ryan. 1982. J. Appl.
Physiol. 53:914-919.

Sugai, W. and S. Ito. 1980. J. Histochem. Cytochem. 28:511-
525.

Lonnerholm, G. 1983. Acta Physiol. Scand. 117:273-279.
Ridderstrale, Y. 1976. Acta Physiol. Scand. 98:465-469.

Lonnerholm, G. and Y. Ridderstrale. 1980. Kidney Int. 17:162-
174.

Feldstein, J.B. and D.N. Silverman. 1984. Ann. N.Y. Acad.
Sci. 429:214-215.

Sapirstein, V.S. and M.B. Lees. 1978. J. Neurochem. 31:505-
517.

Wistrand, P.J. 1980. Uppsala J. Med. Sci. 85:75.

Dodgson, S.J., R.E. Forster ,II, B.T. Storey and L. Mela.
1980. Proc. Natl. Acad. Sci. U.S.A. 77:5562-5566.

Storey, B.T., S.J. Dodgson and R.E. Forster, II. 1984.
Ann. N.Y. Acad. Sci. 429:210-211.

Hansson, H.P. 1965. Life Sci. 4:965-968.

Kumpulainen, T. and LR. Korhonen. 1978. Histochemistry
58:183-192.

Kumpulainen, T. 1979. Histochemistry. 62:271-280.

Panero, C., G. Bonfirraro, E.C. Biagioli and L. Burroni. 1974.
Helv. Paediatr. Acta. 29:157-166.

Spicer, S.S., M.A. Sens and R.E. Tashian. 1982. J. Histochem.
Cytochem. 30:864-873.

Kumpulainen, T. 1981. Histochemistry. 72:425-431.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

32

Kumpulainen, T. and P. Jalovaara. 1981. Gastroenterology.
80:796-799.

Jonas, L., L. Hoffman and D. Serfling. 1983. Z. Urol.
Nephrol. 76:311-317.

Kumpulainen, T. and S.H.M. Nystrom. 1981. Brain Res. 220:220-
225.

Kumpulainen, T., D. Dahl, L.K. Korhonen and S.H.M. Nystrom.
1983. J. Histochem. Cytochem. 31:879-886.

Kumpulainen, T. 1980. Acta Ophthalmol. (kbh) 58:397-405.
Kumpulainen, T. 1983. Histochemistry. 77:281-284.
Holmes, R.S. 1976. J. Exp. 2001. 197:289-295.

Koester, M.K., L.M. Pullman and E.A. Nottmann. 1981. Arch.
Biochem. Biophys. 211:632-642.

Register, A.M., M.K. Koester and E.A. Nottmann. 1978. J.
Biol. Chem. 253:4143-4152.

Carter, N., A. Shiels and R.E. Tashian. 1978. Biochem. Soc.
Trans. 6:552-553.

Carter, N.D., S. Jeffrey, A. Shiels, Y. Edwards, T. Tipler, and
D.A. Hopkinson. 1979. Biochem. Genet. 17:837-854.

Tashian, R.E., S.K. Stroup., Y.-S.L. Yu and D. Henricksson.
1978. Fed. Proc. 37:1797.

Carter, N.D., D. Hewett-Emmett, 8. Jeffrey and R.E. Tashian.
1981. FEBS Lett. 12,8:114-118.

Jeffrey, S., Y. Edwards and N. Carter. 1980. Biochem. Genet.
18:843-849.

Spicer, S.S., P.J. Stoward and R.E. Tashian. 1979. J.
Histochem. Cytochem. 27:820-831.

Spicer, S.S., M.A. Sens, R.A. Hennigar and P.J. Stoward. 1984.
Ann. N.Y. Acad. Sci. 429:382-397.

Hennigar, R.A., B.A. Schulte and 5.8. Spicer. 1983. Anat.
Rec. 207:605-614.

Fernley, R.T., R.D. Wright ‘and J.P. Coghlan. 1979. FEBS Lett.
105:299-302.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

33

Roussel, C., J.-P. Delaunoy, J.-L. Nussbaum and P. Mandel.

1979. Brain Res. 160:47-55.

Vaananen, H.K., T. Kumpulainen and L.K. Korhonen. J.

Histochem. Cytochem. 30: 1109 - 1113.

Cammer, W.T., T. Fredman, A.L. Rose and W.T. Norton.

Neurochem. 27:165-171.

1976. J.

Yandrasitz, J.R., S.A. Ernst and L. Salganicoff. 1976. J.

Neurochem. 27:707-715.

Perlman, D. and 11.0. Halvorson. 1981. Cell 25:525-536.

Fernley, R.T., M. Congiu, R.D. Wright and J.P. Coghlan. 1984.

Ann. N.Y. Acad. Sci. 429:212-213.

Schafer, A. and P. Dietsch. 1984. Ann. N.Y. Acad. Sci.

429 : 241 - 242 .

Graham, D., M.L. Reed, B.D. Patterson and D. Hockley.

Ann. N.Y. Acad. Sci. 429:222-237.

Wehinger, H. 1973. Blut 27:172-185.

1984.

Linser, P. and A.A. Moscona. 1984. Ann. N.Y. Acad. Sci.

429 :430-446 .

Benesch, R.N., N.S. Barron and C.A. Mawson. 1944.

153:138.

Nature.

Wilbur, K. and L. Jodrey. 1955. Biol. Bull. Woods Hole Mass.

108:359.

Freeman, J. 1960. Biol. Bull. Woods Hole Mass.
Costlow, J.D. 1959. Physiol. 2001. 32:177.
Goreau, T. 1961. Endeavor. 20:32.

Tuan, R. 1984. Ann. N.Y. Acad. Sci. 429:459-47

Tuan, R. and J. Zrike. 1978. Biochem. J. 176:6

118:412.

2.

7-74.

Crooks, R. and K. Simkiss. 1975. Q. J. Exp. Physiol. 60:55-

63.

Crooks, R., C. Kyriakides and K. Simkiss. 1976
Physiol. 61:265-274.

Gay, C., H. Schraer, R.E. Anderson and H. Cao.
N.Y. Acad. Sci. 429:473-478.

. Q. J. Exp.

1984.

Ann.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100 .

101.

102.

103.

104 .

105.

106.

107.

108.

34

Sly, W.S., D. Hewett-Emmett, M.P. Whyte, Y.-S. Yu and R.E.
Tashian. 1983. Proc. Natl. Acad. Sci. U.S.A. 80:2752-2756.

Sapirstein, V.S., P. Strocchi, M. Wesolowski and J.M. Gilbert.
1983. J. Neurochem. 40:1251-1261.

Funakoshi, S. and H.F. Deutsch. 1971. J. Biol. Chem.
246:1088-1092.

Trachtenberg, M.C. and V.S. Sapirstein. 1980. Neurochem.
Res. 5:573-577.

Sapirstein, V.S., M.C. Trachtenberg, M.B. Lees and 0. Koul.
1978. Adv. Exp. Biol. 100:55-68.

Sapirstein, V.S., M.B. Lees and M.C. Trachtenberg. 1978. J.
Neurochem. 31:283-287.

Cammer, W. and T.R. Zimmerman, Jr. 1983. Develop. Brain Res.
6:21-26.

Tashian, R.E. and N.D. Carter. 1976. pp. 1-56. In Advances
in Human Genetics. H. Harris and K. Hirshorn, eds. Plenum
Press, New York.

Keilin, D. and T. Mann. 1940. Biochem. J. 34:1163.

Rickli, E.E., S.A.S. Ghazanfar, B.H. Gibbons and J.T. Edsall.
1964. J. Biol. Chem. 239:1065.

Edsall, J.T. 1968. Ann. N.Y. Acad. Sci. 151:41-63.
Kendall, A.G. and R.E. Tashian. 1977. Science 197:471-472.

Denton, M.J., N. Spencer and H.R.V. Arnstein. 1975. Biochem.
J. 146:205-211.

Spencer, N. and S. Peller. 1976. Biochem. Soc. Trans. 4:1153-
1155.

Tashian, R.E., A.G. Kendall and N.D. Carter. 1980. Hemoglobin.
4:635-651.

Ferrell, R.E., W.R.A. Osborne and R.E. Tashian. 1981. Proc.
Soc. Exp. Biol. Med. 168:155-158.

Buetler, E. 1983. Proc. Natl. Acad. Sci. U.S.A. 80:3767-3768.
Whyte, M.P., W.A. Murphy, M.D. Fallon, W.S. Sly, S.L.

Teitelbaum, W.M. McAlister and L.V. Avioli. 1980. Am. J. Med.
69:64-74.

109.

110.

111.

112.

113.

114.

115.

116.

117 .

118.

119 .

120.

121.

122.

35

Hewett-Emmett, D. 1982. Fed. Proc. 41:1385.

Venta, P.J., J.C. Montgomery, K. Wiebauer, D. Hewett-Emmett and
R.E. Tashian. 1984. Ann. N.Y. Acad. Sci. 429:309 323.

Carter, N.D., R. Heath, R.J. Welty, D. Hewett-Emmett,
S. Jeffrey, A. Shiels and R.E. Tashian. 1984. Ann.
N.Y. Acad. Sci. 429:284-286.

Wistrand, P.J. 1981. Acta Physiol. Scand. 113:417-426.

Maren, T.H. and E.O. Couto. 1979. Arch. Biochem. Biophys.
196:501-510.

Heath, R., N.D. Carter, D. Hewett-Emmett, E. Fincani,
S. Jeffrey, A. Shiels and R.E. Tashian. 1983. Fed.
Proc. 42:2180.

Hewett-Emmett, D., R.J. Welty and R.E. Tashian. 1983.
Genetics. 105:409-420.

Carter, N.D., R. Heath, R.J. Welty, D. Hewett-Emmett,
S. Jeffrey, A. Shiels and R.E. Tashian. 1984. Ann.
N.Y. Acad. Sci. 429:287-301.

Shiels, A., S. Jeffrey, I.R. Phillips, E.A. Shephard,
C.A. Wilson and N.D. Carter. 1983. Biosci. Rep.
3:475-478.

Tashian, R.E., D. C. Schreffler and T.B. Shows. 1968.
Am. J. Hum. Genet. 17:259-72.

Curtis, P. 1983. J. Biol. Chem. 258:4459-4463.

Curtis, P.J., E. Withers, D. Demuth, R. Watt, P.J.
Venta and R.E. Tashian. 1984. Gene 25:325-332.

Venta, P.J., J.G. Montgomery, D. Hewett-Emmett, K.
Wiebauer and R.E. Tashian. 1985. J. Biol. Chem.
260:12130-12135.

Boyer, S.H., H. Ostrer, K.D. Smith, K.E. Young and
A.N. Noyes. 1984. Ann. N.Y. Acad. Sci. 429:324-331-

CHAPTER 2

ISOLATION OF THE CHICKEN CARBONIC
ANHYDRASE II GENE*

Corinne M. Yoshihara, Mark Federspiel, and
Jerry B. Dodgson

The carbonic anhydrase (CA) gene family displays considerable
variation in its expression pattern. In amniotes (birds, reptiles, and
mammals) the three genetic loci that have been identified that encode
isozymes CA 1, CA II, and CA III, vary in their eXpression between
classes of amniotes and in tissues within a class. ‘The genes for CA I
and CA II are both eXpressed in most mammalian red blood cells (RBC),
for example, while in avian species only the CA II gene is expressed in
the RBC. In mammals, at least, the genes that encode CA I and CA II
enzymes appear to be linked (1—3).

In order to study the relationship between the structure and
organization of CA genes and CA gene eXpression, it is necessary to
first isolate the genes. We describe here the initial steps toward
isolation of the chicken CA II gene.

A chicken RBC cDNA library was prepared in the plasmid pBR322 with
poly-(A)+ chicken anemic red cell cytoplasmic RNA by dG-dC tailing into
the Pst I site (4). Bacterial colonies that were tetRamps were

transferred to nitrocellulose filters, lysed, and the liberated DNA

fixed to the filters. ‘The filters were first screened for those

 

* Reprinted from Annals of the New York Academy of Sciences, Vol. 429,
1984.

36

37

colonies containing recombinant plasmids with globin DNA inserts by
hybridizing the filters with and globin-specific probes. The
nonglobin colonies were selected and screened with 32P-labe led mouse CA
II cDNA (5). The three colonies whose DNA gave a positive
autoradiographic result were isolated. DNA sequence analysis
demonstrated that one of the three clones isolated whose insert was
approximately 300 bp was a _bg_n_a_ Lid; chicken CA II cDNA clone.

The restriction map of the chicken CA II clone is shown in Figure 1

along with the strategy for sequence analysis. The amino acid sequence

 

 

 

 

pPES-O.3 p73._rT TI . fl 9 3 2 I?! r. i. 4;
' a a 2 66 s A 1 I i ’
: 00st
. AIIII .HhsI §PsII w
9 so. I t Hmc II V Is. I
A Faun I Y HIM I ,, 8c- I
. Paul) ll 0 Nos ll — Tag I
- m. II ? he I :sssnz

FIGURE 1. cDNA clone of carbonic anhydrase II. Restriction map of
pPES-O.3. Arrows indicate the direction and extent of the
sequence determined by Maxam and Gilbert sequence analysis.

predicted from the nucleotide sequence shows extensive homology with the

known amino acid sequences of human (65%), rabbit(6) (631), and mouse(7)

(602) CA 11 (Fig. 2). The chicken cDNA clone contains sequence from the

coding region at the 5'-end of the CA 11 mRNA from amino acids 7 to 86.

A comparison of amino acids 7 to 86 of chicken to the corresponding

amino acids of mammalian CA I, II, 111(8) indicates that chicken CA 11

is identical to 1 of the 12 invariant and unique residues of CA I; 1 of

the 15 for CA 111; and 5 of th 9 for CA 11 (residues 7, 26, 66, 68, and

75). Chicken CA 11 is identical with all of the 6 residues that are

located in the active site regions of the CA isozomes (residues 28, 60,

63, 64, 66, and 68).

FIGURE 2 .

38

(II; ACE O‘C AAC GGA CCC (CG CFC ms YPC GM} YAC 'I'IC CC!
10 20
cmcmcmn Gly Ser ms an Cly Pm Ala Hls m m Glu E§ Phe Pro
Hm CA 11 -- Lys -- -- —- - Glu —- -- His Lys Asp - --
RABBIT CA 11 -- Lys -- -- -— -- Glu -- -- His Lys Asp -- --
muss CA 11 Ser Lys -- -- -- -- Glu Asn - H15 Lys Asp - --
A'IC EC MT (136 GAG co: CAB ICC (CC A'IC (CC A'IC Am ADC

30
CHICKEN CA 11 Ile Ala Mn Gly Glu Arg Gin Ser Pro Ile Ala Ile Ser Thr
HLMAN CA 11 -- —- Lys - - -- -- - - Val Asp - Mp --
RABBIT CAII - -- Asp - - - - -- - - Asp -- Asp —-
[DOSE CA 11 — - — -- Mp -- -- -- - Val Mp — Asp --
AAA CDC (II on: TAC GAC CC! (136 CYG AAC (XI CIC AGC 1"C

40
oncmcmn Lys Ala Ala Arg 1y: Asp Pro Ala {>98 Lys Pro Les Ser Phe
Ham CA 11 His 1311' -- Lys - - -- Sex Leu ~- —- -- - Val
RABBIT CA 11 Asp -- -- Lys His - - Se: Igu - - — Arg Val
I'DUSE CA 11 Ala Thr -— His His -— -- -- Len Gln - -- leu Ile
AGC TAC CAT (DC (11‘ AG; ooc MA (11‘ MC GIC AAC AAC SCI;
50 60
CHICKEN CA 11 Ser 'IYr Asp Ala Gly Thr Ala Lys Ala Ile Val Asn Asn Gly
HlMAN CA 11 - —- -- Gln Ala - Se: Ieu Arg - Leu -- -- --
RABBI'I‘ CA 11 -- -- Glu His Pro Ile Ser Arg Arg -- Ile - - -—
PDUSE CA 11 -- -- - Lys Ala Ala Sex - Ser -- — - - --
CAC 'ICC 'I'IC AAC GIC GAG ‘I'IC GAC GPC ICC roc GAC AAIC 'ICA
70

CHICKDI CA 11 His Set Phe Asn Val Glu Phe Asp mp Ser Ser Asp Lys Ser
HLMAN CA 11 - Ala -— -- - - ~— —- -- Gln - Ala
RABBIT CA 11 -- -- -- -- -- -- - -- - -° His - - -
FDUSE CA 11 -- - - - - - -- -- - - Gln -- Asn Ala

GIG CYG CAA (EA CIEA GE GAT (XI AGI‘

80

madman Vallfgcin Gly Gly AlaieuAspGly Ser

Hum CA 11 - Leu Lys -— - Pro - - —- 'mr

RABBIT CA 11 -- Len Lys Glu -- Pro - Glu - Thr

muss can —leuLys--—Pm-5erpsp-

Nucleotide sequence of pPES-0.3. The predicted amino acid
sequence of chicken carbonic anhydrase II is compared with
homologous amino acid sequences of human, rabbit, and mouse
carbonic anhydrase II. The amino acid sequence predicted by
the nucleotide sequence is given below the coding regions
along with its numbering. Y refers to C or T. Only those
amino acids of the human, rabbit, and mouse CA II proteins
that differ from the predicted chicken CA 11 sequence are
shown. See reference 6 and citations therein for the human
CA 11 sequence.

 

39

The chicken cDNA clone was used to isolate phage from a Charon 4A
chicken genomic library (9). The phage that hybridized to the cDNA
clone presumably contain the chicken CA 11 gene. These recombinant
clones are presently being characterized by restriction enzyme analysis.
Future experiments will involve detailed restriction enzyme analysis,
subcloning, and DNA sequence analysis (particularly the 5' flanking
region) of the clones and will provide preliminary data for the study of

the chicken CA gene family.

Ackngwledgments

We are grateful to Dr. Peter J. Curtis for providing the mouse
carbonic anhydrase cDNA clone. We also thank Dr. Richard E. Tashian and

Dr. David Hewett-Emmett for their advice and encouragement.

Refegences

1. Carter, N.D. 1972. Carbonic anhydrase isozymes in Cavia porcellus,
Carvia apera and their hybrids. Comp. Biochem. Physiol. 4313:743-
747.

2. DeSimone, J., 14. Linda & R. E. Tashian. 1973. Evidence for linkage
of carbonic anhydrase isozyme genes in the pig-tailed macaque,
Mgcgca nemesgging. Nature (New Biol.) 242:55-56.

3. Eicher, E. N., R. H. Stern, J.E. Womack, M. T. Davisson, T. H.
Roderick & S. C. Reynolds. 1976. Evolution of mammalian carbonic
anhydrase loci by tandem duplication: Close linkage of Car-1 and
Car-2 to the centromere region of chromosome 3 of the mouse.
Biochem. Genet. 14:651-660.

4. Maniatis, T., E. F. Fritsch & J. Sambrook. 1982. Molecular cloning
manual. Cold Spring Harbor Laboratory. Cold Spring Harbor, N.Y.

5. Curtis, P. J. 1983. Cloning of mouse carbonic anhydrase mRNA and
its induction in mouse erythroleukemia cells. J. Biol. Chem.
258:4459-4463 .

6. Ferrell, R. E., S. K. Stroup, R. J. Tanis & R. E. Tashian. 1978.
Amino acid sequence of rabbit carbonic anhydrase II. Biochim.
Biophys. Acta 533:1-11.

7. Curtis, P. J., E. Withers, D. Demuth, R. Watt, 1’. J. Venta GI R. E.
Tashian. 1984. The nucleotide sequence and derived amino acid
sequence of cDNA coding for mouse carbonic anhydrase 11. Gene. In
press.

8. Tashian, R. E., D. Hewett-Emmett & M. Goodman. 1983. On the
evolution and genetics of carbonic anhydrase 1, II, and III.
Isozymes: Current Topics in Biological and Medical Research 7:79-
100.

9. Dodgson, J., J. Strommer & J. D. Engel. 1979. The organization of
chicken globin genes. Cell 17:879-887.

CHAPTER 3

THE NUCLEOTIDE SEQUENCE OF THE CHICKEN
CARBONIC ANHYDRASE II GENE

INTRODUCTION

 

Carbonic anhydrase (CA) refers to an ancient family of proteins.
Five different CA isozymes (1-9) (possibly more, 10, 11) have so far
been identified that are believed to be coded for by separate genetic
loci. Together, the CA isozymes are extensive in their distribution.
They are found in practically all organisms and in most tissues of any
higher organism. Although the most obvious and well-studied role of CA
is the hydration of C02 in red blood cells (12), other CA functions
have recently been elucidated. A few of these roles include
involvement in ion fluxes in neurons (13), avian eggshell formation
(14), and eye morphogenesis (15)

The CA gene family is quite variable in its expression pattern.
Among the amniotes (birds, reptiles, mammals) CA II is particularly
interesting because its tissue specificity varies with the class of
organism considered. CA II is the only isozyme expressed in avian red
blood cells (RBC) whereas both CA I and CA II isozymes are expressed in
most mammalian RBC. The CA I and CA 11 genes of mammals are also known
to be linked (16).

Although the CA protein has been thoroughly examined, it is only

quite recently that CA structure at the DNA level has been examined.

40

41

DNA sequence data of the mouse CA 11 gene (17) and a rabbit CA I cDNA
clone (18) have been reported. A closer analysis of the mouse CA II
gene showed that it was composed of 7 exons that were stretched over 16
kb of DNA (17). Three of its introns interrupted codons and three
introns fell between codons. There were also slight differences in the
CA 11 gene sequence between two different mouse strains in the form
of point mutations. Since the human CA II gene has been partially
sequenced.(l9) a comparison of human and mouse CA II was made. There
was considerable sequence homology in the 5'-f1anking regions of the
two genes in the presence and locations of common signal sequences.

In this paper we describe the isolation and primary structural
analysis of the chicken CA II gene. The chicken gene provides another
class of amniotes to be represented in the evolutionary analysis of the
CA gene familyx .A detailed structural study of the CA genes as well as
of their relationship to one another is imperative for the
identification of structural regulatory elements controlling the

expression of the CA gene family.

42

EXPERIMENTAL PROCEDURES
Isolation of cDNA

A Agth cDNA library prepared from chicken red cell poly(A)+

mRNA was screened at a 99% representation of the library (41x 104

phage). The plaques were transferred to nitrocellulose filters and
processed as described (20). CA 11 cDNA was identified by
hybridization to both a 5'-end chicken CA II cDNA clone probe (21) and
a full length mouse CA II cDNA clone probe (22). Probes were prepared
by nick translation (23). CA II cDNA fragments were isolated from
positive Agt10 clones by digestion at the EcoRI linker sites and
inserted into pBR325 plasmid DNA. A fine structure restriction map was
derived for the subclones with the largest inserts and used for the

sequence analysis of the cDNA.

Isolation of AcaIII and AcaXVI

 

A ACharon4A chicken genomic DNA library (24) was screened as des-
cribed above. Twenty petri plates with approximately 5 x 104 pfu per
plate were used. Filters were hybridized at 42° for 12 hours to nick
translated 5'-end chicken CA II cDNA clone previously described (21).
Filters were washed at 65° in 0.1 M Na+ and exposed. Duplicate
positive plaques were purified through two platings at low plaque
density and purified recombinants were grown as liquid cultures (25)
for DNA preparation. Restriction maps of the unique CA II-containing
phage, designated AcaIII and AcaXVI, were prepared by standard multiple

restriction digestion analysis.

43

 

 

AcaIII and AcaXVI were digested to completion with the
appropriate restriction endonucleases (BamHI, EcoRI, HindIII, or KpnI).
The fragments of interest were gel purified and ligated to plasmid
pBR322 or pAT153 DNA cleaved with the same enzyme(s). T4 DNA ligase
was added to the mixture and incubated at 15° overnight. E; £913 HBlOl
was transformed with the ligated DNA. Transformants were selected by
differential drug-resistance characteristics (ampicillin resistance and
tetracycline sensitivity), hybridization to CA II probes and/or
restriction digestion of DNA minipreps. Positive transformants were
further analyzed by restriction enzyme digestion and blot hybri-

dization.

Restriction Enzyme Digestion and Southern Blotting

Restriction digests were run according to the recommendations of
the manufacturers. The products of digestion were run on agarose or
acrylamide gels and visualized by ethidium bromide fluorescence. Where
appropriate, the DNA in the gels was transferred to nitrocellulose (26)
and the resultant Southern blots were hybridized to labeled CA II cDNAs

as described (25).

DNA Sequence Analysis

 

Subcloned CA II DNA was restriction enzyme digested, gel
fractionated and the appropriate fragment isolated (27). The chemical
degradation technique of Maxam and Gilbert (28) as modified by Smith
and Calvo (29) was employed for DNA sequencing. Fragments were treated
with calf alkaline phosphatase, 5'-end labeled with 7 -ATP, recut with

the appropriate secondary enzyme and the resultant singly-labeled

44

fragments isolated. Chemical degradation and gel electrophoresis were

as described previously (28,29,30).

Nuclease Protection A End Analysis

 

 

Restriction fragments were isolated from restriction enzyme
digested pBBca-2.8 by cleavage within exon 1 (RsaI site at +19, Figure
2A) and upstream beyond the likely start site (SinI at -l68). The
resultant 191 bp fragment was treated with calf intestine alkaline
phosphatase and labeled with polynucleotide kinase. The end-labeled
fragment (75 ng) was hybridized to 20pg of adult anemic chicken total
red cell cytoplasmic RNA or to yeast tRNA at 60° in 80% formamide
hybridization buffer for 12 hours. Samples were quenched by dilution
into .3 ml of $1 digestion buffer (24). S1 nuclease (at concentrations
of 1000, 2000, and 4000 units/ml) was added and the DNA-RNA hybrid
processed as described (24).

For the 3'-end analysis, a 600 bp fragment was isolated from
pHHca-0.6 digested with HindIII (Figure 2E). The fragment started 388
bp downstream from the stop codon and included 5 possible poly(A)
addition signals. The fragment was labeled by filling the 5'
overlapping end with AMV reverse transcriptase as described (29). The
end-labeled fragment (70ng) was resuspended in hybridization buffer and
hybridized to 20 pg of chicken red cell RNA as above. Hybridization
was performed at 50° for 12 hours. 81 nuclease was added as described

above.

Materials
Restriction enzymes were obtained from International

Biotechnologies, Inc., New England Biolabs, Inc. and Bethesda Research

45

Labs, Inc. Polynucleotide kinase was from Amersham or International
Biotechnologies, Inc. Other materials and bacterial strains were as

previous ly described (24).

RESULTS

Isolation of Larger Chicken 9A 11 cDNA Clones

 

 

A Agth chicken RBC cDNA library (31) was screened separately by
hybridization to a complete mouse CA II cDNA, pMCA II, (22) and a 5'-
end chicken CA II cDNA, pPE5-0.3 (21). The plaques that hybridized to
both probes were purified through two more screenings.

Two of the recombinant clones, 4A and 2B, were shown to contain CA
II cDNA-hybridizing inserts after digestion with EcoRI. Clone 4A
contained an insert of 1.2 kb and clone 2B contained an insert of 0.4
kb. The cDNA insert of clone 4A was subcloned,(pEcD-l.2), and the
restriction map and DNA sequence of the insert were determined. PEcD-
1.2 was found to contain 1.2 kb of sequence that encodes amino acid 7
through to the 3'-untranslated region ending in a run of A:T base pairs
that presumably result from the DNA complement of the 3' poly(A)

portion of the CA 11 mRNA (Figure 3).

Isolation o_f the Chicken C_A 1; Gene

 

A ACharon 4A chicken genomic library (24) was screened with 32P-
labeled CA II cDNA containing the 5'-end of the cDNA (pPE5-0.3, see
chapter 2). Seventeen recombinant clones were isolated which upon
restriction enzyme analysis proved to contain different recombinant
types. Two of these clones, AcaIII and AcaXVI, have been mapped and

characterized in detail. The restriction maps of the two phage are

46

shown in Figure 1.

Initially, a region was found in both clones which hybridized
strongly to pPE5-0.3. Once the more nearly complete chicken cDNA
clone, pEcD-l.2, was available, the phage digests were also hybridized
to a middle and 3'-end probe as well. AcaXVI showed hybridizing bands
with all three probes. AcaIII hybridized only to the 5'-end and middle
probes. Both mapping and hybridization studies suggested that the
chicken CA II coding region was interrupted by several introns,
that most of the coding region was contained in AcaXVI, and that the
5'-coding region was contained in AcaIII. Since the two clones showed
some common restriction fragments, it appeared that they overlapped.
Figure 1 summarizes the results of the mapping and hybridization data.
The two clones share bands when hybridized to the 5'-end and whole
cDNA probes. However, AcaIII had additional hybrdizing bands to the
5'-end probe and AcaXVI had additional hybridizing bands to the whole
and 3'-end probes. AcaIII had no hybridizing bands to the 3'-end
probe.

The chicken DNA inserts are approximately 16 kb in length.

AcaXVI contained 4 regions that hybridized to chicken CA II cDNA. The
first hybridizing region was a 1.5 kb KpnI, BamHI fragment at one end
of the insert that hybridized strongly to the 5'-end probe. The next
hybridizing region was a 2.8 kb BamHI, HindIII fragment approximately
10 kb from the first hybridizing region. A 1.63 kb HindIII fragment
adjacent to the 2.8 kb fragment was the third hybridizing region and
adjacent to this was a final, very weakly hybridizing HindIII

fragment.

Figure l.

47

Restriction map of the chicken CA II gene locus.

(A) Restriction map of chromosomal DNA contained within
clones AcaIII and AcaXVI. (B) Restriction map of clone
AcaIII which contains exons 1-4. (G) Restriction map of
clone AcaXVI which contains exons 2-7. The solid boxes
represent the coding regions and the open boxes represent
the untranslated regions of the exons. The numbers above
the boxes indicate the exons and the horizontal lines
below the boxes represent the subclones. The arrows above
the line indicate the direction and extent of DNA sequence
determination.

48

 

 

A. 1K!)

‘2 s4 5 s 7

HM J I U
T-
B. Acam
0" an

I—-—I

pBBca-2.8
C.Acam

IAITW—L—II—i—é—Eaee

i—-'1 .—_‘
mega—1,5 asses-3.3 "Iip Hos-0.6
pHHca-1.85

4‘ BamHI
(l, EcoRI
V W m

0 Km 1
,1} EcoRI LINKER

=7 CMA
_|_ CA 1!

£101

49

The other clone, AcaIII, had a 2.8 kb BamHI fragment which
hybridized strongly to the 5'-end probe and a second.L72 kb BamHI,
EcoRI fragment which hybridized to the complete cDNA. In between these
two regions was a 3.5 kb BamHI, EcoRI region which hybridized weakly to
both probes. The 3.5 kb BamHI, EcoRI fragment was also found in
AcaXVI.

The mouse CA II gene was known to be interrupted by 6 introns of
which the second intron is very large (7.2 kb) (17,19). The fragment
that hybridized strongly to pPE5-053‘was subcloned and sequence
analysis showed that it contained exon 2 (Figure 2). This then
allowed us to estimate the location of other exon regions which were
later confirmed and refined by DNA sequence analysis (Figures 2,3)

The restriction map and hybridization data obtained for the two
recombinants were used to develop a subcloning strategy. The subclones
generated, their fine structure restriction maps, and the
sequencing strategy used in their analysis are given in Figure 2.

Since both chicken cDNA clones did not contain the entire exon 1
sequence (see chapter 1), a mouse genomic clone was used to locate exon
1. The mouse clone A6-2.7 (19), a generous gift of P. Venta and R.
Tashian, Univu of Michigan, was prepared-such that the only coding
region it contained was exon 1. The chicken exon 1 was approximately
positioned by cross-hybridization to A6-2{7 and definitively located

by DNA sequencing.

DNA Sequencing Strategy of the Chicken CA 11 Gene

 

 

The fine structure maps of each subclone were used to develop a
sequencing strategy for each of the CA II exons. The direction and

extent of the region sequenced are indicated by the arrows in Figures 1

Fi

Figure 2.

50

Restriction maps of the subcloned DNA fragments of the
chicken CA II gene. Partial restriction maps of

(A) subclone pBBca-2.8 which contains exons 1 and 2,
(B) subclone pKBca-1.5 which overlaps with pBBca-2.8
and also contains exon 2, (C) subclone pBHca-3.3 which
contains exons 3, 4, 5, and a small portion of exon 6,
(D) subclone pHHca—l.85 which contains the greater
portion of exon 6, all of exon 7, and a portion of the
3'-untranslated region, (E) subclone pHHca—0.6 which
contains the remainder of the 3'-untranslated region.
The filled boxes represent the exons which are iden-
tified by numbers. The arrows above the boxes show
the direction and extent of DNA sequence anlaysis.

51

A. pBBca-2.8

<———-I
I———>
I-——>

€———I
|——-)

 

“‘1. .2
3.3% Q‘WE‘? $46 3

 

J,

B. pKBca-1.5

rag—J ‘7 as

C. pBHca-3.3
I—s

 

(-—-'._)(——-‘ I-9

3.5%? 5?? V??? Y
J 7 8

D. pHHca-1.85
6L) I—)

 

 

(——-I

i (-——I

A? gas
’1‘ BamHI

s1: EcoRI

V th m

A Kpnl

A BstNI

V HIMI

0 Rsa I

{7 SacI

1? SlnI

‘ Xma I
EPBRCZZIPAT‘HS
_-_°<7-1.7
.ElCAII

HUT

52

and 2. The sequence of the coding exons has, in some cases, been
determined on both the coding and noncoding strands of DNA. Except for
that sequence 5' to codon 8, all the coding sequences were also
sequenced on one or both strands in the cDNA clone, pEcD-l.2.

Once the sequence data was obtained the identification of the
hybridizing regions as to exon number could be determined.
Furthermore, the exact boundaries of all seven exons could be
identified by comparison.of the cDNA and genomic sequence. .As Figure
2 indicates, subclone pBHca2.8 contains most of exon 3, all of exons 4
and 5, and a small portion of exon 6. Exon 6 is continued on
subclone pHHca-1.63 which also contains exon 7 and the 3'untranslated
region. Since the 3Knuxanslated region is relatively long, it is
continued on the 057 kb subclone. Exons 2 and 3 are also on AcaIII.
Exon 1 as well as the 5'f1anking region is contained in the 2.8
kb BamHI subclone of AcaIII.

Exon 3 is interrupted by a BamHI site used in subcloning and the
2.8 kb BamHI, HindIII fragment containing the 5'end of exon 3 was not
itself subcloned. Therefore, this portion of exon 3 was sequenced
directly from a 3.5 kb BamHI, EcoRI fragment isolated from the AcaIII

clone.

The DNA sequence of the chicken CA II gene is given in Figure 3.
The genomic sequence shown encodes 259 amino acid residues and
approximately 1.5 kb of untranslated region. The coding sequence of
the gene was identified by comparison to the chicken CA II cDNA

( pEcD-lJD sequence. The amino acid sequence predicted from the

Figure 3.

53

DNA sequence of the chicken CA 11 gene. The CCAAT, ATA,
and AATAAA or AACAAA signal sequences as well as the ini-
tiation and stop codons are underlined. Numbers above the
sequence indicate nucleotide numbering from the Cap site.
The upper case letters indicate those sequences that are
transcribed into mRNA and the lower case letters indicate
those sequences that are processed. All the sequences
present in exons within the left and right arrows above
the sequence shown were also sequenced in the cDNA clone,
pEcD-1.2. Asterisks indicate sites where the cDNA and
genomic clone sequences differ (see text). R refers to
A or G. X refers to A, G, T, or C.

54

-200 - - ' I ‘ I
EXON l CGCCGCCGCCGCTCTCCCGGCCC

GGCCCGACGCAGCTCCGCGGCGGGAGGATCGCGGGTTATAAGCGGACCTCTCTCTCTCC
GCCCCCGAOCGAAGTCIEECTCCGCCCCCGCCCGCGCTCCQQAQQCCTICCTCCGGCCO
CGGAGAAGGGCATGGAGTTCGCGGGAGCCiAIAAAAGCCCCTGACAGCCCGCCGAGGCC
+4CGGCGTTGCGATAGCCGACGGAGCCGGGCCGGCGCACCAIQTCCCATCACTGGGGGTA

. ~—4> .
CGACAGCCACAACGGTGAGTGTGGGGCACGGCGG

EXON 2 CGCCCCGCGCTCTCTTGCAG

GACCCGCGCACTGGCACGAGCACTIggCCATCGCCAATGGGGAGCGCCAGTCGCCCATC
GCCATCAGCACCAAAGCCGCCCGCTACGACCCCGCGCIGAAGCCCCTCAGCTTCAGCIA
CGATGC688CACGGCCAAAGCCATCGTCAACAACGGGCACTCCTTCAACGTGGAGTTTG
ACGACTCCTCCGACAAGTCAGGTGAGcGCATCCGTGTGTGC

EXON 3 GGAGGCTCTTTGCTTTGCAG

. . 300 . . .
TGCTGCAAGGAGGAGCGCTGGATGGAGTCTACAGGTTGGTGCAGTTTCACATTCACTGG
GGATCCTGTGAGGGCCAGGGCTCTGAGCACACTGTGGATGGCGTGAAGTACGATGCAGA

GGTATGATGTGCTTTGCCTTT

EXON 4 CCATGTTTTCTTATTCCTAG
uoo . . . .

CTTCATATTGTTCACTGGAATGTAAAATATGGCAAATTTGCTGAAGCTCTGAAGCATCC

lGATGGTTTGéCCGTCGTAGCCATCTTCATGAAGGTTAGTCAAACTTCTTTTTC

55

EXON 5 g TCATGCTATATGTGTTACAG
I 00 I I I I
GTAGGGAATGCCAAACCTGAAATACAGAAAGTTGTTGATGCTCTGAACTCCATTCAAAC

CAAGGTAATATTTTGTGTTGAATG

EXON 6 GCCTACCTTCCTTACTGCAG
GGGAAACAAGCTTCTTTCACAAACTTTGACCCTACTGGACTGCIGCCTCCAIEEAGAGA
CTATIGGACGTACCCTGGCTCCCTGACTACTCCACCACTGCATGAATGTGTGATITGGC
ATGTTCTGAAGGAGCCCATCACTGTCAGCTCTGngAGGTAGCTCTCTGGGGTAGTGC

EXON 7 GTTTTTGCCTTTTCCCACAG
ATGTGCAAACTCCGTGGCCTTTGCTTCAGTGCTGAGAATGAGCCGGTGTGCCGCATGST
GGACAACTGGCGCCCATGCCAGCCTCTAAAGAGCAGgggAGTCAGAGCITCCTTCCAGi
AACCTCAGCéATGAGTGTGITAGAAACTGCTGTGTTTGCGAGGAACCCTITTTGCTAAG
CACAAATCAAACCTTTGCCRQGTGTGCCCTGGCAACATCTIGTCTCCCATATTATTCTC
lTCTTTCGCTCTéCATCTAAAATGCCAGCTAATGAAATGTGAAAGGCTCTTGGCCAAACA
lGGAAGGGGTTCTTCATGTGTGGAGCTGGGGAAAACCTGAGGGéGGCTGTGTGCATTTTG
ATGACITACTGCGACTGACATTTIGGAAAAAAcAAAAACAAACAAACAAACAAAAACTA
TATTéGCTGTTTGGéAGAGCATATGGTGAGAGCAAAATAAGCCAITCTGAGAAACCTCA
TcKAcIGGT6TGATAéTACACAilggTAATGATAACTTGAAGCTTiGTGAAAAGcACAG
GAAAAQAAAATATAGGIATAGIXIKGGAAAAAAAAAAGTATATTGAGAAGGAAAAGIAA

1300 . . .
AATGAATACTGAGAATTAGACTTAGATATTAAG AAACTGACATGTAAATATTTTGACTT

56

1400
AGATATTAAGAAACTGACATGTAAATATTTTGAAGACCATTTTGCATTTTTACCCATGC

TATCAACAACCTTGATCGTCTTCATGGTAGTGTGTTGGTTTTTTTAATTAAAAATGTCC

. 1500
TAAATTAAGTATTATCTTTTGAAGTTATTACTGTAGAGGGTACXXXAAATATCTTTTCA

CTGAAATAATATATGCTCTAATTTAAGGGGGAAATAAAATTGTATTTTAGATAACTTCT

1600
CTAATAAATCTATACTTATATTTTTGCTTCCAAGTTGTTGTTTAATTCAGAAAAATAGT

1700
TACTAATCTGTTCACATCATTGCATACTAAACATTATTAATAAACATTTAGTATATAAG

GAGTGAGGTTTCAAIGACACTGGAGATTAATGCAAAATGATATGGATGTATTAGAAGAT
1800
TTCCTAAGTCGTTCTCTGCTAGACAGCAGAAGGAAGTATTAGTTCTGAGCAACCCTTGC

AAATTGTCAATGAAGTTGATTGCATAATAAAAGATCCCCTTGCAGGAATTTTGAACARC

57

nucleotide sequence is identical to that of the cDNA clone. However,
there are 5 confirmed differences in the cDNA and genomic sequence
(Figure 3). These include the third nucleotide of codon 140 (C in the
genomic and T in the cDNA) and codon 152 (A in the genomic and G in the
cDNA). There are also 3 changes in the 3'untranslated region: at
nucleotide 1001 a G in the cDNA is an A in the genomic DNA; at
nucleotide 1177 a C in the cDNA is an A in the genomic DNA; and at
nucleotide 1113 the cDNA contains an extra A that is not present in the
genomic clone. These changes presumably reflect genetic diversity in
the chickens used to prepare the cDNA and genomic libraries (both were
White Leghorn; the cDNA sequence from a Michigan chicken and the
genomic sequence from a Pennsylvania chicken). The extra A:T base pair
in the cDNA clone with respect to the genomic clone comes in a long
stretch of A:T base pairs interrupted by a few C:G base pairs. This
mutation may have occurred due to slippage during DNA replication. It
cannot be determined at present whether the changes seen are actually
differences in the germ line of the chickens used or whether any or all
of the changes could have occurred during the cloning procedures.
Similar changes were seen by Venta et al.(19) between YBR and BALB/c
mouse strains.

The chicken CA II amino acid sequence has 65% homology to mouse CA
II sequence (17) (Figure 4). There are 69 base changes that result in
silent substitutions (Table I). The greatest number of silent changes
occur in exon 2. There are 164 base substitutions that result in an
amino acid change and a majority of these occur in exons 2 and.7. The
active site residues as well as the unique and invariant residues

(Table II)(32) are fairly well conserved. There are only 2 changes

Figure 4.

58

Amino acid sequence comparison of chicken and mouse CA II
genes. The predicted amino acid sequence of chicken CA II
is compared with homologous amino acid sequences of mouse
CA II. The amino acid sequence predicted by the nucleo-
tide sequence is given above the coding regions along with
its numbering. Only those nucleotides in the mouse se—
quence that differ from that of the chicken sequence are
given along with the resulting amino acid change, if any.
Asterisks indicate those amino acid residues that are
located in the active site region of the CA II protein.

CCAII
MCAII

CCAII
MCAII

CCAII
MCAII

CCAII
MCAII

CCAII
MCAII

CCAII
MCAII

CCAII
MCAII

CCAII
MCAII

CCAII
MCAII

Met
ATG

His
CAC

Ile
ATC

Phe
TTC

A A
Ile

Ser
TCC

80
Gly

GGA

Trp
TGG

Lys
AAG

Lys

G C
Asp

Ser
TCC

Glu
GAG

Lys

Ser
AGC

GA
Asp

Ser
AGC

TCT

Phe
TTC

Gly
GGA

Gly
GGA

Tyr
TAC

*
Phe
TTT

His
CAT

His
CAC

Asp

Thr
ACC

50
Tyr

TAC

Asn
AAC

Ala
GCG

C C
Pro

Ser
TCC

Asp
CAT

Ala

130
Ala

GCT

GA
Gly

His
CAC

Phe
TTC

Lys

GC
Ala

Asp
CAT

Val
GTG

Leu
CTG

Cys
TGT

Ser

Ala
GCA

Glu
GAA

Lys

Trp
T66

20
Pro

CCC

Ala
GCC

A T
Thr

Ala
GCC

Lys

*
Glu
GAG

Asp
GAT

AG
Ser

100
Glu

GAG
Asp

Glu
GAG

Ala
GCT

Gly
GGG

Ile
ATC

Ala
GCC

Gly
GGC

CT
Ala

Phe
TTT

Gly
GGA

AC
Asp

Gly
GGC

Leu
CTT

Leu
CTG

Val

Tyr
TAC

Ala
GCC

Arg
CGC

AG
Gln

Thr
ACG

Ala

70
Asp

GAC

Val
GTC

TC
Ser

Gln
CAG

His
CAT

Lys
AAG

Gln

59

Asp
GAC

AG
Ser

Asn
AAT

Tyr
TAC
C T
His

Ala
GCC

Ser

Asp
GAC

Tyr
TAC

Gly
GGC

Ile
ATT

T G
Leu

His
CAT

A

Gln

Ser
AGC

AG
Lys

Gly
GGG

40
Asp

GAC

Lys

Ser
TCC

Arg
AGG

Ser
TCT

*120

Val
GTT

Pro
CCT

G

His
CAC

Glu
GAG

Asp

Pro
CCC

Ala
GCC

AG
Ser

Ser
TCC

CAG
Gln

Leu
TTG

Glu
GAG

His
CAC

Asp
CAT

A18

AAC

Arg
CGC

Ala
GCG

Ile
ATC

Asp
GAC

*90
Val

GTG

A C
Ile

His
CAC

Trp
TGG

Gly
GGT

A

Gly
GGA

Gln
CAG

Leu
CTG

Val
GTC

Lys
AAG

Asn

Gln
CAG

Thr
ACT

Asn
AAT

Leu
TTG

Pro
CCC

Ser
TCC

Lys
AAG

Gln

*60
Asn

AAC

Ser
TCA

Ala

Phe
TTT

Val
GTG

Val
GTA

ACC
Thr

140
Ala

GCC
T

Ala
GCG

Glu

Pro
CCC

Pro
CCC

Asn
AAC

Val
GTG

*
His
CAC

Asp
GAT

A C
Asn

Lys

Val
GTC

T

His
CAC

Asn

30
Ile

ATC

G G
Val

Leu
CTC

Gly
GGG

Leu
CTG

Ile
ATT

Phe

110
Gly

GGC

Lys

Tyr
TAT

Val
GTA

T G
Leu

Trp
TGG

Ala
GCC

Asp

Ser
AGC

CT
Leu

*
His
CAC

Gln
CAA

Lys

His
CAC

Val
GTG

Lys

Gly
GGC

Gly
GGC

60

150
Ile Phe Met Lys Val Gly Asn Ala Lys Pro Glu Ile Gln Lys Val Val

CCAII ATC TTC ATG AAG GTA GGC AAT GCC AAA CCT GAA ATA CAC AAA GTT GTT

MCAII TAT T T A T A CC TC AA CC C T C C
Tyr Leu Ile Pro Ser Gln Gly Leu Leu

160 170
Asp Ala Leu Asn Ser Ile Gln Thr Lys Gly Lys Gln Ala Ser Phe Thr

CCAII GAT GCT CTG AAC TCC ATT CAA ACC AAG GGG AAA CAA GCT TCT TTC ACA

MCAII A A C T A A G GT G G C T G T
Glu His Lys Arg Ala Ala

180 *190
Asn Phe Asp Pro Thr Gly Leu Leu Pro Pro Cys Arg Asp Tyr Trp Thr

CCAII AAC TTT GAC CCT ACT GGA CTG CTG CCT CCA TGC AGA GAC TAT TGG ACG

MCAII C T TCC TCC T T GG AA TTG C A
Cys Ser Gly Asn Leu
* * * * *200 * * *

Tyr Pro Gly Ser Leu Thr Thr Pro Pro Leu His Glu Cys Val Ile Trp
CCAII TAC CCT GGC TCC CTG ACT ACT CCA CCA CTG CAT GAA TGT GTG ATT TGG

MCAII T C G T TG CC
Leu Thr

* 210 220
His Val Leu Lys Glu Pro Ile Thr Val Ser Ser Glu Gln Met Cys Lys

CCAII CAT GTT CTG AAG GAG CCC ATC ACT GTC AGC TCT GAG GAG ATG TGC AAA

MCAII ATC G C G T AGC CT C T
Ile Arg Ser His

230
Leu Arg Gly Leu Cys Phe Ser Ala Glu Asn Glu Pro Val Cys Arg Met

CCAII CTC CGT GGC CTT TGC TTC AGT GCT GAG AAT GAG CCG GTG TGC CGC ATG

MCAII T ACG G AA A AC GCG T G T AA GAA GCG
Phe Thr Asn Asn Glu Gly Asp Ala Glu Glu Ala
240 250

Val Asp Asn Trp Arg Pro Cys Gln Pro Leu Lys Ser Arg Glu Val Arg
CCAII GTG GAC AAC TGG CGC CCA TGC CAG CCT CTA AAG AGC AGG GAA GTC AGA

MCAII T GCT G AT A A G A A
Ala Asn Lys Ile Lys

Ala Ser Phe Gln Stop
CCAII GCT TCC TTC CAG TAA

MCAII G T A
Lys

 

61

pm as bpp mm o pm s

 

mm mm amp mm or pm o
mm as me Fm 2 pp m
mm mm mm Fm w an :
zm mm opp m.mm 0 mm m
cm mm mop om am pm N
mp m— Pm m.op N 2 P
monomao>au oocownm>av on no mosvnmon vaom acoausuwunpsn omcmso owom coxm
ooauooaosc caom ocas< u Lonasz ocasm no onauooaosc ucoaan oaasa am ca was
u Hmaoa Lopes: Hooch no amass: luasmon momcmso

an no Lonasz

 

.OGOM HH <0 0m503 80L.“ Oflwm HH 40 C030fl£0 .HO OOGOMLO>HU UHOQ OGHEN find QUHuomHOn—Z .H SQ¢H

62

.Nm mesmemmmu ..Hm um uuwaEMquoamm scum vmma>mu we maawh

 

.mmsvemme ouemlopwuow moumowvaw «

.H <0 no woman waeemnasc moamavmm

.HHH <0 no HH <0 aw ucomoea uoc mmsvwmou mmumowuau I

 

 

 

 

 

 

 

 

 

0H0 um2 am< umm 022 020 :00 He <0 :0
sun um< umm :00 uee <0 <2
use g< He <0 <2
use e <0 <2
mmu sen mmu NmN 2mm 0mm 0mm
020 =u0 um> mum wee Has use :02 0u< ouu eu0 :u0 umm :00 :20 cue mH< mam ouu uu <0 20
am< :00 :02 um» use 022 0u< 020 :20 men use =u0 eu0 su0 Hue <0 <2
umm ouu He <0 <2
cu0 mum uom :00 u <0 <2
mmu uNN uwu cum 00m emu 0mm 002 002 e02 mwu meu 002 emu emu mmu mmu esu ans
020 :00 meg uu0 u0> um> mue um> use :20 uu> am< umm uom am< auu am< umm He <0 20
000 20< u :00 wu< _oum 0u< um2 uee use uo> 0u< use Hue <0 <2
nm< 2H0 I pay mug 9H0 cm< HH <0 <2
umm mu< um< am< au0 e <0 <2
emu mmH 0m2 emu 0N2 muu ~22 me 20 e0 00 we we em me He 00 e0 mm
as an i
20< mu< mu< 220 mu< meg umm mue mu< 0u< :u0 mum =u0 mu< am< mum mum . He <0 20
use uu0 use :00 :20 =u0 am< me< =u0 I Hue <0 <2
umm 0u< am< mum mu: 1 He <0 <2
use =u0 umm meg am< am< am< mu< u <0 <2
N0 0m 0m <0 mm on em mm mm AN 0m. 02 02 su 0 s m u
_ .mEhNome HH <0 aoxofinu
one :« mmswemmu msomoaoaos 0:0 0:0 moshnome <0 amHHmEEme mnu uow mmsvamme unmwem>afi 0am waves: .HH mqm<a

63

that involve active site residues and these occur at residue 90
and residue 202. .Residue 90 which is an Ile in mouse changes to Val in
chicken but remains nonpolar. Residue 202, however, is changed from a

nonpolar Leu in mouse to a basic residue, His, in chicken.

Chicken CA I; Gene Intron/Exon Organization

 

The chicken CA II gene is interrupted by 6 introns. Introns 1 and
2 interrupt the codons for Gly-11 and Val-77, respectively. Introns 3,
4, S, and 6 fall between the codons for Glu-116/Leu-117, Lys-147/Val-
148, Lys-168/Gly-169, and Gln-220/Met-221, respectively (Figures 2 and
3). The location of 5 of the 6 introns relative to the amino acid
sequence are conserved between the chicken and mouse CA II genes.
Surprisingly, the location of one of the introns, intron 4, is
different in the chicken gene, falling between codons 147 and 148
rather than within codon 143 as in the mouse» The chicken intron 4
location, however, is also observed in the mouse CA 2 gene (ll)cL (L
Montgomery; personal communication)

The 5' and 3' boundaries of the six introns of the chicken CA II
gene all fit the consensus donor and acceptor sequences (33) seen for
most eucaryotic introns (Table III). This fit to the consensus
sequence holds true for the donor and acceptor sites for intron 4, the
intron whose position is altered in the chicken gene relative to the
mouse CA II gene.

The sizes of the 6 chicken CA II introns are given in Table IV,
in comparison to the corresponding intron sizes in the mouse gene. It
can be seen that some of the intron sizes are similar in both genes as,

for example, are the intron sizes in the corresponding chicken and

64

TABLE III. DNA sequence of intron donor and acceptor sites.
sequences of intron donor and acceptor sites in the

chicken CA II gene are given along with the intron

donor and acceptor consensus sequences.

The

 

 

Intron Donor Acceptor
l ACG/GTGAGT CGCG’CT'CTCTTGCAG/ G
2 CAG/GTGAGC CTCTTTGCTTTGCAG/T
3 GAG/GTATGA TTTTCTTATTCCTAG/ C
4 AAG/GTTAGT CTATATGTGTTACAG/ G
5 AAG/GTAAT CCTTCCTTACTGCAG/G
6 CAG/ GTAG CT TGCCTTTTCCCACAG/ A
co~se~sus gAG/eTéAGT <E),, NgAG/G

 

65

TABLE IV. Intron size comparison of the chicken and mouse CA II genes.

 

Intron Size (kb)

 

Intron Chicken Mouse
1 0.36 1.0
2 10.2 7.2
3 0.45 0.3
4 1.75 1.2
5 0.9 1.1

6 1.21 1.8

 

66

mammalian globin genes (34). However, other introns vary significantly
in size between the two CA II genes. Intron 1 of chicken is 0.36 kb
which is approximately one third the size of mouse intron 1 (17).
Intron 2, on the other hand, is about 3 kb larger than the
corresponding intron in mouse. The remaining 4 introns of chicken
differ from the corresponding introns of mouse anywhere from 051 -(18
kb but the difference is not as striking as in the first two introns.
The portions of the chicken CA II gene introns that have so far
been sequenced have a distinctive pyrimidine rich pattern (Appendix).
These long stretches of'CPS and TVs are apparently not characteristic

of mouse CA II introns (35, P. Venta personal communication)

5_' 329 2; Ease 9.: the Chicken 9.9: II mass

 

The 5' end of the CA II mRNA has been determined by nuclease
protection experiments (Figure 3). An end-labeled fragment extending
from the RsaI site in exon 1 (codons 5 and 6) 5' to the SinI site at -
167 (Figure 2) was hybridized to total chicken reticulocyte mRNA,
digested with $1 nuclease and run on an 8% sequencing gel" The major
protected fragment is about 58 bases in size which places the mRNA
start site (Cap site) in the CCACG sequence about 39 bp upstream from
the ATG initiation codon'(Figure 3). (When the fragment is run next to
a MaxamsGilbert sequencing ladder of the SI probe fragment, the major
protected band corresponds to initiation at the A in the CCACG,
results not shown).

The 3' end of the CA II mRNA can be positioned by the cDNA clone
(pEcD-112) sequence which indicates that polyadenylation of the CA II
message occurs at the A at +1258 about 16 bp downstream of the sequence

AAAACAAAA. 3' protection experiments (Figure 5) tend to confirm this

Figure 5.

67

5' and analysis of chicken CA II cytoplasmic mRNA.
Protection of CA II gene fragment from 81 nuclease
digestion by anemic adult red cell total cytoplasmic
RNA is shown. End-labeled restriction fragments were
hybridized to yeast RNA (lanes B, D, F) and red cell
RNA (lanes C, E, G) as described (Experimental proce-
dures) and then subjected to varying levels of $1
nuclease digestion. Sl-treated fragments and marker
DNA fragments are shown after electrophoresis on an
8% sequencing gel (9). The 5' end protection frag-
ment was generated by hybridization of a 191 bp RsaI/
SinI fragment to RNA followed by treatment with

1000 (lanes B, C), 2000 (lanes D, E), and 4000 (lanes
F, G) units/ml Sl nuclease. Markers (lanes A, H)

are polynucleotide kinase labeled restriction frag-
ments of PBR322 cleaved with Hian.

68

AIODIFIH

880-
”! -

7‘-

 

69

polyadenylation site, although the protected band is not distinct
probably because of a string of 10 contiguous A:T bp downstream from
the putative poly(A) addition site. The string of A:T bp could
hybridize to the 3' poly(A) tail of the mRNA and exhibit partial Sl
resistance. The AAAACAAAA sequence differs slightly from the AATAAA
consensus polyadenylation signal (36). There are several AATAAA
sequences in the 0.6 kb of DNA 3' to the putative 3' end of the mRNA.
Although no bands were observed in the 3' protection experiment
corresponding to the use of poly(A) sites near these AATAAA signals, we
cannot rule out the possibility that one or more of them are used for
at least some CA II mRNAs. Since this region of the genome

contains several lengthy regions of 100% A:T base pairs which are known
to be relatively Sl sensitive, there might be cleavage within a bona
fide mRNAzDNA hybrid that extended to these more distal putative
poly(A) sites, and thus these ends were not observed by $1 analysis.
However, it is noteworthy that only one band was seen on a Northern
blot of gel-fractionated red cell poly(A) RNA hybridized to a chicken

CA II cDNA clone (M. Federspiel personal communication)

_5_'_ and .3_' Flanking Sequences o_f the Chicken % I_I Gene

 

 

A 783 bp region upstream from the CA II coding sequence has been
sequenced. Putative signal sequences that are common to eucaryotic
genes transcribed by RNA Polymerase II are found in the chicken gene
(Figure 6). A Goldberg-Hogness (37) block "ATA" is located 23-30 bp
upstream from the Cap site. Although a sequence that corresponds
accurately to the consensus sequence "CCAAT" (38) cannot be identified,

a region approximately 70-75 bp upstream from Cap (at -74 in Figure 3)

70

.m<zma

HH <0 m:u ou vmwvm we A<V>Hoa £0H£3 um musfion manmnoea mumawfimmv
msoee< .cocoo coaumceaeou man on m>wumame £3050 mew oamw HH <0
:mxoeno was 00:08 050 mo mmuwm was mamawem cowuma%cmpmmaom A00
.mwcuu 002000 22 emumawumme mum Aeuv 0000e00<0e0z00

msmcmmcoo mcu ou wcwccoammeuoo mmocmvuom vmummaou may .mmocwnvmm
use soamn uo m>onm mwcea unwemuum he evacuees“ mum Amuv 000000 was
000000 mmocosamm meannen Ham use .moxon H<<00 Hausa ou m>wumawe
coaufimoa eemnu nue3 wcoam aSOnm one mwcmw HH <0 mnu mo mmxoa <H< 0:0
00 .m mxuoan soeel00 Amv .caosm mwoamsumm wzu cowsumn movauomaosa
wceomam mo emnasc msu mumuapaa memnasz .ssonm mum aovou coeumeuqce
mnu 0:0 .Amuv mmuum m<0 .2020 mmxop <e< .emouoaoe Amuv e<<00 uo
0coemmm .ANHV mmcmw HH <0 cuss: was 00:08 ca mwosmsvmm umafiaem ou
vmummaoo mum ocmw HH <0 amxoeso 0:0 ou .m mmocmsvom vm>emmn00 A<v
.mcmw HH <0 onu wcexcmam mmocmsvmm Hmawwm o>Huwusn mo mocwsvmm <20

.0 ounwam

1
7

m0e0eeme-o2-<<<<e<<-u..-<<e
Ou<00«e-22-<<<<0<<-.u.-<<e

-02-00000L00000uuu000000nm2:000000

<<<<e<e mNuu000000u02ue0<00 Nu 0<00H00<00000H000H00¢0HOHuuns0-00000000000

<<<<H<e 0N 00000000 0 00<00 e 000000000000 ~N< 0000e00<00<<< one 00<0<00<00<00<

0e<-ueu0<0<0uuNu<<<<e<e-0su00<00
0e<uem-0<0<0-uuu<<<<e<eum<-e0<00
0e<-0m-00<00-uu-<<<<e<eum<-00<00

00:02

2020220
.0

z<201
00002
2000210
.0
z<2=x
00:02

2000210
.<

72

that has limited homology (CCACG) to the CCAAT site can be found. The
Goldberg-Hogness box and the CCAAT box usually begin about 30 and 75 bp
upstream of the Cap site, respectively. The absence of good consensus
CCAAT sequences has also been noted for the adult chicken a-globin
genes (39).

The ATA and CCAAT sequences are common to the mouse and human CA
genes that have been sequenced. The ATA box for mouse and human are
located at -99 and -86 bp from the ATG initiation codon, respectively.
The CCAAT boxes for both are located 42 bp upstream from the ATA boxes.

When chicken CCAAT and ATA signal sequences are compared to that
of human and mouse, the most obvious difference is the distance of the
signal sequences from the initiation codon (Figure 7). Chicken ATA is
located at -61 whereas mouse and human ATA boxes are located at -86 and
-99. Although the mRNA initiation sites of mammalian CA II genes have
not been mapped (see below), this presumably reflects a difference in
the length of the 5' untranslated portion of the CA II mRNA. The
chicken Cap site is found at 23 - 30 bp upstream of the transcription
initiation site. ‘The actual Cap site of the mouse and human.CA II
genes have not been experimentally determined but an alternating PyPu
stretch (CACAC) at -70 bp from the initiation site in both mouse and
human CA II genes is likely to be the Cap site (P. Venta personal
communication).

Despite the difference in length and sequence of the 5'
untranslated sequences of the chicken, the CA II gene shows a
significant resemblance to the analogous mouse and human regions (35)
This is shown in Figure 7. The region of greatest homology extends

from the Cap site upstream for about 80 hp with approximately 60%

Figure 7.

73

Comparison of the 5' flanking region of the chicken, mouse,
and human CA II genes. The upper line gives the sequence
of the chicken CA II gene for 120 bp upstream of the CAP
site. The first mRNA nucleotide is numbered as +1 in this
figure. The second line gives those nucleotides of the
mouse gene that differ from the chicken gene and the third
line gives those nucleotides of the human gene which dif-
fer from the chicken gene. Dashes indicate a deletion in
one line relative to the others. The putative TATA and
CCAAT sequences are indicated by a line over the chicken
sequence.

CCAII
MCAII
HCAII

CCAII

MCAII
HCAII

74

-120 -100
CCAII CCGCCCCCGAGCGAAGTCTCCCTCCGCCCCCGCC
MCAII A CT GTCC C CC CA GGT T T C T
HCAII A CT G CC TC CC C -------- T C T

 

-80 -60
CG-CGC-TCCCCACCC---CTTCCTCC----GGCCGCGGAGAAGGGCAT
T---T AGGT T --GG c CCTG cc--- A G
TTCGCTAGGT GAG cc ccce cc--cc A c

-40 +1
GGAGTTCGCGGGAGCCTATAAAAGCCCCTGACAGCCCGCCGAGGCCACG
CA G GGACGGT AC C -A A

A G GGGCCGGC GAC C A A

 

75

overall homology. The region from 45 to 22 bp upstream of the mRNA
start site (which includes the TATAAA sequence 31 to 26 bp upstream
from Cap) is over 95% homologous between the chicken and mouse
sequences. The homology in this 80 bp region between the CA II genes
exceeds that seen between analogous chicken and mammalian.globin gene
promoter regions (34,39,40,4l), suggesting that these sequences may
have an important role in the regulation of CA II gene expression.

McKnight and Kingsbury (42) identified GC rich sequences in the
thymidine kinase gene of Herpes simplex virus that appear to be
involved in efficient transcription, possibly by acting as binding
sites for the Spl transcription factor (43). The consensus Spl
transcription factor binding site is 5' GGGCGG 3' or its complement 5'
CCGCCC 3'. Two such sequences exist 7 bp 5' to the CCAAT box in the
chicken CA II gene (Figure 6B). The mouse and human CA II genes
contain several such sequences both 5' to CCAAT and between.CCAAT and
ATA (Figure 68). While no exact match to the consensus sequence is
seen between CCAAT and ATA in chicken CA.II, this gene does contain a
partially homologous GC-rich sequence, GGCCGCGG, in the appropriate
position which may function in the same manner.

An imperfect tandem repeat of 14 -15 bp that is thought to
function as an upstream promoter element and is found in 5 mammalian
and 1 avian fl-globin genes and in several rat pancreatic genes (44) is
found in the human and mouse CA II genes as "CCNGTCACCTCCGC" (35). In
the fl-globin genes this tandem repeat is found 9 -25 bp upstream from
the CCAAT sequence in mammals and 55 bp upstream in chickens. In the
human and mouse CA II gene these repeat elements have been found 15 and

22 bp upstream, respectively, from the CCAAT boxes (Figure 6B) There

76

are similar repeat elements found in the chicken CA II gene:
CAAAGCACCTCCCC and AGGACCACCACAGC. However, these elements are located
very far upstream of the CA II promoter at -427 and -572 in the chicken
CA II gene, and they are not closely linked to each other. Furthermore
the first of these two elements is a rather poor match to the
consensus. The function of these elements, if any, remains unknown and
whether the homologues near the chicken CA II gene are functional or
merely coincidental is also unclear.

The chicken CA II gene 5' flanking region is very GC-rich as has
also been noted for the mouse and human CA 11 genes (35). Also, in the
case of the chicken gene, the DNA strand which acts as the mRNA
template is very purine-rich in the region from -200 to +1 making the
opposite strand (the strand shown in Figure 3) pyrimidine-rich.

Lengthy polypyrimidine stretches appear in several of the sequenced
intron regions of the chicken CA 11 gene as well. This does not appear
to be the case for the analogous regions of the two mammalian genes.
The significance of these trends, if any, is unknown.

The probable major poly(A) addition sites for the mouse and
chicken CA II genes are shown in Figure 6C. Note that both genes
contain a rather long 3"untranslated region. The unusual
polyadenylation signal seen in the chicken CA II gene is discussed

above in connection with its identification.

DISCUSSION

 

The chicken CA 11 gene has been isolated and its primary structure
has been determined. Sequence data for the coding region, as well as

for the 5'-f1anking, 3'-f1anking, and adjacent regions of the introns

77

is presented. The sequence of a nearly complete CA II cDNA clone has
also been determined, allowing us to position intron sequences and the
3' end of the CA II mRNA.

There are 30 amino acid residues that are postulated to occur in
the active site regions of CA isozymes (1). ‘When chicken CA II amino
acid residues are compared to the homologous active site residues of
mammals most of the chicken residues are found to be conserved. The 16
residues that are invariant in all three isozymes across the species
are also invariant in chicken. .A similar conservation is seen in the
two active site residues (Asn at residue 66 and Glu at residue 68) that
are unique and invariant to CA 11 (Refer to Table II)(32); chicken has
Asn and Glu at the appropriate positions. (Chicken CA II possesses 9
of the 15 residues in the molecule as a whole that are unique and
invariant to CA II, 1 of the 18 residues that are unique and invariant
to CA I, and 8 of the 39 residues that are unique and invariant to CA
III). At position 90 of the active site region the residue that is
present in most mammals is Ile except for ox which has Val. Chicken is
similar to ox in that it too has Val at this position. 'Val at position
90 is just a slight divergence since both residues are nonpolar.
Chicken does differ from mammals at active site residue 202. Mammalian
CA IIs that have been sequenced all have Leu at this postion whereas in
chicken the residue is His. 'The amino acid change is from a nonpolar
to a basic residue in this case.

When the chicken CA II amino acid sequence is compared to those of
mouse and human there is greater than 60% homology to both mammalian
sequences. Chicken CA II appears to be more homologous to human CA II

than to mouse CA II. There is 65% sequence homology of the chicken

78

sequence to the mouse sequence and 70% homology of the chicken to the
human sequence.

In comparing the exon/intron organization between the chicken and
mouse CA 11 genes, the most surprising result is the different
locations of intron 4 relative to the respective coding sequences. In
almost all cases studied to date, intron positions within coding
regions are conserved between.homologous mammalian and avian genes.
Even in genes such as the a- and fi-actin genes where intron positions
are known to vary considerably between different species, the rat and
chicken genes retain identical intron locations (45). The change in
intron position between the two CA II genes shifts this intron 4 2/3
codons toward.the 33 end in the chicken relative to the mouse CA.II
gene (Figure 8). This "new" intron position has also very recently
been found in the human CA 2 (mitochondrial) (11) gene (J11
Montgomery, personal communication). The similarity of the chicken CA
II intron position to that of the mouse CA 2 intron suggests that the
chicken CA II gene structure is the more ancient form compared to the
mouse CA II gene. Given the similarities observed in the rest of the
two CA II genes, it appears that the intron shift occurred via a small
number of mutational events as opposed to the possibility that the two
genes are the product of two lines of CA II gene evolution that have
been separate for longer than most mammalian/avian homologues. The
mutational shift may have occurred after the divergence of birds and
mammals. Figure 8 shows that the intron boundary sequence at the intron
4 acceptor site retains a considerable level of homology to the
corresponding sequence of the mouse gene which in mouse is used to code

for amino acids 144 -l47. It seems likely that a mutation in the

79

.mocmw 3000 fig :oewme

:oeuae 0:0 00 mowemvcson mau mumoevafi mmnmmam .meswem mnu mo mam: eusoa 030 no
:Bo:m me 0:00 HH <0 :mxoeno 0:0 mo :oewmu weavaoammeeoo use .Amev q :oeuaa vaaoem
meow HH <0 canoe 050 no mucosvmm weom ocean 0cm mveuomausa msu 030nm seawam msu no
new: Heme: 05H .mmsmm HH <0 @0908 van amxoanu ca coaquOH 0 :oeucw mo comaewnao0

00H 00H eae
>00 0<> 0>0 e02 010 000 >00 0<> 0<>

000 <H0\0<0 <ee 0H0 e<e <e0 ... e0< He0\0<< 0h< 0ee 0e< 000 <h0 0H0

<00 ee< 0<< 0e» bee e<e 00\0 <00 H00 00H ... .0e eke ee< e0\0 0he ee0

>00 000 0>0 000 0:0 m>e >0 0 000 0<>
wee 00H and

.0 seamen

zmxuuxu

00:0:

80

mammalian evolutionary line resulted in a shift to previously cryptic
splice donor and acceptor sites. For example, if the ancestral gene to
the mouse CA II gene had the chicken arrangement, and the donor site at
codon 147 was partially inactivated, a cryptic donor site around 14 bp
upstream might have become active followed by a similar shift of the
acceptor to the present intron 4 acceptor site in mouse. Note that the
codon 147 - 148 junction in the mouse gene shows a good fit to the
consensus intron acceptor sequence even though it is apparently not
used in 2139. Of course, it is impossible to say exactly what the last
common precursor to the mammalian and avian CA II genes looked like or
what order of events might have led to the observed intron shift.

Other than the difference in position of intron 4, the
organization of the chicken and mouse CA 11 genes is quite similar.

All other introns have identical locations within the coding sequence,
both genes have relatively long 3'tumxenslated regions, and except for
a few nucleotide differences in the donor/acceptor sites, the junctions
are fairly well conserved. There are some differences in the size of
the homologous introns, but the effect of these changes is likely to be
minimal. Likewise, we have yet to understand the significance of the
long CT stretches that appear to be characteristic of chicken CA II
introns and 5' flanking region, or of the GC-rich character of the 5'
flanking DNA.

Consensus sequences commonly found in eucaryotic promoter regions
transcribed by RNA Polymerase II are present in the chicken CA II gene.
These sequences have been described in detail in the Results section.
Note that all these sequences appear to be components of a variety of

eucaryotic promoters. We have yet to identify any sequences that might

81

confer tissue-specificity in or near the CA II gene promoter.
Preliminary results have proven, however, that the chicken CA II mRNA
level is greater than IOO-fold induced in anemic reticulocytes over
that seen in liver and other adult tissues. Much of this induction
occurs late in erythroid differentiation since the HD3 erythroid
precursor cell line (transformed with avian erythroblastosis virus) has
about 10 times as much CA II mRNA as does liver but 10-50 fold less
than in terminally differentiated reticulocytes (J. Dodgson, personal

communication).

9.

10.

ll.

12.

13.

14.
15.

16.

82

REFERENCES

Tashian, R.E., D. Hewett-Emmett and M. Goodman. 1983. pp. 79-100.
_I; Isozymes: Current Topics in Biological and Medical Research.
M.C. Rattazzi, J.G. Scandalios and G.S. Whitt, eds. Alan R. Liss,
Inc., New York.

Sapirstein, V.S., P. Strocchi and J.M. Gilbert. 1984. Ann. N.Y.
Acad. SCio 429:481-4930

Whitney, P.L. and T.V. Brigg le. 1982. J. Biol. Chem. 257:12056-
12059.

Henry, R.P. and J.N. Cameron. 1983. Exp. Biol. 103:205-223.

McKinley, D.N. and P.L. Whitney. 1976. Biochim. Biophys. Acta
445:780-790.

Sanyal, C., N.I. Pessah and T.B. Maren. 1981. Biochim. Biophys.
Acta 657:128-137.

Wistrand, P.J. 1984. Ann. N.Y. Acad. Sci. 429:195-206.

Dodgson, S.J., R.E. Forster, II, B.T. Storey and L. Mela. 1980.
Proc. Natl. Acad. Sci. USA 77:5562-5566.

Vincent, S.H. and D.N. Silverman. 1982. J. Biol. Chem. 257:6850-
6855.

Feldstein, J.B. and D.N. Silverman. 1984. Ann. N.Y. Acad. Sci.
429:214—215.

Montgomery, J.G., P.J. Venta and R.E. Tashian. Isozyme Bull. 14
(in press).

Maren, T.B. 1967. Physiol. Rev. 47:595-781.

Cammer, W.T., T. Fredman, A.L. Rose and W.T. Norton. 1976. J.
Neurochem. 27:165-171.

Benesch, R.N., N.S. Barron and C.A. Mawson. 1944. Nature 153:138.
Linser, P. and A.A. Moscona. 1984. N.Y. Acad. Sci. 429:430-446.
Tashian, R.E. 1977. pp. 21-62. _Ig Isozymes: Current Topics in

Biological and Medical Research. M.C. Rattazzi, J.G. Scandalios
and G.S. White, eds. Alan R. Liss, Inc., New York.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.
29.

30.

31.

32.

33.

34.

35.

36.

37.

83

Venta, P.J., J.C. Montgomery, D. Hewett-Emmett, K. Wiebauer and
R.E. Tashian. 1985. J. Biol. Chem. 260:12130-12135.

Boyer, S.H., B. Ostrer, R.D. Smith, R.E. Young and A.N. Noyes.
1984. Ann. N.Y. Acad. Sci. 429:324-331.

Venta, P.J., J.C. Montgomery, R. Wiebauer, D. Hewett-Emmett and
R.E. Tashian. 1984. Ann. N.Y. Acad. Sci. 429:309-323.

Benton, W.D. and R.W. Davis. 1977. Science 196:180-2.

Yoshihara, C.M., M. Federspiel and J.B. Dodgson. 1984. Ann. N.Y.
Acad. SCI... 429:332‘3340

Curtis, P., J.B. Withers, D. Demuth, R. Watt, P.J. Venta and R.E.
Tashian. 1983. Gene 25:325-332.

Rigby, P.W.J., M. Diekmann, C. Rhodes, and P. Berg. 1977. J. Mol.
Biol. 113:237.

Dodgson, J.B., J. Strommer and J.D. Enge 1. 1979. Cell 17:879-887.

Maniatis, T., E.F. Fritsch and J. Sambrook. 1982. Molecular
Cloning. Cold Spring Harbor Laboratory, New York.

Southern, E. 1975. J. Mol. Biol. 98:503.

Girvitz, B.C., S. Bacchetti, A.J. Rainbow and P.L. Graham. 1980.
Anal. Biochem. 106:492.

Maxam, A.M. and W. Gilbert. 1980. Methods Enzymol. 65:499-560.
Smith, D.R. and J.M. Calvo. 1980. Nucleic Acids Res. 8:2255-2274.

Simoncsits, A. and I. Torok. 1982. Nucleic Acids Res. 10:7959-
7964.

Yamamoto, M., N.S. Yew, M. Federspiel, J.B. Dodgson, N. Hayashi and
J.D. Engel. Proc. Natl. Acad. Sci. USA 82:3702-3706.

Hewett-Emmett, D., P.J. Hopkins and R.E. Tashian. 1984. Ann. N.Y.

Mount, S.H. 1982. Nucleic Acids Res. 10:459-472.

Dodgson, J.B., 8.J. Stadt, O.-R. Choi, M. Dolan, R.D. Fischer and
J.D. Engel. 1983. J. Biol. Chem. 258:12685-12692.

Venta, P.J., J.C. Montgomery, D. Hewett-Emmett and R.E. Tashian.
1985. Biochim. Biophys. Acta 826:195-201.

Proudfoot, N.J. and C.C. Brownlee. 1976. Nature 263:211-214.

Goldberg, M. 1979. Ph.D. Thesis, Stanford University.

38.

39.
40.

41.

42.
43.

44.

45.

84
Efstratiadis, A., J.W. Posakony, T. Maniatis, R.M. Lawn, C.
O'Connell, R.A. Spritz, J.K. DeRiel, B.C. Forget, S.M. Weissman,
J.L. Slightom, A.E. Blechl and O. Smithies. 1980. Cell 21:653-
668.
Dodgson, J.B. and J.D. Engel. 1983. J. Biol. Chem. 258:4623-4629.

Dolan, M., Dodgson, J.B. and J.D. Engel. 1983. J. Biol. Chem.
258:3983-3990.

Engel, J.D., D.J. Rusling, R.C. McCune and J.B. Dodgson. 1983.
Proc. Natl. Acad. Sci. USA 80:1392-1396.

McKnight, S.L. and Ringsbury, R. 1982. Science 217:316-324.
Gidoni, D., W.S. Dynan and R. Tijan. 1984. Nature 312:409-413.

Dierks, P., A. van Ooyen, M.D. Cochran, C. Dobkin, J. Reiser and C.
Weismann. 1983. Cell 32:695-706.

Fornwa 1d, J.A., G. Runcio, I. Peng and C.P. Ordahl. 1982. Nucleic
Acids Res. 10:3861-3876.

APPENDIX

Figure 9.

85

APPENDIX

Nucleotide sequence of 5'—flanking and intron regions

of the chicken CA II gene. (A) This stretch of sequence
is a continuation of the sequence presented in Figure 3
(5'-f1anking region and exon 1). The sequence given
here extends the 5'-f1anking region from —200 bp upstream
to -770 bp upstream of the CAP site. (B - G) Partial
sequence from introns 1 - 6 is presented. Exons mark
the start of the intron sequence and the intron-exon
junction is indicated by a slash line. The numbers in
the intron sequence refer to the approximate distances
that separate sequenced portions of the intron reading
from a 5' to 3' direction. Refer to the arrows above the
subcloned DNA fragments in Figure 2 for the sequencing
strategy.

86

APPENDIX

5’ FLANKING ngCCTCTGGCAGCGCATCCTCACCTTCTCCCATGGCTACTGCT
AGTAGCTGATCCTACAAATATTAGTCEQgGTGCTCTGGTCTGACCAAGGACCACCACAG
CACAGCTGGGGGCCGGTCCCAGTTTCCATGCACATATTCCCACTGTTTAATTACAGTTG
TTGGAGCAERgTACATAAATAAATAAATAAATAAATCCGAAATGGTGTCTGGAAACCAT
GCAGGCGATTTCTGGTGGCAAAGCACCTCCCCGGGCTCAGAGGGTCCCAgggCTCCCAC
AGCCAGCTGCTCTGCGCTCTGGCCAAGGGCATCGCAAGAGCCCGAGGGGAGTTGGGGAA
CCCGGGGGANTCCCYTTAACCAGCTCTYCAGTTTGACTCCTGCAGGCGACCCCGCACCC
GCGGCATTYTCACACACATTYYCGCGCTTXCCTGGCTGCTCGGTGYCGCTGCTYTCGCT
GCACGCCCGCGGYCAgCGCGGCGCGAGAAGCAGGAGCCGTCCCGGGGCCGGCATGGGTG

I I I | ' -200
CGGGCAGGGCCGGGCCGGGCACTAAGTGTCTCTGACGCGCGGGGCCGCCCCGCTGCC

INTRON l EXOXCG/GTGAGTGTGGGGCACGGCGGGACGCTTGCCTGCCGGCGCA
AXXXCCCGGG-IN}CGGGGGCGCTTCGGTTTTGCTTCTGTAACYXTGAGATGYCACCTT
TTGAGTGGTGCGCCGAGCATCCTYTGCGTTCCGGCACGCGGGGATGGTGCACAGGGTTG
AACCCAGCCCGGAGGAACCCGGCTGGCGTTCCCCTCACCTCGGGGACTGCCCCGCCTGC
TCCCCGAGCGGGTTTTTCCCTCAGCCGTACAGGTGCCGCGGACGTCACGACCGCAGAGC

- EXON 2
GGTACCTCTCTGACCCCCGAGCCCTGAGCGCCCCGCGCYCYCRRRRRAG/GACC

87

INTRON 2 CAG/GTGAGCGCATCCGTGTGTGCGTGCGGCCAGAGCGCGACGCTGT
CGTCAATGCTGCTATTAATGCTAAGCATCTACTGGAGCCGCAGCGTTTCTTTTTTATTA
AAAATACATATATGTTGTAAAACTGAGCTCCTCTTAGAAGAACGCTATGGCTACCTGGG
CTGCCCTACTCTAAACTCAGTAGAAGTGAACTCTACCTTTGCAGTTAATTAAGGAGGAT
TTGGGCATGGGCTTGGATGGGAAGGAGGAAGAAGGGAAAAAAAAGATGGATTATGAGAA
CAGTTATAGGTATAGCTCATAGTAAGCTGCTGGAACAAACGGTTAACGAGGAGGCAAAA
GGCACCACCRRGRGAAACGGGCTGTAGAATAACAGGRGATTACAGTTACCAACATGGAG
CAGAAATAGTTTXGAAATAGTTTCTTTGGTTACCAAAATATTATTTCATATTTCXGAAG
CCAGTATGCTCCCA6GTTAGTACAGAAACAGAGAAATGATTAGGAAATACCTCTCTGTT
CTCCTGTAACTTAAACAAAAATCCCTGACGTTTCTTGACXGAAGGCTTTCTAGATAAAG
ACAGTGTATACCTGGCTCATTCCCACAAAGTTTGCAGGCAGGCAGCTGATCCGCCGTGT
CCCATGGCTCACTGTGCTGTGCAGTGTGGTTCAGCACCGTGTCCTGCAAACCCCGCAGT
AACCCTGCTGGAGGGAAYGGCCCCATCGGCGGGAAAAGGGANTC-200-GCGCGATTAA
CTTGTGXGGCTAGTGCCCTGTTTTGTXGGGCGCTTCCTTAGCGAGTGTCACTTAGXAAT
GGTGCTGAAACGAAACGAAACCGATGTCTACCGAAAAAAATGTGGGGTTTTTTTTGCGG
ATAATACTATATGAAAGGTATGTCAGACAGCACTGCACTGGGAATTTGTATTTCACTTG
TGATCTTTATACXRACAGCGTTAGTGGGTCAGCTAAGTGCTTCACAGCTGCATAGATAC
ACATAAGCCTATGTATAATTTATTCXGAACTTCTGTATTCATATCTGTATAGCTTTAAA
AAACGGAAAAGAAGGGAAAAAGGTCTTCTGAGTTCTTAGCATTTATTGCTCXGAAGCAC
XGAGGAGCCGGAGGXGAAGTGCATGTGTTTTTAGCAGTGTTAGGAGGAGTAGATCCCAA

AGAGTATAAAGAGAGRGGGATCC- 5 5 0 0'TCCCTCTGTCTCTGGAAGCAATCTTATAGT

88

TCAGCCATTTTGATTACTTCTTTGACAAGTTTATTGCTTCAGTGAGATTTATTAGTCTC
CTGCCCTGTTACATCCTTACACAGGGAGCTATGGTAATTGCACAGTATTCCCGCCTAAC
TTAGGGCATGTGACTGCTGCGACTTCCTGAAGAAAAATACAAAGTAGTCCACAGAATAT
CCTGTAAAGATGCCCTGTTCTGCATGGGAAAAGTAACTGCCCAGTGACAGCCTTTGGGA
ACAGCAGTGAGGTTGTGTGGCCTGAGTGAGCAGCTGTCCCGGGTGCAGCACAGTGCCCC
TTGGTTTCTG-TOGO-CTTTATTGCAGCAGCGTGCCTTXGAGAAAGCCTGGAGTAGCGTG
GCAGCATGCAGAGCAGTAAACTAAXXXCACTTGGGAGACTTGAGTGGATGTTCCAGTCT
GAAGTGAGACTTGATAGCAAATGTTCCAGAGGTGCTAATGGAGGCTCTTTGCTTTGCAG

EXON 3
/TGC

INTRON 3 ExogAg/GTATGATGTGCTTTGCCTTTCCAGTATTCTTACCTAAGAGTAT
TTTGATTTTCACCTTCCACTCTGGTAGTAATAGCTCTACAAACCTTCACGGGGCAAGAG
ACTGTGGGGGTTACCATGGAAGATTTTGGTGCAGTGCCATATTAGGATXXAGYCCCTTG
TAAATTATAGANTC-loo-GGCAAAACACCATCTGCTCTCAATGTCCAACAAATTACTT
CCCTGCATCCTCATTTGCAAAATATTATGGATXGAATATAATTTTCACAGTGCTCCATG

EXON 4
TTTTCTTATTCCTAG/CTT

EXON 4
INTRON 4 AAG/GTTAGTCAAACTTCTTTTTCTATTATTTGTGTTAGTGTAG-
1250-TXTXTXXGATTAGCCTAAXTTAACAACTATTTTGATTTYXCTGCTATTGTTAAA
TGTCCAXGAGCATTTCXTATCGGCCAAATTGTCTTGCAGCTTCTTGAGGAAGGTGATTG

TAACCCAACTGTGTACACTCTGXC'13 0 -AAACGGACGGCTCACATCCAAAAATAARCC

89

CACATGCTGTAAACXGAAATAATTTTATTCTGACCAGTGTCCTCAGCCGACCTATTTGT
GCTGTCAACTGGCTXGATTTTTAGRGAGACAGTTGCAGTCACTTCTGCTTAATATTCCT
GTTATGTAAXGAAATAATGCTCXGAAAATGTGCTACAAAATGTGTCCTAGAGAATTGTA
ACAGAAAACTAAGGTGTCGAGAAACTGTGCCATTAAGTGCTTTTCATGCTATATGTGTT

EXON 5
ACAG/GTA

INTRON S EXORAg/GTAATATTTTGTGTXGAATGCCACCAGACACAAXCRYGTGCCC
GAXXGGGAGATTCTCTATCTTGCATGTCCACTGTGTGTATGAGAGTAAAACCATATCGG
CACAAATGCTCCTGTAATCGTTAAAGCCAAAGATGTTCAAAGTAAGTTAGGGTAGAGGG
CAGGTAATTTGTATCTGTGACTGCATGTATTTGTCTTTCAGCTGCATTGAAACCTGTAA
GGTAAGGATAATCTTCCAGACAAGAAGTTTGATCAGGATAAAAACTGGGGAAAAAATTA
ACCTTTTGTATTTTAAATACTGTATAAGGCATATATGGGATCCTAAC-375-CTGTATT
CCCRXTGTATACTGTGTGXGAAATCTCATTATTCCTTCXXGTAGGATCAAGTCATATCT
GTAAXGAAAAGCCCATAAATTCAACATTXGAAATGCXGAAATGTCTTGACTGTCCTAAAA
AAAATAAXGAATATAGTGCAGACGTACATAATGCCTGTACTGTAAGAGGTACTAACCCT
GTATGCCCAACCCAGTTTCTCCCTTCCTCCAAATATAGAGCTTCTGTAGAAGACTTCAC
TGTGACGAGCCCAGTATTTGTATGTTGTCACTTCCTTAACAATAGTCTGCCTACCTTCC

EXON 6
TTACTGCAG/GGG

EXON 6
INTRON 6 CAG/GTAGCTCTCTGGGGTAGTGCCAATTGTATTGATTGCTTTAATT

TAGT- 70-AAACGGCTTGCCTTTTTCTGGCAAGCACTGCTTCTTTATGTGGAAGTAAAG

90

GTTGATGTAAAATGAAAGGTTGTCCTCCATCACAAA-17-ATTAATGAAATTACAAAGA
GGCAAGATTTTATXXTGAGGTTTGCAXCTATGTGCT-750—AGCCTCATGGATCGGGAT
GATGATACATAATTCAGATAGTCCTGCTGCAATGAGTATGCATTTGAGGATGTTTTTGC

EXON 7
CTTTTCCCACAG/ATG