LIBRARY
menu” State

University J

a

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

NW'
DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

rfi“

 

 

 

 

 

 

 

 

 

‘-

5. n

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmetive Action/Equal Opportunity Institution —!
cmmnt

. a»? _ __ —.-_____ 7 v-

 

THE DEVELOPMENTAL EXPRESSION OF CD45 GLYCOPROTEIN IN MURINE T LYMPHOCYTES

BY

Hsun-Lang Chang

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Microbiology and Public Health

1990

ABSTRACT
THE DEVELOPMENTAL EXPRESSION OF CD45 GLYCOPROTEIN IN MURINE T LYMPHOCYTES
BY

HSUN- LANG CHANG

The CD45 family of lymphoid and myeloid cell surface glycoproteins
exhibits molecular weight heterogeneity arising in part from alternate
exon use among exons 4, 5, and 6. I developed a new method combining
reverse transcription and polymerase chain reaction (RT-PCR) using only
3,000-30,000 cells to study the exact alternate exon use in the murine
lymphoid cells. Based on the results from size measurement, Southern
hybridization, and direct sequencing of the gel-purified fragment, I
concluded that exon 7 and exon 8 could also be alternately used. My data
indicated that, under the experimental conditions, smaller isoforms of
CD45 cDNA were amplified more efficiently than larger ones at large number
of PCR cycles. However, each individual cDNA isoform, regardless of the
length, was amplified at approximately the same rate at least before the
end of the 25th cycle. Therefore, RT-PCR has been limited to 24 cycles in
examining the CD45 isoform expression in murine T cells. My results
showed that Stage I thymocytes (CD4'CD8') expressed only trace amounts of
minus-one [Ex(-l)] and minus-two exon [Ex(-2)] isoforms, with no other
isoforms detected. A Ex(-2) isoform was also detected in all thymocytes
and '1' cells analyzed but only in trace amounts. Stage II thymocytes
(CD4*CD8*) expressed high and approximately equal amounts of Ex(-l) and
zero alternate exon [Ex(0)] isoforms, with minor quantities of one exon
[Ex(l)] and two exon [Ex(2)] isoforms. Among Stage III thymocytes, both

CD4*CD8' and CD4’CD8+ cells expressed significant quantities of only the

Ex(—l) and Ex(0) isoforms. Comparison of CD45 alternate exon use in
resting CD4+ and CD8+ lymph node T cells revealed evidence of divergent
exon use. Examination of allogeneically activated T cells revealed that
the CD4+ BC-3 helper T cell clone expressed less of the Ex(l) isoform,
while the CD8+ 8.2.2 CTL clone increased its production of higher alternate
exon isoforms, including Ex(2) and Ex(3) isoforms. Analysis of isoform
expression among thymocytes and T cells suggests that shuffling of CD45
alternate exons occurs in.an.organized and predictable sequence during the

process of T cell maturation and activation.

ACKNOWLEDGEMENTS

I would like to thank my thesis advisor, Dr. Walter Esselman, for
his long and constant support academically and financially. I would also
like to thank members of my guidance committee, Dr. Pamela Fraker, Dr.
Richard Schwartz, Dr. John Vang, and Dr. Jeffrey Williams, for their
advice and interest in.my research topics. Special thanks to Dr. Michael
Zaroukian for his financial support and constructive discussion, and Mary
Morrison, Howard Beittenmiller for their fine technical assistance. I
would also like to express my sincere appreciation to Dr. Jia-Ling Yang,
Dr. Jin-Dwan Lee, Dr. Veronica Maher, Dr. Justin McCormick, Dr. Jerry
Dodgson, Dr. Lekan Latinwo, and Dr. Ronald Patterson for helping me to

start the molecular biology research.

iv

TABLE OF CONTENTS

95mm; Page
List of tables .................................................... vi
List of figures ................................................... vii
Introduction ...................................................... 1
Chapter One

Literature review ........................................... 7

List of References .......................................... 31
Chapter Two

T200 alternate exon use in murine lymphoid cells determined

by reverse transcription-polymerase chain reaction .......... 40

List of References .......................................... 69

Chapter Three
Developmental expression of CD45 alternate exons in murine

T cells: evidence of additional alternate exon use .......... 73
List of References .......................................... 112
Summary and conclusions ........................................... 116

II

LIST OF TABLES

Bag:
Chapter Two
PCR primers used to amplify the region of alternate
exon use in T200 cDNA ................................ 49
Analysis of alternative exon use and identification
of RT-PCR products ................................... 57
Chapter Three
Oligonucleotides used for amplification of the
alternate exon region in CD45 cDNA and the probes
used for Southern hybridization ...................... 82

vi

LIST OF FIGURES

Page
Literature Review
Genomic structure of mouse CD45 gene ................. 10
Structure of CD45 glycoprotein ....................... 17
Chapter Two
Strategy for PCR amplification of the region of
alternate exon use in T200 cDNA ....................... 44
PCR of T200 cDNA-containing plasmids pLy-5-68 and
702/3-3 and RT-PCR of cell lines ...................... 52
Immunnoprecipitation and SDS-PAGE analysis of T200
glycoprotein .......................................... 54
Neil and BanI restriction enzyme mapping of gel-
purified products of RT-PCR of T200 alternate exon
forms in 383 cells .................................... 58
Survey of T200 alternate exon use in lymphoid and
myeloid cells as demonstrated by RT-PCR ............... 62
Chapter Three
Illustration of the amplified region of CD45 and
the relative position of RT-PCR primers and probes
for Southern hybridization ............................ 77
RT-PCR pattern of CD4+ thymocytes showing the
presence of the 124 bp and 100 bp isoforms ............ 86
Southern analysis of the Ex(0) form and the 124 bp
and 100 bp isoforms ................................... 88
Sequencing data showing the absence of exon 7 in
the 124 bp fragment isolated from CD4+ LN T cells ..... 91
Autoradiography of RT-PCT pattern of CD4JCD8'
thymocytes at the end of every 5 PCR cycles ........... 93
Amplification curves of CD45 isoforms in CD4’CD8'
thymocytes ............................................ 95
CD45 RT-PCR pattern of murine T cell subsets and
alloreactive T-cell clones ............................ 98

vii

Restriction enzyme analysis of Ex(l) form purified
from CD4+LNT-cells, CD8+ LNT-cells, and Te
clone 8.2.2

Summary of developmental expression of CD45
isoforms in murine T cells

viii

INTRODUCTION

The overall objective of the experiments described in this
dissertation was to identify the CD45 isoforms expressed in murine
lymphoid cells at the mRNA level. The main focus of the research was on
the characterization of CD45 expression in.T cells during development and
activation. CD45, also known as Ly-5, T200, 8220, and LCA, is an
alloantigen expressed on all hemopoietic cells except erythrocytes. Since
its first discovery by Dr. I. Trowbridge in 1978, CD45 has been of
interest as a differentiation antigen and has recently been implicated in
the regulation of the immune response. A few examples include 8 cell
proliferation, cytolytic T cell lysis, target cell binding,and lysis by NK
effector cells. Recent data showed that CD45 was a substrate for protein
kinase C, suggesting an involvement in transmembrane signal transduction.
Most recently CD45 was found to have protein tyrosine phosphatase activity
in its cytoplasmic domain. It is possible that this phosphatase activity
may strip the phosphate groups from kinases themselves or directly
dephosphorylate the substrates for tyrosine ‘kinase. This finding,
combined with the observation that cells with higher phosphatase activity
were more recalcitrant to transformation, suggested that CD45 may function
as a tumor suppressor gene.

CD45 consists of a family of high molecular weight transmembrane
glycoproteins with apparent molecular weight normally ranging from 180 kDa

to 220 kDa on sodium dodecyl sulfate-polyacrylamide gel electrophoresis

2

(SDS-PAGE) . CD45 isoform of extraordinarily high molecular weights (about
260 kDa) have been reported in intraepithelial lymphocytes (IEL) . It has
been shown that there is only one copy of CD45 gene in the mouse genome
and the gene spans a length of about 130 kb, consisting of 34 exons. The
molecular weight heterogeneity observed by SDS-PAGE is due to both
differential glycosylation and alternate exon use among exons 4, 5, and 6.
Evidence for alternate exon use comes from sequencing cDNA clones isolated
from thymocytes and pre-B cells. It is believed that each of the three
exons can be used independently of each other, which generates the
possibility of eight combinations of the three exons. Exon la and lb are
also alternately used but in a mutually exclusive way. Since la and lb
are in the 5' untranslated region, the selective use of either exon does
not affect the molecular weight of CD45 glycoprotein. So far, there have
been no reports of the existence of other alternate exons.

The expression of CD45 has been found to be cell lineage-specific
and dependent on the state of maturation and activation. It was reported
that thymocytes expressed predominantly the smallest CD45 mRNA (4.8 kb),
pre-B cells the largest (5.5 kb), and lymph node T cells the intermediate
sizes. The 4.8 kb mRNA was considered to be the isoform without any of
the three alternate exons, and the largest one to be the isoform
containing all the three exons. There have been no conclusive data
showing exon use in T cells expressing the intermediate sizes of mRNA.

By using either exon-specific probes in Northern hybridization or
labeling the cells with exon-specific monoclonal antibodies, it was
possible to determine whether a certain exon was present in the isoform
but it was impossible to determine how many CD45 isoforms were present or

whether a specific exon was used in each isoform, alone or in conjunction

3
with the other two exons. In addition, these methods were not sensitive
enough to detect minor isoforms of CD45 which may 'have important
functional roles. Therefore we attempted to establish a methodology which
could not only detect the major and minor isoforms of CD45 mRNAs in cells
but also the exact exon use in each of the isoforms.

In chapter two, a methodology is described which combines reverse
transcription, polymerase chain reaction (RT-PCR) and restriction enzyme
digestion to precisely define alternate exon.use of CD45 mRNA. PCR is an
extremely powerful method for the production of large quantities of DNA
for various kinds of analysis, but has not been used widely until the
isolation of thermally stable Taq polymerase from Thermus aquaticus. The
procedure used consisted of multiple runs of two DNA primer extensions
(extending toward each other on separate strands) in the presence of a
large excess of primers. By raising the temperature to 94'C. the double-
stranded template DNA was denatured into two single-stranded DNAs. The
primers were then allowed to anneal when the temperature was dropped to
the annealing temperature. Finally the primer was extended to the end of
its template at 72°C, whidh was about the optimal temperature of Taq
polymerase. The temperature change from denaturing to annealing, and then
to extension consisted of one cycle of PCR. PCR normally was conducted
for 20-35 cycles depending on the purpose of the experiment. All of the
extended products or the amplified products from the previous cycle are
denatured and serve as potential templates for the following cycles. This
allows PCR to increase the quantity of the target DNA exponentially for
the first 15-25 cycles depending on the starting template concentration.

The basic concept behind the protocol was to use RT-PCR to analyze

the CD45 mRNA isoforms by first reversely transcribing RNA into cDNA which

4

was then used as template in the subsequent PCR. After PCR, the double-
stranded DNA products were digested by restriction enzymes to confirm the
presence of the alternate exons in each individual isoform of interest.
PCR primers flanking the alternate exon region were designed in such a way
that the sizes of PCR products were between 620 bp and 190 bp
representing, respectively, the mRNA containing all three exons (three
exon form) and none of the three exons (zero exon form). Since each of
the exons is about 140 bp in length, PCR products containing one of the
three exons would be around 330 bp in length (one exon form), while the
PCR products around 470 bp in length would contain two of the three exons
(two exon form). Therefore, by determining the sizes of the PCR products
by electrophoresis, the number of exons in each isoform was readily
determined. Neil and BanI digestion of the PCR products provided
conclusive information about which exons were present in particular CD45
mRNA isoforms.

By use of RT-PCR, we have confirmed 1) that one cell line can
express more than one CD45 mRNA isoform; 2) that pre-B cell lines
expressed predominantly the largest isoform together with smaller amounts
of some intermediate forms; and 3) that T cells expressed the smallest and
the intermediate mRNAs as the major isoforms.

In chapter three, the nature of two unidentified bands observed in
the initial studies was determined by directly sequencing the isolated PCR
products. The results indicated that the unknown amplified products were
new CD45 isoforms. One of the newly identified isoforms did not have any
of the exons between exon 3 and 8 (minus one form), thereby showing that
exon 7 is another alternate exon. Evidence from Southern hybridization

with exon specific probes suggested that exon 8 was also alternately used

5
giving rise to a CD45 isoform missing exons between exon 3 and 9 (minus
two form).

In addition, the potential of using PCR to compare relative amounts
of mRNA by using 32-P end-labeled primer was explored. It was found that
the PCR procedure amplified shorter templates more efficiently than larger
templates. Under the experimental conditions, between cycle 10 and 25,
the one exon form, the zero exon form, and the minus one form were
amplified exponentially at about the same rate. However, after the 25th
cycle the increase of the one exon form reaches saturation faster than the
shorter forms. Based on this study, the method was modified to terminate
the PCR amplifications at cycle 24, so that the intensity of each isoform
at the end of PCR better represented the relative quantity of the mRNA
isoforms present at the outset of the experiment.

The reverse transcription-polymerase chain reaction method has
several advantages. First, successful amplification requires very few
cells, only 300-30,000, which makes possible analysis of rare T cell
subpopulations. Second, the method is effective with fresh or frozen
cells, thus enabling the analysis of samples collected over a long period
of time. Third, RT-PCR saves time and reagents since it does not require
prior RNA isolation. Finally, RT-PCR is sensitive enough to detect the
presence of minor isoforms in complex mixtures.

Our analysis of CD45 expression during thymocyte and T cell
development indicated that, except for double negative thymocytes, the
zero and the minus one exon isoforms were among the most abundant mRNA
isoforms present. The putative minus two exon isoform was also
detectable, in trace amounts, in all thymocytes and T cells analyzed. A

comparison of other subsets of T cells indicated that double negative

6
thymocytes had low expression of CD45 in general and these cells only
expressed the minus one and minus two exon isoform. 0n the other hand
double positive thymocytes expressed comparably higher levels of CD45,
which consisted mainly of zero and minus one exon isoforms. In general,
mature T cells, except the CD4+ T cell clone, exhibited increased
expression of higher exon isoforms as they moved to higher differentiation
and activation states. After the digestion with Neil, Hian, and BanI,
the one exon isoforms observed in both CD4+ and CD8+ cells were found to

contain mainly exon five with very little exon 6 and exon 4.

LITERATURE REVIEW

W The existence of the CD45 glycoprotein
family (also known as LCA, Ly5, T200, and 3220) was first described in
1975 by Trowbridge et a1. (1). These glycoproteins were found on the
surface of mouse T and B cells, as well as on mouse T lymphoma cells with
an apparent molecular weight around 200 kDa as determined by SDS—
polyacrylamide gel electrophoresis (SDS-PAGE)(1). CD45R (restricted) has
been used to indicate monoclonal antibodies recognizing unique epitopes of
CD45 whose expression was found to be restricted to certain isoforms (2,
3). Initially, the expression of CD45 was thought to be lymphocyte-
specific (4), but later it was demonstrated that CD45 was expressed on all
hemopoietic cells except erythrocytes (5-7).

Based on antibody binding, it was estimated that thymocytes and
spleen cells had about 50,000-100,000 CD45 molecules per cell surface,
covering about 10% of the cell membrane (8). CD45 together with Thy-l
glycoprotein constitute the 'major' proteins on the thymocyte plasma
membrane (8). Compared to splenocytes, bone marrow cells have lower
amounts of CD45, while the T cell lymphoma cell line, BW5147, expresses
about four times more CD45 than splenic lymphocytes (8). In general, the
expression of CD45 on macrophages and granulocytes is lower than that
found in lymphocytes (9).

CD45 expression.was found to be enhanced.by IL-2, interferon, or by
certain degranulating agents (10, 11). Recently, an intracellular pool
(in the tertiary granules) has been reported to contain at least 50% of
the total CD45 molecules in human neutrophils (10). Upon stimulation with
a degranulation-inducer, much of the intracellular CD45 relocalized to

plasma membranes. It is not clear whether other cell types also have a

8

similar intracellular pool. Phorbol myristyl-13-acetate (PMA) was found
to cause the reexpression of CD45 in a CD45 negative mutant derived from
a human T cell leukaemia line (12).

WW}... CD45 is a molecularly
heterogenous family of transmembrane glycoproteins that have been found in
all the mammalian systems studied so far, including mouse, human, and rat
(13-18). The apparent molecular weights observed by SDS-PACE analysis
normally ranged from 175 kDa to 240 kDa depending on the investigator.
The molecular weight heterogeneity has been proposed to be due to variable
N-terminal peptide sequences coded by three alternate exons, namely exons
4, 5 and 6 (16-18); and to large differences in carbohydrate content (14,
16-18). Recently, a large isoform of CD45 (260 kd) was reported on intra-
epithelial lymphocytes (IEL) from the mouse (19, 20). Since other
alternate exons have not been reported, the molecular weight increase over
the normal range was believed to be due to extra glycosylation (20).
Experiments from many different laboratories suggested that one cell type
can express more than one CD45 isoform, although some of the isoforms may
be present in barely detectable levels (13-18, 21). Normally, B cells and
thymocytes expressed the highest and lowest forms respectively, while T
cells, macrophages, and granulocytes expressed multiple forms of
intermediate size (5, 8, 22). In addition, each cell lineage may express
CD45 differently at different developmental stages (23-25).

W Mouse CD45 cDNA from a T cell clone, Cl.Lyl-Tl,
was cloned and sequenced by Shen and Saga (13). Based upon their data
from Southern hybridization, as well as from restriction mapping of
overlapping genomic clones, they concluded that there was only one copy of

the CD45 gene in the mouse genome. They further proposed that CD45

9
isoforms were generated by differential processing of transcripts rather
than by gene rearrangement (13). The hypothesis was later confirmed by
sequencing cDNA clones (16-18).

At least three alleles of CD45 have been identified by observation
of restriction fragment length polymorphisms and by serological
phenotyping (26) and.have been designated Ly-S‘, L.y-5b and Ly-5° (27, 28).
Ly-Sb appeared as the most common genotype (27) and was found in Balb/c,
CS7BL/6J , DEA/2 and in many other common strains (27). On the other hand,
Ly-5° was the least common and was only found in mice of ST/bJ strain (27) .
The nomenclature for the Ly-S alleles has recently been revised (29).

The mouse and human CD45 gene locus have both been mapped to
chromosome 1 (14, 27). The mouse gene spanned about 130 kb including 34
exons, 1a or lb, and 32 additional exons (16, 17, 28; Figure l). The
mature mRNA was about 5.1 kb coding for 1,268 amino acids (30). Exons 3-
15 were found to encode amino acid residues 1-538 of the extracellular
domain, exon 16 encoded residues 539-574, encompassing the transmembrane
domain, and exon 17-33 encoded residues 575-1268 of the cytoplasmic domain
(30). The exons were found to vary in size, but most were between 50 and
200 base pairs in length (28, 30). The largest exon, 33, was about 1.1 kb
in length and contained a termination codon plus the 3' untranslated
region (30). The splice junctions of all exons followed the AC/GT rule
(16, 28). The sizes of the introns were widely different, from the
shortest of 81 base pairs between exons 2 and 3, to about 50 kb in length
between exons 3 and 4 (28).

Based upon.primer extension studies, the first two exons, namely la
and lb, were shown to be expressed in a mutually exclusive way (16), and

were presumably from two different transcription initiation sites (16).

10

Figure 1. Genomic structure of mouse CD45 gene. Exons are represented by
the vertical lines. Exons 1a to 15 are in the extracellular domain, while

exons 17 to 33 in the cytoplasmic domain. TM: transmembrane region.

II

no. 09

.H MSME

 

5...

43141333413111

an

omIHN

:

o.

L—
owl:

so u «n «£0—

OHIH

12

In human, exons 1a and lb were used differentially but not in a cell type-
specific manner (16) . According to the accumulated data, lb seemed to be
more frequently selected (16). However, it was reported that in mice the
la containing species of mRNA was much more abundant in T cells than in B
cells (28). Since each of these two exons encodes the 5' untranslated
region, the use of either exon does not appear to affect the peptide
sequence. The significance of the alternative use of exons 1a and lb is
still unknown. Exon 2 encodes part of the 5' untranslated region and most
of the leader sequence (23 amino acid residues). Exon 3 includes the last
2 nucleotides for the last amino acid in the leader sequence and encodes
the first seven amino acids of the mature protein, a sequence that is
found to be common to all CD45 isoforms discovered so far.

The main region for alternative splicing was found from exon 4 to
exon 6 (17, 28). In the murine system these exons were 129, 147, and 141
base pairs or 43, 49, and 47 amino acid residues in length, respectively
(28). The cDNA clone from a pre-B cell line (702/3.12), which expressed
mainly a 220 kd isoform, contained all the exons except la (due to the
presence of the mutually exclusive exon 1b) , whereas the cDNA from
thymocytes expressed the 180 kd form lacking exons 4, 5 and 6 (l6, 17).
It was further established that each of the three exons was independently
used, alone or in combination. Potentially each cell type could contain
up to eight distinct isoforms (28). The use of the three exons not only
changed the peptide sequence but also the extent and pattern of
glycosylation, which in turn determined the molecular weight of the CD45
isoforms on SDS-PAGE. It is now believed that the expression of CD45
isoforms is controlled in a cell type-specific way (1, 4, 8, 16, 28) and

that any one type of cells can express more than one isoform. So far, six

13
of the isoforms have been identified (l4, l6, 17).

.Although it is possible that there are still some alternate exons to
be discovered for CD45, there has been no evidence for the existence of
other alternate exons other than 1a, lb, 4, 5, and 6. Therefore epitopes
on the peptide sequences encoded by exon 3 and by exons 7 to 32 are
believed to be present on all CD45 isoforms, and have been termed common
epitopes (31, 32). By contrast, epitopes that were found on the protein
sequences coded by the alternate exons were not common to all the
isoforms, and thus were referred to as restricted epitopes (CD45R). In
addition, the restricted epitopes were not necessarily due to differences
in the peptide sequence but were also believed to be due to differential
glycosylation.

W Although the mechanism controlling cell
lineage-specific CD45 splicing is still unknown, it is believed that
selective use of the transcription initiation sites does not determine
splicing patterns (33). Consequently, the existence of both trans-acting
and cis-acting factors was proposed to account for the regulation of the
splicing of CD45 RNA. Some knowledge of the regulatory elements for
splicing has been acquired from studies of a human-mouse chimeric system
(34, 35). In order to study the factors involved in tissue-specific
splicing, Streuli et al. transfected the murine pre-B cell line 300-19 and
thymocyte cell line EL-4 with mini-gene constructs that contained only
part of the human CD45 sequence, including several of the alternate exons
(exon 4 and/or exon 6) and the SV40 promoter. The transcripts were then
analyzed by primer extension to determine whether pre-B cells retained the
alternate exons in their mature mRNA, and whether the thymocyte isoform

excluded all of the alternate exons. The results indicated that the same

14

primary transcripts were discriminately processed in the pre-B cell line
and thymocyte line as expected (36). This finding suggested.that; (a) the
phenomenon of alternate splicing for CD45 mRNA probably was not the
consequence of different pre-mRNA structures resulting from alternative
promoter usage or alternative poly(A) site usage; (b) there was some
trans-acting factor that regulated the specificity of the CD45 pre-mRNA
splicing; and (c) whatever the trans-acting factor was, it must have been
well conserved between human and mouse, in that the trans-acting factor
from the murine system seemed to be functional on the human DNA sequence
and this factor worked with both exon 4 and exon 6 (36). It was also
concluded that exon 5 was used in a poorly regulated way and thus its use
might not be tissue-specific (36).

Transfection of CD45 into the nonhemopoietic cells, NIH-3T3, HeLa,
and L cells indicated that the splicing pattern was essentially the same
as that in thymocytes (36) . This suggested that thymocytes and non-
hemopoietic cells have the same or functionally identical trans-acting
factor, if present. It was hypothesized that there was no such trans-
acting factor in the thymocytes and nonhemopoietic cells and the primary
transcript was thus processed to remove all the alternate exons. This
hypothesis has been supported by the findings from Saga's group (37).
They examined the splicing pattern of mouse CD45 in hybrid cells made by
fusing T cells and B cells and found that all the hybrids used the
alternate exons in a B cell-specific pattern. Thus, they concluded that
all the information required for faithful splice-site selection according
to cell type was contained within the resultingypre-mRNA, and the splicing
pattern manifested by nonexpressor cells or thymocytes may represent a

default, nonregulated type (37). So far, no trans-acting factors have

15
been identified.

There are several models proposed to explain the alternate splicing
phenomenon observed for CD45 (36) , but none of these has been found to be
entirely satisfactory. The results obtained in studies of cis-acting
elements have been more conclusive. By use of various kinds of deletion
mutants, it has been demonstrated that the critical cis-elements for
specific splicing are inside the alternate exons themselves and in their
imediate flanking intron regions (from 10 to 200 bp) (36). In addition,
the absence of neighboring constant exons, Ex-3 and Ex-7, did not seem to
affect alternate splicing (36).

By use of linker scanning, it has been found that for Ex-4 to be
alternatively used, at least three different cis-elements within Fix-4 were
crucial. In other words, if certain sequences were modified without
significantly changing the length, the splicing of the mutated transcripts
were no longer tissue-specific. The critical sequences were between
nucleotide positions 8 and 17, 40 and 91, and 127 and 137 in Ex-4 (36).
For Ex-6 the critical sequences lie between nucleotide positions 16 and
144 (34) . However, Saga et al. reported that some of the intron sequences
flanking both sides of exon 6 were also important in deciding the
exclusion of that exon in mouse T cells (37). Although it has not been
confirmed, it is possible that Ex-S has crucial sequences for tissue-
specific splicing. Although sis-elements have been identified, the
mechanism by which they affect specific splicing is still obscure. It is
possible that they consist of binding sites for one or more trans-acting
factors, or perhaps they are important for secondary structure formation
(36).

W The complete structure of CD45

l6

glycoproteins of mouse, human, and rat have been elucidated by analysis of
cDNA clones (13-18, 38). The protein was found to consist of 1118 to 1281
amino acids depending on the species and alternative mRNA splicing. The
mature murine CD45 glycoprotein was be divided into three distinct
domains: the N-terminal, extracellular domain (402-541 amino acids
residues depending upon the variable exon usage), the transmembrane domain
(22 amino acids residues), and the C-terminal, cytoplasmic domain (705
amino acids residues) (Figure 2). Electron microscopic examination of the
rat CD45 molecule indicated that it had the shape of a comma (39). The
extracellular domain appeared to be the elongated (tail) portion and the
cytoplasmic domain the globular (head) portion.

Based on protein sequence analysis and biochemical studies, the
extracellular domain was further divided into at least two different
areas: an O-linked, glycan-rich area and an N-linked, glycan-rich,
cysteine-rich area. Peptide sequence coded by exons 3-8 was rich in
serine and threonine (around 34%) but contained no cysteine, which is
characteristic of 0-linked carbohydrate sites (40). Therefore this area
contained potential sites for O-linked glycosylation, and the use of the
alternate exons changes the number of glycosylation sites. It has been
shown that in the rat thymocyte form (lacking exon 4,5 and 6), all the 0-
linked carbohydrates were found within the first 32 amino acids (15),
which were equivalent to the sequences encoded by mouse exon 3, 7 and 8.
Although the overall homology of the alternate exon coded sequences among
the three species (human, rat, and mouse) was found to be only 40%, the
location of the O-linked sites was well conserved. Lefrancois has
proposed that for some cell types the carbohydrates in this area contain

N-acetylgalactosamine in a fil,4-linkage to galactose and sialic acid in an

17

Figure 2. Structure of CD45 glycoprotein. CD45 glycoprotein can be
divided into three domains: extracellular, transmembrane and cytoplasmic
domains. The cytoplasmic domain contains two protein tyrosine phosphatase
-like subdomains. Exons 4, 5 and 6, are the alternate exons. Dark area

represents the transmembrane region.

 

18

 

 

 

CD45 GLYCOPROTEIN
Extracellular CytOplasmiC
34 5 6 73
um ll 1 IIIIIIIIIIHIIIUIIII IIIIIILLIIIIITIIIIIIII
Alternate Trans- PTPase I PTPase ll
Exons membrane

 

Figure 2.

l9
a2,3 linkage (41, 42).

The peptide sequence codedey exons 9 to 15 was found to contain the
cysteine-rich area. Although the homology among the three species was
quite low (only 33%), many key conserved residues were found indicating
that the exterior domainlmay have a conserved three-dimensional structure
(43). This region was found to contain 16 cysteine residues in human and
rat CD45 and 18 in.mouse, and the positions were well conserved except for
the two extra cysteines in mouse. These cysteines were believed to be
disulfide-bonded (38), presumably to maintain the critical structure.
There were also 11-17 potential N-linked sites (depending on the species)
in the extracellular domain (15).

Since observations indicated a lack of protein sequence conservation
in the exterior domain, Thomas et a1. suggested that this sequence may act
as a platform for the carbohydrate structure (19). The carbohydrates,
rather than the protein sequence of the extracellular domain, may be
crucial for the function of CD45. The observations supporting this are:
first, the protein sequence homology was low among the species compared so
far; second, the 0-linked glycosylation was precisely regulated via
alternate exon usage; and third, the locations and number of cysteine
residues were well conserved. Additionally, there has been evidence
showing the importance of carbohydrate residues in NK cell functions (44) .

In contrast to the extracellular domain, the large cytoplasmic
domain of CD45 was highly conserved among the species with >85% identity.
There were sequences of over 100 amino acids that were 95% homologous.
Thomas suggested that the discrepancy in conservation between the
extracellular domain and the cytoplasmic domain implies that these domains

perform different functions (43). In the cytoplasmic region two

20
subdomains were found, each of 300 amino acids in length, sharing 33%
homology between each other (30). Each of the two cytoplasmic domains was
found to have about 40% homology to placental protein tyrosine phosphatase
(45). The cytoplasmic domain was also phosphorylated (46). This
discovery raised the possibility that CD45 could serve as a signal
transducer.

W CD45 has been proposed to function in the
following cells; stem cells (47, 48), NK cells (11, 22, 44, 49-53) and
lymphoid cells (2, 18, 20, 47, 54-61). Most of the hypothetical functions
were based on observations involving the use of antibodies to disrupt
cellular functions. Much of the data suggests that CD45 probably serves
as the receptor/ligand molecule which transduces signals across the
membrane resulting in cell lysis, proliferation, activation or anti-
transformation. However, since the specificity of many of the antibodies
is not clear, contradictory results using different antibodies were
obtained. The development of new exon-specific or cell lineage-specific
monoclonal antibodies with accurate epitope definition will greatly help
our understanding of the function of CD45.

NK cells were found to express multiple forms of CD45 (22, 49). A
number of antibodies against CD45 have been generated that have the
ability to inhibit NK activity (11, 49, 52, 53, 62-65). Based upon the
abundance of accumulated data, CD45 was proposed to be one of the critical
molecules implicated in conjugate formation between the NK cells and
target cells, and the ensuing target cell lysis (ll, 52, 61, 66).
Attempts to identify receptors on NK cells associated with conjugate
formation or cytolytic activity by generating and screening monoclonal

antibodies which blocked NR functions revealed that almost all such

21

antibodies have reacted with one or more epitopes of CD45 (67) . Anti-Ly-S
sera and monoclonal antibody 13.1 against CD45 were examples of inhibitors
of NR function in the mouse and human, respectively (44, 50, 52). Since
there were multiple forms of CD45 on NR cells, it was difficult to assign
a specific role to each individual isoform. It was not clear whether all
the isoforms perform the same kind.of function" To clarify this question,
identification of all the isoforms expressed and definition of the
epitopes recognized by the monoclonal antibodies for each CD45 isoform was
considered imperative.

The involvement of CD45 carbohydrate structure in NR cell function
has been explored. Gilbert et a1, discovered that purified CD45
incorporated into liposomes inhibited conjugate formation between the NR
cells and the target cells (44). However, no inhibition was observed if
the liposomes were pretreated with Ly-S antisera or enzymes to remove some
of the carbohydrate (44). This research ‘was among the first to
demonstrate a connection between CD45 function and its carbohydrate
structure.

When small resting human B cells were incubated with antibodies
against CD45, the inducible proliferation by anti-IgM and T cell-replacing
factors was abolished (55). Anti-Ly-S sera greatly affect the generation
of plaque-forming cells (plasma cells) specific to T cell-dependent
antigens, but not T cell-independent antigens (54). The antibodies were
also capable of inhibiting the IgG response when B cells were induceduwith
lipopolysaccharide, whereas proliferation of IgM secreting cells or IgM
response was not changed (56). The phenomenon was later rationalized as
a decrease in.the number of cells committed to class-switching rather than

a defective clonal expansion of IgG-producing cells (57).

22

The involvement of CD45 in.T cell functions is more complicated due
to complex patterns of CD45 expression on T cell subsets (13-18, 21).
The observation of the general correlation between the differential
expression of CD45 and the distinct functions of T cell subsets, implied
that CD45 was important in moderating their cellular functions (18). It
has been observed that the ratio of T cell subsets, defined by the CD45
epitope expression, was altered in the course of some diseases such as
rheumatoid arthritis (68), multiple sclerosis (69, 70), and multiple
myeloma (71). In mice with recessive lymphoproliferative disorders, lpr
or gld, lymph node T cells abnormally expressed CD45 antigen with more of
the high molecular weight isoform (similar to that of the B cell form)
than that in normal animals (72). As a result of abnormal proliferation,
lymph nodes enlarged up to 50-fold. It remains unclear whether the
diseases result from the irregularity of CD45 expression.

There has been evidence showing that CD45 was indeed involved in T
cell activation or differentiation. In the mouse model, the addition of
anti-Ly-S sera to the culture resulted in the inhibitory effect on the
generation and lytic ability of cytotoxic T lymphocytes in the mixed
lymphocyte reaction (47, 58, 59). Another line of evidence was provided
by the anti-CD45 monoclonal antibody 2H4, which distinguished CD4+ T-helper
cells and CD4+ suppressor inducer T cells (23). When the suppressor
inducer T cells were pretreated with 2H4 antibody, they lost the ability
to induce suppression (73). In another experiment, the same antibody was
added directly into the culture, but surprisingly it enhanced. the
suppression (60). Thus far, it is still not clear why the same antibody
'had the opposite effect: under different conditions. In addition,

antibodies against different epitopes of CD45 may have totally opposite

23
effects (2, 20). Monoclonal antibody against the restricted epitope on
CD45 augmented the mitogenic response to suboptimal dose of PHA (2). 0n
the other hand, antibody against the common epitope suppressed PHA and
ConA responses (20). It is possible that antibodies against different
epitopes act on a different subset of T cells, consequently inducing
opposite functions. 0n the other hand cells may express more than one
CD45 isoform and, depending on the isoform that the antibody binds, can be
induced to perform diverse functions.

Several recent studies indicated that CD45 might be involved in
early T cell activation. It was proposed that antibodies stimulate early
T cell activation by making some alteration in the IL-2/IL-2 receptor
pathway (2, 20). Monoclonal antibody against the common epitope on human
CD45 together with Sepharose-coupled anti-CD3 antibody acted as a mitogen
for T cells and resulted in T cell activation. Neither of the two alone
was mitogenic (61) . The most direct evidence showing the involvement of
CD45 in the T cell response came from the work of Pingel and Thomas in
which they found that a T cell mutant lacking CD45 could not proliferate
in response to antigen even though they expressed other T cell surface
proteins normally (74).

CD45 was found to be physically associated with the cytoskeletal
protein fodrin through its cytoplasmic domain and participated in receptor
patching and capping (75, 76). Although the significance of this
association is not certain, it is probably an indication of a role in
signal transduction or in regulation of cell membrane organization and
mobility (43). CD45 was also found to be physically associated with CD2
in human T lymphocytes (77) .

W The mechanism of function of CD45 was better

24

understood upon the discovery that CD45 was a protein tyrosine phosphatase
(PTPase) (78). In lymphoid and myeloid cells, considerable interest has
emerged from the recent discovery that the cytoplasmic domain of CD45
possesses two domains bearing significant amino acid sequence homology
with placental protein tyrosine phosphatase lB (PTPase-18) (45). Tonk's
group presented evidence showing that CD45 had PTPase activity (78) . They
suggested that the binding of CD45 to its ligand modulated the CD45
enzymatic activity as a protein tyrosine phosphatase, thus resulting in
the regulation of protein tyrosine kinase activity, whose activity is
related to cell proliferation and transformation (78). By using a CD45-
negative mutant, Roretzky et a1. concluded that CD45 was essential for
coupling T cell antigen receptor to the phosphatidyl inositol (PtdIns)
pathway to generate the PtdIns-derived second messenger (12). In
addition, several investigators proposed that CD45 may dephosphorylate
p56“‘" (79, 80), a protein tyrosine kinase encoded by the 1ck proto-
oncogene. CD45 has been shown to activate p561“ in a lymphoma cell line
by dephosphorylation of a regulatory site.

Perhaps the most interesting proposed function of CD45 is that of an
anti-oncogene or tumor suppressor (81). This concept was derived from an
experiment in which the results of transfection with mutated human neu
oncogene were compared between normal mouse 3T3 cell line and the same
cell line with over-expressed PTPase 18 activity. They found that
introduction of the mutated human neu oncogene into the normal cells
aggressively induced transformation presumably by acting as an unregulated
tyrosine kinase. By contrast, a severe reduction in transformation was
observed in the PTPase over-expressing cells. Since CD45 shares homology

(about 40%) with PTPase 18 and has been shown to have PTPase activity, it

25
was thus suggested that CD45 may be an anti-oncogene.

Overall, the above evidence suggested a model in which CD45
extracellularly interacts with other molecules (or its ligand), transduces
a signal inside through the cytoplasmic domain of CD45, which then
regulates the activity of some cytoplasmic component, and finally results
in the cellular response. There are, however, several unanswered
questions. Are all the isoforms of CD45 capable of performing the
functions as proposed, or is it just a privilege of certain isoforms?
Since all the isoforms have the same cytoplasmic domain, upon interacting
with their specific molecule (stimulus) on the outside, is the signal
transduced inside of the same quality and intensity for all the isoforms?
Perhaps, the difference in alternate exon usage and glycosylation of each
isoform is to construct the specificity for the individual isoform so that
each of them can react specifically with certain molecules or ligands. If
every different isoform has its own distinct function, it will be
essential to know both the exact isoform expression and the relative
quantity of the isoforms. It should also be kept in mind that all the
complicated functions proposed for CD45 probably involved the co-presence
of other molecules.

W There have been many antibodies developed
against CD45, but the specific epitopes recognized by a given antibody is
still not well defined. Initially, antibody specificity was determined by
comparing the data obtained by Northern hybridization and by anti-CD45
cell surface labeling. The ambiguity of this approach resulted from the
multiple expression of CD45 on a single cell. This was solved by the
establishment of transformants obtained by infecting CD45“ cell lines with

expression vectors containing only one specific cDNA isoform of CD45 (82) .

26

Based on these data, the epitope recognized by some of the monoclonal
antibodies could be said, at most, to be dependent on the presence of a
specific exon, but the property of the epitope was still obscure. This
was because of the fact that the antibody may have recognized the sequence
around the exon junction, or it may have recognized unique carbohydrate
residues rather than the peptide sequence (41, 58, 83). It is possible
that one particular isoform may have dissimilar glycosylation in different
cell types. Evidence supporting the tissue-specific glycosylation of CD45
was provided by the existence of CTl and CT2 antigens only on murine CTL
(41) , and the extraordinarily high molecular weight (260 kd) only observed
in IEL (19, 20).

Antibodies or anti-sera believed to recognize the common epitopes on
CD45 include I3/2 (8) and Ly-5.l (84) in mice, and EO-l in humans (25).
Antibodies that have been claimed to be directed against a specific
epitopes include: 2H4 for exon 4 in humans (60), UCHLl for the junction
between exon 3 and exon 7 in humans (85) , NR-9 for carbohydrate on human
NR cell CD45 (49), CTl and CT2 for carbohydrate on CD45 of the murine CTL
(41), 14.8 and RA3-2C2 for exon 4 in mice (86), C363.16A, M82362 and MB484
for exon 5 in mice (82, 86), and OX-22 for exon 5 on rat CD45 (31). It
should be kept in mind that the epitopes assigned to a certain monoclonal
antibody may be changed with the emergence of more conclusive data.

WM Although it is believed that
the expression patterns of CD45 glycoproteins are not only cell lineage-
specific but also related to the degree of maturation and activation (35,
82, 87) , only small progress has been made on the characterization of CD45
expression during lymphoid cell development (21, 23, 31, 35, 82, 87-89).

It has been reported that the size of mouse CD45 mRNA is 4.8 kb for

27
thymocytes, 5.2 kb for cytotoxic T lymphocytes, 4.8 kb for T; clones, and
5.5 kb for B cells (21). Most of the study has been centered on T cells
with very limited research on B cells (21, 23, 31, 35, 88), whereas study
of non-lymphoid cells is almost absent.

The isoform expression on pre-B cells and immature thymocytes seems
to be much simpler. Based on cDNA screening and sequencing, the former
has the full length isoform with all the three alternate exons, namely
exon 4, 5, and 6 (17), while the latter has the shortest isoform (13).
Since some of the minor cDNA clones may be lost during library
construction or colony screening, it is possible that by a.more sensitive
method, other minor isoforms may be detected. By primer extension, minor
isoforms containing only one or two of the three exons were also detected
in B cells in addition to the major isoform, which included all three
exons (37).

Conclusions regarding the expression pattern on other cell types
were either incomplete or sometimes conflicting in different reports.
Most of these conclusions were based on Northern hybridization or antibody
labeling. It was reported that when thymocytes became mature, some
subsets expressed the larger CD45 isoforms (CD45R) including exon 4, 5
and/or 6 (35, 82, 87, 89). After stimulation with either antigens or
ConA, the expression of larger isoforms was again down-regulated, while
the smallest isoform was up-regulated (35, 89). Birkeland et a1. (82)
reported similar results. They found that when resting T cells (either
CD4+ or CD8+) were stimulated with allogenic dendritic cells, they lost the
190 kDa isoform (presumably the exon 5 dependent isoform)‘with an increase
of the 180 kd form (82). According to Bottomly et al., about 45% of CD4+

T cells expressed exon 5 dependent antigen on the surface (90). Among

28
them, one third expressed lower levels of exon 5, which were the Th2
subpopulation, with the more brightly stained cells representing the TM
subpopulation (90).

The majority of the results regarding expression were obtained from
cDNA sequencing (13, 17) , Northern hybridization with exon-specific oligo-
probes, or monoclonal antibody labeling. These methods have certain
disadvantages and limitations. Sequencing cDNA clones from the cell types
to be studied certainly would be the most accurate way, but it is
impractical. Northern hybridization with exon-specific probes or surface
labeling with monoclonal antibodies can only offer accurate information
about the presence and relative quantity of each particular exon, but it
can not answer how many or which combinations of the exons are present.
Furthermore, the epitopes recognized by some of the monoclonal antibodies
are not well defined as discussed above. Consequently, the conclusions
based upon the antibody labeling can be misleading.

mm It was hypothesized that the polymerase
chain reaction (PCR) could resolve problems about the exon usage of CD45.
Although PCR is a newly developed technique (91, 92) , its application has
been found in a large variety of research areas as described in reviews
(93). The widespread use of PCR was due to its ability to amplify very
small amounts of DNA into sufficient quantities for later manipulation
(92) . In brief, PCR consisted of consecutive, multiple runs of DNA primer
extension mediated by a thermal stable DNA polymerase (Taq polymerase,
111m M15335) in the presence of a large excess of two specific
primers. The primers were usually around 20-30 bases in length and
synthesized according to the known sequences flanking the DNA segment of

interest (target DNA). Under precise control, the PCR reagents were

29

subjected to about 30 cycles of the temperature changes. In each cycle,
the double-stranded template DNAuwas first denatured into single-stranded
DNA at about 94’C. When the temperature drops, the primers annealed to
their own complementary strand of template DNA. After the annealing, the
temperature was brought to 72’C, the optimal temperature for Taq
polymerase, and the primers extended to the end of their sequence. The
process of denaturing, annealing, and extension consisted of one PCR
cycle. Both the original template DNA and the product DNA were available
to be used as the template in the following cycle. Normally each PCR
reaction included about 30 cycles. It has been estimated that at the end
of 60 PCR cycles, amplification of up to a factor of 1012 can be achieved
(94). This accounts for the usefulness of PCR in detecting the trace
amount DNA extracted from tiny amount of samples, such as hair roots (95)
or saliva (96) in forensic cases.

Since PCR was used with very specific primers to the flanking
sequences, and the primer extension was carried out at high temperature,
the chance of mispriming was very limited. Consequently, residual amounts
of target sequence in a very noisy background still operated as templates
without prior purification. A.standard PCR reaction of 30 cycles took
only about 3 hours and.after that the product was ready for various assays
such as electrophoresis, sequencing, or restriction enzyme digestion.

Nevertheless, there were several pitfalls and limitations recognized
in the application of PCR. First, if a contaminant sequence was present
in the reaction tubes, a false positive result might be obtained due to
the high level of amplification. This problem was particularly obvious in
experiments designed to detect rare sequences of pathogens in the early

phase of infectious diseases such as human T cell leukaemia or AIDS (97,

30

98) . However, this problem was avoided by taking precautions as described
(93). Second, the error rate associated with Taq polymerase was about 1
in 9,000 for single base substitutions and l in 41,000 for frame-shift
errors (99). Accordingly, extreme caution should be taken in sequencing
DNA from PCR. Third, the sequences flanking both ends of the target DNA
have to be known in order to design the primers for PCR. Another
limitation was the length of sequences to be amplified. Although distance
up to 10 kb have been successfully amplified (100), most of the
applications have been for sequences within several kb. When the length
of target DNA increased, PCR became much less efficient.

There have been successful examples of using PCR in mRNA phenotyping
(92, 94) or for quantitation (101, 102). In those applications reverse
transcription was necessary to convert mRNA into cDNA to function as the
template. The primer for reverse transcription was either oligo (dT)era
or the anti-sense PCR primer. After reverse transcription and PCR, the
sequences of the products represent those of the mRNA in the corresponding
segment. Thus, analysis of the PCR.products enabled information for'mRNA
content to be obtained.

Since the entire sequence of CD45 mRNA was known (l3, 17), it was
feasible to use reverse transcription. together"with. PCR. to explore
alternate exon usage in CD45 mRNA. It was possible to synthesize primers
flanking the alternate exon region such that the PCR products were within
700 bp in length depending how many and which alternate exons were used.
The PCR products were then subjected to more detailed analysis for

characterization.

LIST OF REFERENCES

10.

11.

LIST OF REFERENCES

Trowbridge, I. S., P. Ralph, and M. J. Bevan. 1975. Differences in
the surface proteins of mouse B and T cells. Proc. Natl. Acad. Sci.
USA 72:157.

Ledbetter, I. A., L. M. Rose, C. E. Spooner, P. G. Beatty, P. J.
Martin, and E. A. Clark. 1985. Antibodies to common leukocyte
antigen p220 influence human T cell proliferation by modifying IL-2
receptor expression. J. Immunol. 135:1819.

Pilarski, L. M. , R. Gillitzer, H. Zola, R. Shortman, and R. Scollay.
1989. Definition of the thymic generative lineage by selective
expression of high molecular weight isoforms of CD45 (T200) . Eur. J.
Immunol. 19:589.

Trowbridge, I. 8., and C. Mazauskas. 1976. Immunological properties
of murine thymus-dependent lymphocyte surface glycoproteins. Eur. .
J. Immunol. 6:557.

Omary, M. B., I. S. Trowbridge, and H. A. Battifora. 1980. Human
homologue of murine T200 glycoprotein. J. Exp. Med. 152:842.

Scheid, M. P. , and D. Triglia. 1979. Further description of the Ly-5
system. Immunogenetics 9:423 .

Judd, W., C. A. Poodry, S. Broder, S. M. Friedman, L. Chess, and J.
L. Strominger. 1980. High molecular weight antigens present on human
T cells. Proc. Natl. Acad. Sci. USA 77:6805.

Trowbridge, I. S. 1978. Interspecies spleen-myeloma hybrid producing
monoclonal antibodies against mouse lymphocyte surface glycoprotein,
T200. J. Exp. Med. 148:313.

Shah, V. 0., C. I. Civin, and M. R. Loken. 1988. Flow cytometric
analysis of human. bone marrow. IV. Differential quantitative
expression of T-200 common leukocyte antigen during normal
hemopoiesis. J. Immunol. 140:1861.

Lacal, P., R. Pulido, F. Sanchez-Madrid, and F. Mollinedo. 1988.
Intracellular location of T200 and M01 glycoproteins in human
neutrophils. J. Biol. Chem. 263:9946.

Minato, N., L. Reid, H. Cantor, P. Lengyel, and B. R. Bloom. 1980.
Mode of regulation of natural killer cell activity by interferon. J.

31

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

32
Exp. Med. 152:124.

Roretzky, G. A., J. Pious, M. L. Thomas, and A. Weiss. 1990.
Tyrosine phosphatase CD45 is essential for coupling T-cell antigen
receptor to the phosphatidyl inositol pathway. Nature 346:66.

Saga, Y., J. S. Tung, F.-W. Shen, and E. A. Boyse. 1986. Sequences
of Ly-5 cDNA: isoform-related diversity of Ly-S mRNA. Proc. Natl.
Acad. Sci. USA 83:6940.

Ralph, S. J., M. L. Thomas, C. C. Morton, and I. S. Trowbridge.
1987. Structural variants of human T200 glycoprotein
(leukocyte-common antigen). EMBO J. 6:1251.

Barclay, A. N., D. 1. Jackson, A. C. Willis, and A. F. Williams.
1987. Lymphocyte specific heterogeneity in the rat leucocyte common
antigen (T200) is due to differences in polypeptide sequences near
the NH2-terminus. EMBO J. 6:1259.

Saga, Y., J. S. Tung, F.-W; Shen, and E. A. Boyse. 1987. Alternative
use of 5' exons in the specification of Ly-S isoforms distinguishing
hemopoietic cell lineages. Proc. Natl. Acad. Sci. USA 84:5364.

Thomas, M. L., P. J. Reynolds, A. Chain, Y. Ben-Neriah, and I. S.
Trowbridge. 1987. B-cell variant of mouse T200 (Ly-5): evidence for
alternative mRNA splicing. Proc. Natl. Acad. Sci. USA 84:5360.

Streuli, M., L. R. Hall, Y. Saga, S. F. Schlossman, and H. Saito.
1987. Differential usage of three exons generates at least five

different mRNAs encoding human leukocyte common antigens. J. Exp.
Med. 166:1548.

Thomas, M. L., and L. Lefrancois. 1988. Differential expression of
the leucocyte-common antigen family. Immunol. Today 9(10):320.

Thomas, G. G., H. L. Chang,‘W. J. Esselman, and L. Lefrancois. 1990.
Characterization of the CD45 molecule on murine intestinal
intraepithelial lymphocytes. J. Immunol. (in press)

Lefrancois, L., M. L. Thomas, M. J. Bevan, and I. S. Trowbridge.
1986. Different classes of T lymphocytes have different mRNAs for
the leukocyte-common antigen, T200. J. Exp. Med. 163:1337.

Newman, W., S. Re Targan, and.L. D. Fast. 1984. Immunobiological and
immunochemical aspects of the T-200 family of glycoproteins. Mol.
Immunol. 21:1113.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

33

Morimoto, C., N. L. Letvin, J. A. Distaso, W. R. Aldrich, and S. F.
Schlossman. 1985. The isolation and characterization of the human
suppressor inducer T cell subset. J. Immunol. 134:1508.

Rudd, C. E., C. Morimoto, L. L. Wong, and S. F. Schlossman. 1987.
The subdivision of the T4 (CD4) subset on the basis of the
differential expression of L-C/T200 antigens. J. Exp. Med. 166:1758.

Smeland, E. B., H. Holte, H. R. Blomhoff, F. C. Asheim, T. Stokke,
P . Torj esen, and S . Funderud. 1990 . Inhibition of
polyphosphoinositide breakdown and c-myc induction accompanying
inhibition of human B-cell activation by two monoclonal antibodies

against the leukocyte comon antigen (CD45). Scan. J. Immunol.
31:583.

Shen, F. W., Y. Saga, G. Litman, G. Freeman, J. S. Tung, H. Cantor,
and E. A. Boyse. 1985. Cloning of Ly-5 cDNA. Proc. Natl. Acad. Sci.
USA 82:7360.

Shen, F. W., J. S. Tung, and E. A. Boyse. 1986. Further definition
of the Ly-5 system. Immunogenetics 24:146.

Saga, Y., J. S. Tung, F. W. Shen, T. C. Pancoast, and E. A. Boyse.
1988. Organization of the Ly-5 gene. Molec. Cell. Biol. 8:4889.

Morse, H. C., F. W. Shen, and U. Hammerling. 1987. Genetic
nomenclature for loci controlling mouse lymphocyte antigens.
Immunogenetics 25:71.

Johnson, N. A., C. M. Meyer, J. T. Pingel, and M. L. Thomas. 1989.
Sequence conservation in potential regulatory regions of the mouse
and human leukocyte common antigen gene. J. Biol. Chem. 264:6220.

Woollett, G. R., A. N. Barclay, M. Puklavec, and A. F. Williams.
1985. Molecular and antigenic heterogeneity of the rat
leukocyte-common antigen from thymocytes and T and B lymphocytes.
Eur. J. Immunol. 15:168.

Maddox, J. F., C. R. Mackay, and M. R. Brandon. 1985. The sheep
analogue of leucocyte common antigen (LCA). Immunology 55:347.

Hall, L. R., M. Streuli, S. F. Schlossman, and H. Saito. 1988.
Complete exon- intron organization of the human leukocyte common
antigen (CD45) gene. J. Immunol. 141:2781.

Tsai, A. Y., M. Streuli, and H. Saito. 1989. Integrity of the exon
6 sequence is essential for tissue-specific alternative splicing of
human leukocyte common antigen pre-mRNA. Mol. Cell. Biol. 9:4550.

Clement, L. T., N. Yamashita, and A. M. Martin. 1988. The
functionally distinct subpopulations of human CD4+ helper/inducer T
lymphocytes defined by anti-CD45R antibodies derive sequentially
from a differentiation pathway that is regulated by

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

34

activation-dependent post-thymic differentiation. J. Immunol.
141:1464.

Streuli, M. , and H. Saito. 1989. Regulation of tissue-specific
alternative splicing: exon- specific cis-elements govern the splicing
of leukocyte common antigen pre-mRNA. EMBO J. 8:787.

Saga, Y., J. S. Lee, C. Saraiya, and E. A. Boyse. 1990. Regulation
of alternative splicing in the generation of isoforms of the mouse
Ly-5 (CD45) glycoprotein. Proc. Natl. Acad. Sci. USA 87:3728.

Thomas, M. L., A. N. Barclay, J. Gagnon, and A. F. Williams. 1985.
Evidence from cDNA clones that the rat leukocyte-common antigen
(T200) spans the lipid bilayer and contains a cytoplasmic domain of
80,000 Mr. Cell 41:83.

Woollett, G. R., A. F. Williams, and D. M. Shotton. 1985.
Visualisation by low-angle shadowing of the leucocyte-common

antigen. A major cell surface glycoprotein of lymphocytes. EMBO J.
4:2827.

Triglia, D. 1980. Expression of Ly-5 on yolk sac and fetal liver
cells of the mouse. Immunogenetics 11:303.

Lefrancois, L., L. Puddington, C. E. Machamer, and M. J. Bevan.
1985. Acquisition of cytotoxic T lymphocyte-specific carbohydrate
differentiation antigens. J. Exp. Med. 162:1275.

Conzelmann, A. , and L. Lefrancois. 1988. Monoclonal antibodies
specific for T cell-associated carbohydrate determinants react with
human blood group antigens CAD and SDA. J. Exp. Med. 167:119.

Thomas, M. L. 1989. The leukocyte common antigen family. Annu. Rev.
Immunol. 7:339.

Gilbert, C. W., M. H. Zaroukian, and W. J. Esselman. 1988. Poly-N-
acetyllactosamine structures on murine cell surface T200
glycoprotein participate in NR cell binding to YAC-l targets. J.
Immunol. 140:2821.

Charbonneau, H., N. R. Tonks, R. A. Walsh, and E. H. Fischer. 1988.
The leukocyte common antigen (CD45): a putative receptor-linked
protein tyrosine phosphatase. Proc. Natl. Acad. Sci. USA 85:7182.

Autero, M., and C. G. Gahmberg. 1987. Phorbol diesters increase the
phosphorylation of the leukocyte common antigen CD45 in human T
cells. Eur. J. Immunol. 17:1503.

Harp, J. A., B. S. Davis, and S. J. Ewald. 1984. Inhibition of T
cell responses to alloantigens and polyclonal mitogens by Ly-5
antisera. J. Immunol. 133:10.

Ralph, S. J., and M. V. Berridge. 1984. Expression of antigens of

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

35

the 'T200' family of glycoproteins on hemopoietic stem cells:
evidence that thymocyte cell lineage antigens are represented on
'T200'. J. Immunol. 132:2510.

Nieminen, P., and E. Saksela. 1986. NR-9, a distinct sialylated
antigen of the T200 family. Eur. J. Immunol. 16:513.

Newman, W., L. D. Fast, and L. M. Rose. 1983. Blockade of NR cell
lysis is a property of monoclonal antibodies that bind to distinct
regions of T-200. J. Immunol. 131:1742.

Pospisil, M., J. Rubrycht, R. Bezouska, O. Taborsky, M. Novak, and
J. Rocourek. 1986. Lactosamine type asialooligosaccharide
recognition in NR cytotoxicity. Immunol. Lett. 12:83.

Seaman, W. E., N. Talal, L. A. Herzenberg, L. A. Herzenberg, and J.
A. Ledbetter. 1981. Surface antigens on mouse natural killer cells:
use of monoclonal antibodies to inhibit or to enrich cytotoxic
activity. J. Immunol. 127:982.

Rasai, M., J. C. Leclerc, F.-W. Shen, and H. Cantor. 1979.
Identification of Ly-5 on the surface of natural killer cells in
normal and athymic inbred mouse strains. Immunogenetics 8:153.

Yakura, H., F. Shen, E. Bourcet, and E. A. Boyse. 1983. On the
function of Ly-5 in the regulation of antigen-driven B cell
differentiation: comparison and contrast with Lyb-2. J. Exp. Med.
157:1077.

Mittler, R. S., R. S. Greenfield, B. Z. Schacter, N. F. Richard, and
M. R. Hoffman. 1987. Antibodies to the common leukocyte antigen
(T200) inhibit an early phase in the activation of resting human B
cells. J. Immunol. 138:3159.

Yakura, H., I. Rawabata, F. W. Shen, and M. Ratagiri. 1986.
Selective inhibition of lipopolysaccharide-induced polyclonal IgG
response by monoclonal Ly-5 antibody. J. Immunol. 136:2729.

Yakura, H., I. Rawabata, T. Ashida, and M. Ratagiri. 1988.
Differential regulation by Ly-5 and Lyb-2 of IgG production induced
by lipopolysaccharide and B cell stimulatory factor-1 (IL-4). J.
Immunol. 141:875.

Lefrancois, L., and M. J. Bevan. 1985. Functional modifications of
cytotoxic T-lymphocyte T200 glycoprotein recognized by monoclonal
antibodies . Nature 314 : 449 .

Nakayama, E. 1982. Blocking of effector cell cytotoxicity and T cell
proliferation by Lyt antisera. Inlmunol. Rev. 68:117.

Morimoto, C., T. Matsuyama, C. E. Rudd, A. Forsgren, N. L. Letvin,
and S. F. Schlossman. 1988. Role of the 2H4 molecule in the
activation of suppressor inducer function. Eur. J. Immunol. 18:731.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

36

Brooks, C. C., R. Ruribayashi, G. E. Sale, and C. S. Henney. 1982.
Characterization of five cloned murine cell lines showing high
cytolytic activity against YAC-l cells. J. Immunol. 128:2326.

Sparrow, R. L., and I. F. C. McKenzie. 1983. A function for human
T200 in natural killer cell cytolysis. Transplantation 36:166.

Targan, S. R., and W. Newman. 1983. Definition of a ”trigger” stage
in the NR cytolytic reaction sequence by a monoclonal antibody to
the glycoprotein T-200. J. Immunol. 131:1149.

Zaroukian, M. H., C. W. Gilbert, and W. J. Esselman. 1986. Surface
Ly-S glycoprotein in murine natural killer (NR) cell development,
target binding and cytotoxicity. Immunol. Invest. 15:813.

Starling, G. C., S. E. Davidson, J. L. McKenzie, and D. N. J. Hart.
1987. Inhibiton of natural killer-cell mediated cytolysis with
monoclonal antibodies to restricted and non-restricted epitopes of
the leucocyte common antigen. Immunology 61:351.

Chang, H. L., andNW; J. Esselman. 1988. Analysis of alternative exon
usage in murine T200 cDNA by polymerase chain reaction. J. Cell.
Biol. 107:368a.

Omary, M. B., and I. S. Trowbridge. 1980. Disposition of T200
glycoprotein in.the plasma membrane of a murine lymphoma cell line.
J. Biol. Chem. 255:1662.

Ilowite, N. T., R. J. Wedgwood, L. M. Rose, E. A. Clark, C. G.
Lindgren, and H. D. Ochs. 1987. Impaired in vivo and in vitro
antibody responses to bacteriophage phi X 174 in juvenile rheumatoid
arthritis [published erratum appears in J Rheumatol 1988
Feb;15(2):386]. J. Rheumatol. 14:957.

Rose, L. M., A. H. Ginsberg, T. L. Rothstein, J. A. Ledbetter, and
E. A. Clark. 1985. Selective loss of a subset of T helper cells in
active multiple sclerosis. Proc. Natl. Acad. Sci. USA 82:7389.

Sobel, R. A., D. A. Hafler, E. E. Castro, C. Morimoto, and H. L.
Weiner. 1988. The 2H4 (CD45R) antigen is selectively decreased in
multiple sclerosis lesions. J. Immunol. 140:2210.

Serra, H. M“, M. J. Mant, B. A. Ruether, J. A. Ledbetter, and L. M.
Pilarski. 1988. Selective loss of CD4+ CD45R+ T cells in peripheral
blood of multiple myeloma patients. J. Clin. Immunol. 8:259.

Tung, J. 8., Y. Saga, and E. A. Boyse. 1987. The incongruous Ly-S
phenotype of lpr/lpr and gld/gld T cells. Immunogenetics 25:126.

Takeuchi, T., S. F. Schlossman, and C. Morimoto. 1987. The 2H4
molecule but not the T3-receptor complex is involved in suppressor
inducer signals in the AMLR system. Cell. Immunol. 107:107.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

37

Pingel , J . T. , and M. L. Thomas . 1989. Evidence that the
leukocyte - common antigen is required for anti gen- induced T
lymphocyte proliferation. Cell 58:1055.

Bourguignon, L. Y. W., S. J. Suchard, M. L. Nagpal, and J. R.
Glenney Jr. 1985. A T-lymphoma transmembrane glycoprotein (gp 180)
is linked to the cytoskeletal protein, fodrin. J. Cell. Biol.
101:477.

Glenney, J. R., P. Glenney, and R. Weber. 1982. Erythroid spectrin,
brain fodrin and intestinal brush border protein (TW 260/240) are
related molecules containing a common calmodulin-binding subunit
bound to a variant cell type- specific subunit. Proc. Natl. Acad.
Sci. USA 79:4002.

Schraven, B., Y. Samstag, P. Altevogt, and S. C. Meuer. 1990.
Association of CD2 and CD45 on human T lymphocytes. Nature 345:71.

Tonks, N. R., H. Charbonneau, C. D. Diltz, E. H. Fischer, and R. A.
Walsh. 1988. Demonstration that the leukocyte common antigen CD45 is
a protein tyrosine phosphatase. Biochemistry 27:8695.

Ostergaard, H. L., D. A. Shackelford, T. R. Hurley, P. Johnson, R.
Hyman, B. M. Sefton, and I. S. Trowbridge. 1989. Expression of CD45
alters phosphorylation of the lck-encoded tyrosine protein kinase in
murine lymphoma T-cell lines. Proc. Natl. Acad. Sci. USA 86:8959.

Ostergaard, H. L., and I. S. Trowbridge. 1990. Coclustering CD45
with CD4 or CD8 alters the phosphorylation and kinase activity of
p561ck. J. Exp. Med. 172:347.

Streuli, M., N. X. Rrueger, A. Y. Tsai, and H. Saito. 1989. A family
of receptor-linked protein tyrosine phosphatases in humans and
Drosophila. Proc. Natl. Acad. Sci. USA 86:8698.

Birkeland, M. L., P. Johnson, I. S. Trowbridge, and E. Purze. 1989.
Changes in CD45 isoform expression accompany antigen- induced murine
T-cell activation. Proc. Natl. Acad. Sci. USA 86:6734.

Lefrancois, L. , and M. J. Bevan. 1985. Novel antigenic determinants
of the T200 glycoprotein expressed preferentially by activated
cytotoxic T lymphocytes. J. Immunol. 135:374.

Tung, J. S., M. P. Scheid, M. A. Pierotti, U. Hammerling, and E. A.
Boyse. 1981. Structural features and selective expression of three
Ly-5+ cell surface molecules. Immunogenetics 14:101.

Pulido, R., M. Cebrian, A. Acevedo, M. 0. de Landazuri, and F.
Sanchez-Madrid. 1988. Comparative biochemical and tissue
distribution study of four distinct CD45 antigen specificities. J.
Immunol. 140:3851.

Johnson, P., L. Greenbaum, R. Bottomly, and I. S. Trowbridge. 1989.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

38

Identification of the alternatively spliced exons of murine CD45
(T200) required for reactivity with B220 and other T200-restricted
antibodies. J. Exp. Med. 169:1179.

Lefrancois, L. , and T. Goodman. 1987. Developmental sequence of T200
antigen modifications in murine T cells. J. Immunol. 139:3718.

Streuli, M., C. Morimoto, M. Schrieber, S. F. Schlossman, and H.
Saito. 1988. Characterization of CD45 and CD45R monoclonal
antibodies using transfected mouse cell lines that express
individual human leukocyte common antigens. J. Immunol. 141:3910.

Takeuchi, T., C. E. Rudd, S. F. Schlossman, and C. Morimoto. 1987.
Induction of suppression following autologous mixed lymphocyte
reaction; role of a novel 2H4 antigen. Eur. J. Immunol. 17:97.

Bottomly, R., M. Luqman, L. Greenbaum, S. Carding, J. West, T.
Pasqualini, and.D. B. Murphy. 1989. A monoclonal antibody to murine
CD45R distinguishes CD4 T cell populations that produce different
cytokines. Eur. J. Immunol. 19:617.

Saiki, R. R., S. Scharf, F. Faloona, R. B. Mullis, G. T. Horn, H. A.
Erlich, and N. Arnheim. 1985. Enzymatic amplification of beta-globin
genomic sequences and restriction site analysis for diagnosis of
sickle cell anemia. Science 230:1350.

.Mullis, R. B., and F. A. Faloona. 1987. Specific synthesis of DNA in
vitro via a polymerase catalyzed chain reaction. Methods. Enzymol.
155:335.

Bell, J. 1989. The polymerase chain reaction” Immunol. Today
10(10):351.

Saiki, R. R., D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi,
G. T. Horn, R. B. Mullis, and H. A. Erlich. 1988. Primer-directed
enzymatic amplification of DNA with a thermostable DNA polymerase.
Science 239:487.

Higuchi, R., C. H. van Beroldingen, G. F. Sensabaugh, and H. A.
Erlich. 1988. DNA typing from single hairs. Nature 332:543.

Lench, N., P. Stanier, and.R. Williamson. 1988. Simple non-invasive
method to obtain DNA for gene analysis. Lancet. 1:1356.

Saag, M. S., B. H. Hahn, J. Gibbons, Y. X. Li, E. S. Parks, W. P.
Parks, and G. M. Shaw. 1988. Extensive variation of human
immunodeficiency virus type-l in vivo. Nature 334:440.

Bangham, C. R., S. Daenke, R. E. Phillips, J. R. Cruickshank, and.J.
I. Bell. 1988. Enzymatic amplification of exogenous and endogenous
retroviral sequences from DNA of patients with tropical spastic
paraparesis. EMBO J. 7:4179.

99.

100 .

101 .

102 .

39

Tindall, R. R. , and T. A. Runkel. 1988. Fidelity of DNA synthesis by
the Thermus aquaticus DNA polymerase. Biochemistry 27:6008.

Jeffreys, A. J., V. Wilson, R. Neumann, and J. Reyte. 1988.
Amplification of human minisatellites by the polymerase chain
reaction: towards DNA fingerprinting of single cells. Nucleic.
Acids. Res. 16:10953.

Rappolee, D. A., C. A. Brenner, R. Schultz, D. Mark, and Z. Werb.
1988. Developmental expression of PDGF, TGF-alpha, and TGF-beta
genes in preimplantation mouse embryos. Science 241:1823.

Rappolee, D. A., D. Mark, M. J. Bands, and Z. Werb. 1988. Wound
macrophages express TGF-alpha and other growth factors in vivo:
analysis by mRNA phenotyping. Science 241:708.

Chapter Two

T200 ALTERNATE EXON USE IN MURINE LYMPHOID CELLS DETERMINED BY

REVERSE TRANSCRIPTION-POLYMERASE CHAIN REACTION1

Hsun-Lang Chang, Michael H. Zaroukian, and Walter J. Esselman2

Departments of Microbiology and Public Health, and Medicine

Michigan State University, East Lansing, MI 48824-1101

Running Title: T200 Alternate Exon Use Determined By RT—PCR

40

FOOTNOTES
1 Supported in part by a grant from the National Institute of Health
(GM35774), and by a Michigan State University Collaborative Research

Grant .

2 Address correspondence to Dr. Walter J. Esselman, Dept of Microbiology
6: Public Health, Michigan State University, 146 Giltner Hall, East

Lansing , MI 48824-1101 .

Abbreviations: PCR, polymerase chain reaction; RT, reverse transcriptase;
RT—PCR, reverse transcription-polymerase chain reaction; bp, base-pairs;
Ex, exon; M—MLV, Moloney murine leukemia virus; P2, a synthetic
oligonucleotide that primes from a point beginning within Ex—2 of T200;
P9, a synthetic oligonucleotide that primes from a point beginning within

Ex-9 of T200 .

41

ABSTRACT

T200 glycoproteins of lymphoid and myeloid cells exhibit cell
lineage-specific structural heterogeneity; Peptide heterogeneity appears
to arise from alternate 5'-exon use (Ex-4, 5 and 6), potentially giving
rise to eight distinct forms of T200 mRNA containing 0-3 of these
alternate exons. A method is described for determining the number and
identity of the three alternate T200 exons expressed in cells using the
polymerase chain reaction (PCR) and the reverse transcription-polymerase
chain reaction (RT-PCR) without prior purification of RNA. Synthetic
primers flanking the alternate exon region of T200 were designed to yield
products for each possible exon combination having unique size and
restriction enzyme sites. PCR amplification of plasmids containing T200
cDNA with none (pLy-5-68) or all three (p7OZ/3-3) known alternate exons
resulted in the amplification of 186 and 603 base-pair (hp) products,
respectively. That the amplified products were derived from T200 cDNA was
verified by restriction enzyme mapping of each PCR product. T200 cDNA
prepared from cell lines utilizing no alternate exons (BW5147) or all
three exons (702/3.12) were analyzed by RT—PCR and contained amplified
products of 186 bp (zero alternate exons) and 603 bp (containing Ex-
4+5+6), respectively. RT-PCR of EL4 cells revealed approximately 186 and
330 bp products suggestive of zero and one alternate exon forms.
Restriction mapping confirmed that EL4 cells contained a zero-exon form
and a one-exon form containing Ex—5. Analysis of the 3B3 pre-B cell line
yielded 186, 330, 460 and 603 hp products; restriction mapping revealed
T200 mRNA for a zero alternate exon form, two distinct one— and two-exon
forms (Ex-4; Ex-5; Ex—4+5; Ex-5-l-6) , and a three-exon form (Ex-4+5+6).

Other lymphoid cell lines were heterogeneous in T200 alternate exon use,

42

43
with distinct patterns distinguishing B and T cells. RT-PCR can
facilitate the analysis of variations in T200 alternate exon use among

developmentally and functionally distinct lymphoid and myeloid cells.

INTRODUCTION

T200 (also called CD45, B220, Ly—5 and L—CA) is a molecularly
heterogenous family of transmembrane glycoproteins ubiquitously expressed
on the surface of all mammalian lymphoid and myeloid cells (1-3).
Heterogeneity is believed to result from variations in N—terminal peptide
sequences coded for by up to three alternately used exons, designated Ex-
4, 5 and 6 (Figure l), as well as by differences in glycosylation (4-8).

T200 has a number of unusual features that suggest an important role
for the glycoprotein in lymphoid cell function. T200 has a large (705 aa)
cytoplasmic domain (9) which is highly conserved (>90%) between mice and
humans (5,10) and is a substrate for protein kinase-C—mediated
phosphorylation (11). Recent structural and functional studies have
suggested a possible regulatory role for T200 in the phosphorylation state
of intracellular proteins. Specifically, the cytoplasmic domain of T200
has been shown to share significant sequence homology with placental
tyrosine phosphatase (12). In addition, significant protein tyrosine
phosphatase activity has been identified in purified preparations of T200
(13). These and other data suggest that the cytoplasmic domain of T200
may transduce or regulate the signal derived from cell surface
interactions involved in the generation of intracellular responses. The
N-terminal extracellular domain of T200 is more variable, suggesting that

this segment may regulate interactions with other cells or molecules,

44

Figure 1. Strategy for PCR amplification of the region of alternate exon
use in T200 cDNA. The complete T200 cDNA (top) includes the alternate
exon region (thick segment) and coding region (*-+ +—*). The expanded
first kb (middle) contains the three alternately used exons (Ex—4, 5 and
6) and shows the position and orientation of the flanking PCR primers P2
and P9. The 603 bp amplified product is shown (bottom), including the
three alternate exons. Exon positions are from Saga et a1. (7), and are
numbered according to Johnson et a1. (29). Base positions of cleavage

sites (474, 522) are given according to the T200 cDNA clone p702/3-3 (5).

 

 

1115

4 83m:

 

 

 

 

 

 

Zoom
8 meme v m m 580%
'TIITII: OMEDQE
8 I...
1.8 Egg
szm a m A o m w m m 9
“IITT: :
8 on «N we E a: we a He 290
u _ _ _ _
ll _ _ Tull-J. <29ng
mx m e m w e e

rte exon
.tema‘l
xpardel
l 5 if»
.mersli
.ing 35?
and all
leal'aie
.-3(fi.

46

thereby acting as a cell receptor for certain transmembrane signaling
processes. This possibility is supported by previous work in this
laboratory and others of a role for cell surface T200 in B cell
recognition/processing of macrophage signals (14), B cell proliferation
(15) and antibody responses (16), certain examples of CTL CMC (17,18), T
cell proliferation (19,20), NR-target cell binding (21—24), and the Ca2+-
dependent programming phase of NR CMC (25-27) . Since some epitope—
specific T200 mAb affect the function of certain lymphoid cell types but
not others (28), it seems likely that distinct epitopes on T200 are
involved in the function of T200 in different cell systems.

The degree to which cell lineage-specific heterogeneity is observed
in the T200 glycoprotein family remains to be determined, but appears to
occur despite a common origin from a single gene (5,10) , resulting in part
from alternate exon shuffling involving three 5' exons (Ex—4, 5 and 6
[7,29]; Figure l) . This process involves the synthesis of T200 containing
one of up to eight possible combinations of from 0—3 exons, with cells of
different origin often expressing different isoform phenotypes (4,6,30) .
Individual cell lines may also express more than one T200 isoform as a
result of alternate exon shuffling, possibly including major and minor
forms.

Specific determination of alternate exon usage has been reported for
some tumor and continuous lymphoid cell lines using transfer blot
techniques. Thomas et a1. (5) used a probe derived from cDNA containing
all three alternate exons to determine that murine B cells and some
suppressor T cell line T200 mRNA contained alternate exons, while helper
T cell lines and thymocytes did not. Similarly, a study of human T200

showed variable expression of T200 alternate exons in different human

47
tumor cell lines (6).

The polymerase chain reaction (PCR) is an important recent
development in recombinant DNA technology which allows single copy genes
from a few cells to be amplified to microgram quantities which may be
easily detected by conventional agarose gel electrophoresis without the
use of radioactive tracers (31—34). The amplified product is produced in
adequate amounts for subcloning or for sequence analysis and may be
directly gel-purified and evaluated by restriction enzyme analysis. In an
extension of PCR designated reverse transcription-PCR (RT—PCR), purified
mRNA may be analyzed by initial reverse transcription into cDNA which is
used as the template in a subsequent PCR (35). This method is extremely
sensitive and has been reported to be capable of revealing the specific
mRNA content of a single cell (35).

In this report, we describe a sensitive technique involving RT-PCR
to amplify the region of alternately-used exons of T200 which enables
identification of specific exon usage in lymphoid cells. This method does
not require purification of RNA prior to RT-PCR. The pattern of both

major and minor alternate exon usage of B and T cell lines is presented.

METHODS

WW 702/3.12. cm—z. D10. m. nwsm.

YAC—l, P388D1 and NIH 3T3 cells were obtained from the American Type
Culture Collection (Rockville, MD). 3B3 cells were obtained from Dr. R.
Brooks (Michigan State University, East Lansing, MI) and were grown
without interleukins as described (36). CTLL—2 and D10 cells were grown

with Gen A—activated rat spleen conditioned supernatants or with

48
recombinant IL—2 (Cetus Inc, Emeryville, CA). Radioiodination, Ly-5.l mAb
imunoprecipitation and SDS—PAGE were preformed as described previously
(24).

W Plasmids containing T200 cDNA having none of the
alternately-used exons (pLy-5-68) and bearing all three alternate exons
(p7OZ/3—3) were obtained from Drs. Y. Saga (10) and I. S. Trowbridge (5)
respectively .

W The PCR and RT-PCR primers were made in
the Macromolecule Synthesis Laboratory of Michigan State University by the
phosphoramidite method with an Applied Biosystems 380B automated DNA
synthesizer (Applied Biosystems, Foster City, CA). The P2 and P9 primers
used to amplify the region of alternate exon use in T200 cDNA are
described in Table I. Primer sequences were selected on the basis of the
known sequence of T200 cDNA as described previously (5) , and the expected
restriction enzyme fragment size was predicted from the sequence data.

W The PCR protocol was adapted from
methods previously described (31-34). The RT-PCR method used was a
modification of a method of Drs. J .L. Yang, J. McCormick and V. Maher
(personal communication). cDNA was prepared by addition of reverse
transcriptase (M—MLV, Bethesda Research Laboratories, Inc. , Gaithersburg,
MD) to a pellet of 1000 to 3000 cells in 50 p1 of 1X reverse transcription
buffer containing 2% Triton X-100, 5 pg BSA, 500 uM spermidine, 36 units
RNasin, 500 pM dNTPs and 0.5 pg anti-sense PCR oligonucleotide or
oligo(dT)l2-18 (Pharmacia LRB Biotechnology Inc., Piscataway, NJ) as the
RT primer. The mixture was incubated for 90 min at 37°C and one tenth Of
the resulting cDNA was amplified by 35 PCR cycles with 2.5 units Taq

polymerase, 400 ng sense and anti-sense primers and 500 pM dNTPs in 1X Taq

49

Table 1. PCR Primers Used to Amplify the Region of Alternate Exon Use in

 

T200 cDNA.
2'}. '1 Didi. '\ J. '1' _. oas‘ 90‘ .09 ‘_- , 0
P2 SENSE Ex-2 GGATCCCCTTCTGGACACAGAAGTCTTTGT (163-192‘,
BamHIb)
P9 ANTISENSE Ex—9 GAATTCACAGTAATGTTCCCAAACATGGC (765-737,
EcoRI)

 

a. Base position given according to cDNA clone p7OZ/3-3 (5).

b. Restriction sites were incorporated into the 5' end of the primers.

50
buffer (50 mM RC1, 10 mM Tris-HCl pH 9.3, 3 mM MgC12 and 0.1% w/v gelatin).
One PCR cycle consisted of incubations of l min at 94°C. 1 min at 50"C and
2 min at 72’C using a DNA Thermal Cycler (Perkin Elmer Cetus Inc.,
Norwalk, CT). The amplified products were analyzed by 2% agarose
(Bethesda Research Laboratories) electrophoresis and visualized with
ethidium bromide using a 123 ladder (Bethesda Research Laboratories) as

the size standard.

 

Amplified
fragments were gel-purified by electroelution in dialysis bags and
analyzed by restriction enzyme digestion. NciI (Boehringer Mannheim
Biochemicals, Inc. , Indianapolis, IN), BanI and Hian (New England Biolabs
Inc., Beverly, MA) were used according to the manufacturer's protocols.
The fragments were analyzed by 2% agarose electrophoresis as described

above .

RESULTS

Way—WWW. The synthetic

oligonucleotides, P2 (30 nucleotides, with a 5' BamHI site) and P9 (29
nucleotides, with a 5' EcoRI site) were prepared (Table I) which flanked
the region of the T200 cDNA containing the three known alternately—used
exons (Figure l; Ex-4, 5 and 6). These primers were designed to have
minimum possible mispriming sites and to yield PCR products for each of
the eight possible combinations of 0-3 exons having unique size and
restriction enzyme cleavage sites. PCR amplification of a plasmid
containing T200 cDNA having none of the alternately-used exons (pLy-5-68;

10), yielded a 186 bp product (Figure 2B), while a second plasmid

51
containing a distinct form of T200 cDNA hearing all three alternate exons
(p702/3-3; 5) resulted in the amplification of a 603 bp product (Figure
2H, depicted in Figure 1). These corresponded exactly to the predicted
sizes for the zero and three alternate exon forms, respectively.

We then

 

amplified T200 cDNA from 1000 to 3000 lymphoid cells using RT—PCR. The
method involved the use of viral reverse transcriptase (M—MLV) to make
cDNA copies from mRNA, with subsequent use of selective primers to amplify
T200 cDNA. cDNA was prepared from cell lines that were expected, on the
basis of T200 mAb immunoprecipitation and SDS-PAGE protein analysis, to
utilize none of the alternate exons (BW5147; Mr - 180 kDa, Figure 3C) , all
three alternate exons (702/3.12; M, - 220 kDa, Figure 3E), or multiple
combinations of alternate exons (EL4 and 3B3; Figure 3B and D). Reverse
transcriptase (M—MLV) was used with the PCR-primer (P9) as a RT primer and
added to a pellet of 1000 to 3000 cells. The PCR primers and Taq
polymerase were added after 90 min incubation and 35 cycles of the PCR
reaction commenced. After determining the critically important optimal
conditions of buffers, cycle temperature and time, purification of mRNA
was found to be unnecessary prior to production of T200 cDNA.

RT-PCR resulted in amplification of 186 hp products for BW5147 and
603 bp product for 702/3.12 (Figure 2D and C, respectively). RT—PCR of
EIA cells revealed approximately 186 and 330 bp products suggestive of
zero and one alternate exon forms (Figure 2E). RT-PCR of the 3B3 pre-B
cell line yielded amplified products of approximately 186, 330, 460 and
603 bp (Figure 2F) , consistent with the simultaneous presence in a single
cell line of T200 mRNA coding for zero, one, two and three alternate exon

forms, and raising the possibility of multiple forms within each alternate

52

Figure 2. PCR of T200 cDNA-containing plasmids pLy-5-68 and p7OZ/3-3 and
RT-PCR of cell lines. Ethidium bromide-stained agarose gel
electorphoresis of PCR amplification products of T200 cDNA in plasmids as
template; B, pLy-5-68 (no alternate exons); and H, p7OZ/3-3 (three
alternate exons). Results of RT-PCR with P2/P9 are for the following
cells: C, NIH 3T3 (negative control); D, BW5147 (T cell lymphoma); E, EL4
(T cell thymoma); F, 3B3 (B cell lymphoma); and G, 702/3.12 (pre-B

lymphoma). A size marker (123 ladder) is in lane A.

 

53

lflo.ol

Exons

base
pairs

O|5 _ 3
492 — 2
369 _ 1
246

- O
123

 

Figure2.

54

Figure 3. Immunoprecipitation and SDS-PAGE analysis of T200 glycoprotein.
Select cell lines used in this study were radioiodinated, lysed in
detergent and immunoprecipitated with anti-Ly-5.1 (B to E) or
unimmunoprecipated (A and F) for Mr analysis on an SDS-gradient
polyacrylamide gel. A, EL4 whole cell lysate; B, EL4 T200; C, BW5147
T200; D, 3B3 T200; E, 702/3.12 T200; F, whole cell lysate of

splenocytes. The position of size markers is given in kDa .

 

55

min. I
sed it ‘
E) at i
radian: ‘

573137

 

ate of

Figure 3.

56
exon group. NIH 3T3 cells (Figure 20) were used as a negative control

throughout the experiments.

 

332‘, That the amplified plasmid products were derived from T200 cDNA.was
verified by restriction enzyme digestion mapping of sites unique to each
PCR product. The details of the strategy for PCR and restriction enzyme
analysis of each amplified product are shown in Table II. The number and
nature of alternate exons expressed in plasmid T200 cDNA could be readily
determined from the predicted size of PCRramplified products before and
after restriction enzyme treatment with Neil (and other enzymes). P2/P9
primer-directed.PCRsamplified'products‘were gel-purified.and treateduwith
NciI, with the results shown in Figure 4. As predicted, the PCR product
derived from p702/3-3 T200 cDNA with all three alternate exons (Figure 4L)
was cleaved to fragments of 48, 243 and 312 base pairs (Figure 4R), while
the product from the pLy—5-68 T200 cDNA having none of the alternately—
used exons was digested with BanI but resisted digestion with Neil (data
not shown). Digestion with other restriction enzymes further confirmed
the identity of the amplified fragments (data not shown).

To determine and confirm the number and identity of the alternate
T200 exons used by each of several lymphoid cell types, restriction enzyme
mapping of the RT—PCRramplified products from 702/3.12, BW5147, EL—4 and
3B3 cells was carried out. The usefulness of the RT—PCR method in
producing and amplifying T200 cDNA was further confirmed by analysis of
383 cells which simultaneously contained T200 mRNA coding for zero, one,
two and three alternate exon forms of T200 cDNA. Each RT—PCR amplified
product of 3B3 cells was gel-purified and further characterized by size

and NciI restriction enzyme mapping (Figure 4, Table II).

57

Table II. Analysis of alternative exon use and identification of RT—PCR

 

 

 

products'.
Alternate Product lciI Fragments Alternate Exon Observed
Exon; (hp) Predigteg §y5147 4514 333b 7oz
sx-3+s+s 303 48, 233, 312 ++c +++
Ex-4+5 462 150, 312 ++
sx-4+s 453 213. 243
Ex-5+6 474 4a, 133, 243 ++
sx-s 315 313d +0
sx-s 333 130, 133 ++ ++
Ex-B 327 84, 233
none 1861 136d +++ ++ ++

 

a. Data are typical of at least two experiments with each cell line.

b. Electrophoresis data are shown in Figure 4.

c. Expression of alternate exon forms was determined by the size of RT-PCR
products and NciI digested fragment(s), which were identical to
those predicted from sequence data, within the resolution of the
technique (i 10 bp). Rey: +++, major form; ++, moderately expressed
form; +, minor form. Blank spaces indicate that these forms were
not observed or identified.

d. No change.

e. A.very small amount of the one-exon form resisted digestion by NciI and
was shown to contain Ex-4 by Hian digestion.

f. The form without alternative exons was further identified by BanI

digestion.

58

Figure 4. Neil and BanI restriction enzyme mapping of gel-purified
products of RT-PCR of T200 alternate exon forms in 3B3 cells. Pairs of
digested and undigested fragments are as follows: B, zero-exon form
treated with BanI; C, zero-exon form: D, one-exon form treated with Neil;
E, one-exon form: C, two-exon form treated with NciI: H, two-exon form:
I, three-exon form treated with Neil; J, three-exon form: R, p702/3-3
product treated with Neil; L, PCR-amplified plasmid p702/3-3 product.
The position of the faint, 48 bp fragment in Lanes G, I and R is
indicated. Size standards, A, F and M. Fragment sizes are given in Table

II.

 

base

pairs

59

C’)
N

 

00
V

Figure4,

60

Digestion of the 3B3 three alternate exon form (Figure 4J) produced
fragments (Figure 41) which were indistinguishable in size to those
produced by digestion of the p7OZ/3-3 product, indicating the presence of
one three-exon form, Ex-4+5+6 (Table II). In contrast, digestion of the
two exon form (Figure 4H) produced five fragments (Figure 46) which were
indistinguishable in size to those predicted for the digestion of a
mixture of two distinct two-exon forms containing Ex-4-l-5 and Ex-5+6.
Digestion of the one-exon form (Figure 4E) produced two fragments (Figure
4D) which matched the size of those predicted for the digestion of a
product containing Ex-5, along with a very small amount of Neil-resistant
material consistent with a form containing Ex-4 (further confirmed by
Hian digestion). Finally, the zero-exon form of 333 cells (Figure 4C)
was resistant to Neil digestion (data not shown) and was digested with
BanI into fragments which agreed with the predicted size (59 and 127 bp)
for the zero-exon form (Figure 4B). The BanI site is created at the
junction between Ex-3 and 7 when Ex-4, 5 and 6 are not present. Hence,
multiple forms for some of the alternate exon groups were observed, with
six of the eight possible T200 alternate exon forms detected. Only the
Ex-6 one-exon form and the Ex-4+6 two-exon form were not apparent.

The gel-purified zero alternate exon product of BW5147 cells (186
bp) resisted digestion with Neil but was cleaved into two appropriate size
fragments by BanI which were indistinguishable to those shown in Figure
4B. Similarly, the amplified product of 702/3.l2 cells was cleaved by
Neil to fragments having sizes consistent with the expression of all three
alternate exons as shown in Figure 4R and I.

Restriction mapping of gel-purified RT-PCR products from EL4 cells

(from Figure 2E) revealed a zero alternate exon form; a one-exon form (Ex-

61

5) and very small amounts of a two exon form which could not be gel-
purified in amounts adequate for further characterization.

WWW After
demonstrating that the combination of RT-PCR and restriction enzyme
analysis was suitable for precise identification of the alternate exon
expression of T200 cDNA, we undertook to estimate the degree of
heterogeneity of T200 alternate exon use in a variety of lymphoid and
myeloid cells (Figure 5). T cell lines predominantly expressed the zero-
exon form as well as smaller amounts of one-exon forms (Lanes D, F-J).
Among T cell lines, BW5147 cells (D) almost exclusively expressed the
zero-exon form, while YAC-l cells (F) also expressed a small amount of an
intermediate form (390 bp). This intermediate form was also observed in
other T cell lines (Lanes G-I), P815 mastocytoma cells (C) and the P388D1
macrophage cell line (B). The zero alternate exon T200 isoform also
predominated in P815 and P388D1 cell lines. While 702/3.12 cells exhibited
predominant expression of the three exon form, smaller forms, including a
zero-exon form, similar to those in 3B3 cells were evident in small
amounts.

DISCUSSION

The method described herein permitted selective amplification of a
small amount of mRNA coding for a portion of a complex lymphoid cell
surface glycoprotein that is generated by a process involving alternative
exon shuffling. The amplified products were identified by standard
agarose electrophoresis and individually mapped by restriction enzymes.
This method is a significant advance of PCR technology and to our

knowledge is the first report describing the amplification of mRNA

62

Figure 5. Survey of T200 alternate exon use in lymphoid and.myeloid.cells
as demonstrated by RT-PCR. The cells analyzed were as follows: B, P388D1
(macrophage line); C, P815 (mastocytoma); D, BW5147 (T cell lymphoma); F,
YAC-l (T cell lymphoma); G, D10 (Tkz‘helper cell line); H, HT-2 (T cell
line); I, CTLL-2 (cytotoxic T cell line); J, EL-4 (T cell thymoma); L, 383
(B cell lymphoma); M, 702/3.12 (pre-B lymphoma); A, E, and R, size

standards (123 ladder).

 

63

mcoXm ,

$0.075

1.

0.0.3.3..-

 

.m o.:m_m

mm—

cum

00m

mow
m_o

mt_aa
omen

64
containing alternately used exons without prior purification of RNA. This
makes RT-PCR more sensitive and direct in the analysis of mRNA derived
from small numbers of cells.

The features of this protocol that enabled direct cDNA production
from mRNA without prior purification include the addition of reverse
transcriptase directly to cell pellets (1000 to 3000 cells) with
detergent, nucleotides and buffer. An aliquot of the resultant cDNA was
then added to Taq polymerase, PCR primers, and a specific buffer system
for PCR amplification. Other reagents and conditions that appear to
contribute to the success of direct.mRNA phenotyping using RT-PCR include
the use of a pH of 9.3 for the PCR reaction, the use of primers
approximately 30 nucleotides in length, and optimization of reaction
temperatures and number of cycles. We observed that RT-PCR proceeded only
in a small range around these optimal conditions, and expect that these
reaction conditions may require adjustment for different template and
primer systems. In addition, the use of one of the PCR primers for RT
initiation proved superior to the use of oligo-dT (data not shown).
Primer sequences were designed in positions which have no more than five
matches out of ten 3' nucleotides and no homology in the last few 3'
nucleotides in order to eliminate mispriming artifacts. The primers were
also designed with 5' restriction sites for subsequent subcloning.

RT-PCR.was chosen.for these studies because no other suitable method
was available for determining T200 alternate exon usage in small numbers
of cells. RT-PCR permits amplification of mRNA/cDNA from a few hundred
cells, enabling visualization on ordinary agarose gels. This capability
was felt to be important to advance current understanding regarding the

relationship between T200 structural heterogeneity and its possible role

65

in the differentiation and function of various lymphoid and myeloid cells.
We found mRNA phenotyping to be superior to Northern blot transfer
analysis in.the determination of T200 alternate exon.expression (data.not
shown). T200 mRNA is large (5.5 kb), which can impair the stability and
resolution of closely related isoforms. In this report, the RT-PCR method
identified and distinguished apparent major and minor T200 alternate exon
isoforms as well as complex mixtures, while such isoforms could not be
clearly distinguished in Northern blots. In addition, the products of PCR
are obtained in sufficient quantity for detailed analysis by restriction
mapping, direct sequencing and subcloning.

Although analysis of lymphoid cell lines revealed considerable
heterogeneity in T200 alternate exon use, several predictable patterns
were observed, As predicted from SDS-PAGE analysis, cells expressing the
lowest M, (180 kDa) isoform.of T200 glycoprotein (e.g., BW5147) generally
showed a predominance of the zero alternate exon T200 isoform by RT-PCR
analysis, while cells expressing the highest M, isoform (B cell lines)
expressed the highest alternate exon form (Ex-4+5+6). The pattern of
alternate exon use was more complex in cell lines expressing several
intermediate surface T200 glycoprotein isoforms (190 to 215 kDa). For
example, EL4 cells expressingflboth one-exon.end zero-exon forms by RT-PCR
exhibited several T200 glycoprotein isoforms by SDS-PAGE, with M, of 190-
210 kDa (Figures 2, 3 and 5).

Correlation of T200 glycoprotein expression and alternate exon use
in the 383 B cell line was even more complex. These cells exhibited.mRNA
yielding PCR products of zero- to three-exons and protein isoforms with M,
of 200-220 kDa. It is proposed that these higher M, protein isoforms

contained proteins translated from all of the alternate exon products.

66

The apparent M, of a particular T200 isoform as determined by SDS-PAGE may
have been more a function of the extent and complexity of glycosylation
than of the number of alternate exons present. This possibility is
supported by the finding that T200 isolated from B cells grown in the
presence of tunicamycin had an apparent M, of only 190 kDa, compared to 220
kDa in the absence of tunicamycin (8). Similarly, thymocyte T200 from
tunicamycin-treated cells had an apparent M, of only 160 kDa, while
untreated thymocyte surface T200 had a M, of 180 kDa. Consequently, for
the 220 kDa isoform of B cell T200 using all three alternate exons,
glycosylation accounts for about 30 kDa (14%) of apparent M, while each
alternate exon would only account for about 5 kDa (2%) of apparent 11,. If
only peptide structural differences accounted for apparent M, differences
observed on SDS-PAGE, RT-PCR results for 383 cells would have predicted
four T200 isoforms. That only higher M, forms were observed suggests that
the lower alternate exon isoforms are extensively glycosylated resulting
in an apparent M, between 200 and 220 kDa. Despite the apparent existence
of multiple alternate exon isoforms, the major T200 glycoprotein expressed
on 3B3 cells is a 220 kDa form. The M, heterogeneity of T200 observed by
SDS-PAGE is therefore contributed significantly to by variations in
glycosylation (glycoforms), which may obscure minor differences due to
alternate exon usage (protein isoforms). It is also possible that
alternate exon use directs the extent and complexity of T200
glycosylation.

An alternative explanation for the above phenomena which requires
further study is that not all of the mRNA isoforms are translated into
cell surface T200 glycoproteins. It is also possible that unequal

amplification of cDNA representing different alternate exon isoforms

67
occurred after reverse transcription, leading to the disproportionate
appearance of minor forms. Since the precise relationship between the
number of copies of'mRNA for a particular isoform.and the intensity of the
band seen after RT-PCR is not known, we are unable to estimate the degree
to which unequal amplification of cDNA contributed to variations in the
amounts of final products produced in these experiments.

The observation. of’ small amounts of intermediate size RT-PCR
products (Figure 5) requires further study. These products (e.g. 390 bp,
Figure SB,C,F) do not correspond exactly to the predicted.sizes for RT-PCR
products of known alternate exon isoforms of T200 (approximately 330 or
460 bp). Of particular interest is the possibility that this intermediate
form may include an as yet unproven additional alternate exon as suggested
for a macrophage cell line by Saga et a1. (7). The presence in trace
amounts of this intermediate form in most of the T cell lines studied by
RT-PCR may not have been previously identifiable using other less
sensitive methods. Similarly, the occurrence of a minor product at
approximately 115 bp in some cell lines (Figure 5B,C,F) also suggests the
possible lack of an additional alternately used exon of approximately 70
bp (possibly Ex-7). The faint bands of lowest size are believed to
represent excess primers. Mispriming artifacts may also be responsible
for the generation of products of intermediate size. Sequence analysis
should allow resolution of these possibilities.

The significance of the multiple T200 isoforms occurring in a
particular cell line remains to be determined. The possibility that each
mRNA isoform is translated into a functional member of the T200 membrane
glycoprotein family further suggests that T200 alternate exon usage

contributes to the generation of individual T200 isoforms which may have

68
distinct roles in the function of individual lymphoid cells. RT-PCR will
be useful in determining T200 isoform expression in small numbers of
lymphoid cells or even individual lymphocytes (35). The general method
will also be useful in the study of mRNA and protein expression in a wide
variety of other systems involving immune cell differentiation and

function.

W. The authors wish to thank Drs. V. Maher, J.L.
Yang, and J. McCormick, for initially suggesting the RT-PCR protocol. We
also wish to thank J. Dodgson and R. Schwartz for their invaluable advice
and discussion during this project. M. Morrison and H. Beittenmiller

provided expert technical assistance.

LIST OF REFERENCES

10.

ll.

LISTS 0F REFERENCES

Omary, M. B., I. S. Trowbridge, and H. A. Battifora. 1980. Human
homologue of murine T200 glycoprotein. J.. Exp. Med. 152:842.

Scheid, M.P., and D. Triglia. 1979. Further description of the Ly-S
System. Immunogenet 9:423.

Judd, W., C. A. Poodry , S. Broder, S. M. Friedman, L. Chess, and
J.. L. Strominger. 1980. High molecular weight antigens present on
human T cells. Proc. Natl. Acad. Sci. USA 77:6805.

Saga,'Y., J. S. Tung, F. W. Shen, and E. A, Boyse. 1987. Alternative
use of'5' exons in the specification.of'Ly-5 isoforms distinguishing
hematopoietic cell lineages. Proc. Natl. Acad. Sci. USA 84:5364.

Thomas, M. L., P. J. Reynolds, A. Chain, Y. Ben-Neriah, and I. S.
Trowbridge. 1987. B-cell variant of mouse T200 (Ly-5): evidence for
alternative mRNA splicing. Proc. Natl. Acad. Sci. USA 84:5360.

Streuli, M., L. R. Hall, Y. Saga, S. F. Schlossman, and H. Saito.
1987. Differential usage of three exons generates at least five

different mRNAs encoding human leukocyte common antigens. J. Exp.
Med. 166:1548.

Saga, Y., J. S. Tung, F. W. Shen, T. C. Pancoast, and E. A. Boyse.
1988. Organization of the Ly-S gene. Molec. Cell. Biol. 8:4889.

Tung, J. S., M. C. Deere, and E. A. Boyse. 1984. Evidence that Ly-S
product of T and B cells differ in protein structure. Immunogenet.
19:149.

Thomas, M. L., A. N. Barclay, J. Gagnon, and A. F. Williams. 1985.
Evidence from cDNA clones that the rat leukocyte-common antigen

(T200) spans the lipid'bilayer and contains a cytoplasmic domain of
80,000 Mr. Cell 41:83.

Saga, Y., J. S. Tung, F. W. Shen, and E. A. Boyse. 1986. Sequences
of Ly-S cDNA: isoform-related diversity of Ly-S mRNA. Proc. Natl.
Acad. Sci. USA 83:6940.

Shakelford, D. A., and I. S. Trowbridge. 1986. Identification of
lymphocyte integral membrane proteins as substrates for protein
kinase C: phosphorylation of the interleukin-2 receptor, class I HLA
antigens, and T200 glycoprotein. J. Biol. Chem. 261:8334.

69

12.

13.

14.

15.

l6.

17.

18.

19.

20.

21.

22.

23.

24.

70

Charbonneau, H., N. R. Tanks, R. A. Walsh, and E. H. Fischer. 1988.
The leukocyte common antigen (CD45): a putative receptor-linked
protein tyrosine phosphatase. Proc. Natl. Acad. Sci. USA. 85:7182.

Tonks, N. R., H. Charbonneau, C. D. Diltz, E. H. Fischer, R. A.
Walsh. 1989. Demonstration that the leukocyte common antigen CD45 is

a receptor-linked protein tyrosine phosphatase. Biochemistry
27:8695.

Yakura, H., F. Shen, E. Bourcet, and E. A. Boyse. 1983. On the
function of Ly-S in the regulation of antigen-driven B cell
differentiation: comparison and contrast with Lyb-2. J. Exp. Med.
157:1077.

Mittler, R. S., R. S. Greenfield, B. Z. Schacter, N. F. Richard, and
M. R. Hoffman. 1987. Antibodies to the common leukocyte antigen
(T200) inhibit an early phase in the activation of resting human B
cells. J. Immunol. 138:3159.

Yakura, H., I. Rawabata, F. W. Shen, and M. Ratagiri. 1986.
Selective inhibition of lipopolysaccharide-induced polyclonal IgG
response by monoclonal Ly-5 antibody. J. Immunol. 136:2729.

Nakayama, E., H. Shiku, E. Stokart, H. F. Oettgen, and L. J. Old.
1979. Cytotoxic T cells: Lyt phenotype and blocking of killing
activity by Lyt antisera. Proc. Natl. Acad. Sci. USA. 76:1977.

Nakayama, E. 1982. Blocking of effector cell cytotoxicity and T cell
proliferation by Lyt antisera. Immunol. Rev. 68:117.

Ledbetter, I. A., L. M. Rose, C. E. Spooner, P. G. Beatty, P. J.
Martin, and E. A. Clark. 1985. Antibodies to comon leukocyte
antigen p220 influence human T cell proliferation by modifying IL-2
receptor expression. J. Immunol. 135:1819.

Martorell, J., R. Vilella, L. Borche, I. Rojo, and J. Vives. 1987.
A second signal for T cell mitogenesis provided by monoclonal
antibodies CD45 (T200). Eur. J. Immunol. 17:1447.

Pollack, S.B., M. R. Tam, R. C. Nowinski, and S. L. Emmons. 1979.
Presence of T cell associated surface antigens on murine NR cells.
J. Immunol. 123:1818.

Seaman, W.E. , N. Talal, L. A. Herzenberg, and J. A. Ledbetter. 1981.
Surface antigens on mouse natural killer cells: use of monoclonal
antibodies to inhibit or to enrich cytotoxic activity. J. Immunol.
127:982.

Zaroukian, M. H., C. W. Gilbert, and W. J. Esselman. 1986. Surface
Ly-5 glycoprotein in murine natural killer cell development, target
binding, and cytotoxicity. Immunol. Invest. 15:813.

Gilbert, C. W., M. H. Zaroukian, and W. J. Esselman. 1988. Poly-N-
acetyllactosamine structures on murine cell surface T200

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

71

glycoprotein participate in NR cell binding to YAC-l targets. J.
Immunol. 140:2821.

Targan, S.R., and W. Newman. 1983. Definition of a ”trigger" stage
in the NR cytolytic reaction sequence by a monoclonal antibody to
the glycoprotein T-200. J. Immunol. 131:1149.

Newman, W., L. D. Fast L. D, and L. M. Rose. 1983. Blockade of NR
cell lysis is a property of monoclonal antibodies that bind to
distinct regions of T-200. J. Immunol. 131:1742.

Newman,‘W., S. R. Targan, and L” D. Fast. 1984. Immunobiological and
immunochemical aspects of the T-200 family of glycoproteins. Mol.
Immunol. 21:1113.

Starling, G.C., S. E. Davidson, J. L. McKenzie, and D. N. J. Hart.
1987. Inhibiton of natural killer-cell mediated cytolysis with
monoclonal antibodies to restricted and non-restricted epitopes of
the leucocyte common antigen. Immunology 61:351.

Johnson, N. J., C. M. Meyer, J. T. Pingel, and M. L. Thomas. 1989.
Sequence conservation in potential regulatory regions of the mouse
and human leukocyte common antigen gene. J. Biol. Chem. 264: in
press.

Ralph, S. J., M. L. Thomas, C. C. Morton, and I. S. Trowbridge.
1987. Structural variants of human T200 glycoprotein (leukocyte-
common antigen). EMBO J. 6:1251.

Mullis, R.B., and F..A. Faloona. 1987. Specific synthesis of DNA in
‘vitro via a polymerase catalyzed chain reaction. Meth. in Enzymol.
Part F 155:335.

Saiki, R.R., D. H. Gelfand, S. Stoffel, S. J. Scharf, R, Higuchi, G.
T. Horn, R. B. Mullis, and H. A. Erlich. 1988. Primer-directed
enzymatic amplification of DNA with a thermostable DNA polymerase.
Science 239:487.

McMahon G., E. Davis, and G.N. Wogan. 1987. Characterization of C-
Ri-ras oncogene alleles by direct sequencing of’ enzymatically

amplified DNA from carcinogen- induced tumors. Proc. Natl. Acad. Sci.
USA 84:4974.

Wrischnik, A. L., R. G. Higuchi, M. Stoneking, H. A. Erlich, N.
Arnheim, and. A. C. Wilson. 1987. Length mutations in. human
mitochondria DNA: direct sequencing of enzymatically amplified.DNA.
Nucl. Acids Res. 15:529.

Rappolee, D.A., D. Mark, M. J. Bands, and Z. Werb. 1988. Wound
macrophages express TGF-a and other growth factors in 'vivo:
analysis by mRNA phenotyping. Science 241:708.

Brooks, R., D. Yuan, J. W. Uhr, P. H. Rrammer, and E. S. Vitetta.

72

1983. Lymphokine-induced IgM secretion by clones of neoplastic B
cells. Nature 302:825.

Chapter Three
DEVELOPMENTAL EXPRESSION OF CD45 ALTERNATE EXONS IN MURINE T CELLS:

EVIDENCE OF ADDITIONAL ALTERNATE EXON USE1

Hsun-Lang Chang, Leo Lefrancois, Michael H. Zaroukian,

and Walter J. Esselman2

Departments of Microbiology and Public Health, and Medicine
Michigan State University, East Lansing, MI 48824-1101

and the Upjohn Corporation, Kalamazoo MI 49001

Running Title: CD45 Alternate Exon Use

73

FOOTNOTES
1 Supported in part by a grant from the National Institutes of Health (GM-
35774), a Michigan State University Biomedical Research Support Grant and

an All-University Research Initiation Grant.

2 Address correspondence to: Dr. W. J. Esselman, Department of Microbiology
& Public Health, Michigan State University, 339 Giltner Hall, East

Lansing, MI 48824-1101.

3 Abbreviations: bp, base-pairs; Ex, exon; LCA, leukocyte-comon antigen
(one of several alternative designations for CD45); IEL, intraepithelial
lymphocytes; LN, lymph node; M-MLV, Moloney murine leukemia virus; PCR,
polymerase chain reaction; P2, P4, P5, P6, and P9, synthetic
oligonucleotide primers from a point beginning within CD45 exons 2, 4, 5,
6 and 9 respectively; RT, reverse transcriptase; RT-PCR, reverse

transcription-polymerase chain reaction.

74

ABSTRACT

The CD45 family of lymphoid and myeloid cell surface glycoproteins
exhibits cell lineage-associated.structural heterogeneity arising in part
from alternate 5'-exon shuffling. Previous studies from other
investigators about exons involved in the final glycoprotein structure
have provided evidence of alternate exon use only for exons 4, 5 and 6.
However, our prior RT-PCR data on CD45 alternate exon use implied the
presence of additional alternate exons. We employed the reverse
transcription-polymerase chain reaction (RT-PCR), Southernglotti g using
exon-specific or exon splice junction-specific oligonucleotide probes, and
direct DNA sequencing of RT-PCR.products to provide evidence that exons 7
and 8 are also alternately used. In order to estimate the relative amount
of mRNA for each CD45 isoform present in lymphocytes, the effect of
template length on RT-PCR amplification efficiency was tested by using
radiolabeled primers and quantitating products at intervals of five PCR
cycles. Smaller CD45 cDNA isoforms amplified more efficiently than larger
ones when RT-PCR exceeded 25 cycles, but the amplification was
proportional when PCR 'was limited to 20-25 cycles. Under these
conditions, examination of CD45 isoform expression revealed that Stage I
thymocytes (CDATCD8') expressed only trace amounts of minus-one [Ex(-1)]
and minus-two exon [Ex(-2)] isoforms, with no other isoforms detected, A
Ex(-2) isoform was also detected in all thymocytes and T cells analyzed
but only in trace amounts. Stage II thymocytes (CD4ICD8I) expressed high
and approximately equal amounts of Ex(-l) and zero alternate exon [Ex(0)]
isoforms, with minor quantities of one exon [Ex(l)] and two exon [Ex(2)]
isoforms. Among stage III thymocytes, both CD4+CD8" and CD4'CD8+ cells

expressed significant quantities of only the Ex(-l) and Ex(0) isoforms.

75

76
Comparison of CD45 alternate exon use in resting CD4+ and CD8* lymph node
T cells revealed evidence of divergent exon use, with CD8+ cells expressing
detectible quantities of an Ex(2) isoform, while producing proportionately
less of the Ex(-1) and Ex(0) isoforms. Examination of allogeneically
activated T cells revealed that the CD4+ BC-3 helper T cell clone expressed
less of the Ex(l) isoform, while the CD8+ 8.2.2 CTL clone increased its
production of higher alternate exon isoforms, including Ex(2) and Ex(3)
isoforms. Our evidence of the alternate use of exons 7 and 8 suggests
that at least five alternate exons exist in the CD45 glycoprotein family.
Analysis of isoform expression among thymocytes and T cells suggests that
shuffling of CD45 alternate exons occurs in an organized and predictable
sequence during the process of T cell maturation and activation.
INTRODUCTION

The CD45 family of lymphoid and myeloid cell membrane-associated
glycoproteins (also termed LCA, T200, and Ly-S) has attracted considerable
interest recently as a potential participant and regulator of cellular
activation and proliferation (1-3). Molecular heterogeneity in the CD45
glycoprotein family is known to result both from variations in the N-
terminal extracellular peptide region and in glycosylation (4-10).
Protein heterogeneity has been attributed to variable incorporation of
N-terminal peptide sequences coded for by up to three alternately
transcribed exons, designated Ex-4, 5 and 6 (Figure l; 4, 6). These exons
share in common a large number of potential O-linked glycosylation sites
(11, 12) suggesting that alternate exon use directs the nature and extent
of glycosylation of the extramembranous portion of CD45.

The constant cytoplasmic domain of CD45 possesses two tandemly

repeating subdomains homologous to placental protein tyrosine phosphatase-

77

Figure 1. Illustration of the amplified region of CD45 and the relative
position of RT-PCR primers and probes for Southern hybridization. The Ex-
8/Ex-3 junction-specific probe is not shown. The thick line represents
the alternate exon region and exons 7 and 8. Based upon published
sequences (6), NciI has one cleavage site in each of exons 5 and 6
(nucleotide no. 473 and 521); Hian has one cleavage site in each of exon

4 and 5 (nucleotide no. 287 and 392).

78

A 22m:

 

 

:51
2 j 3 .33....
:uZ
1. mm choc;
on. .Y tun. .552;
3 E «A .3 A: 8. E 35 :83
A TTITIIILI-IIJ. “ J.
o m n o n v m m n: coxm

 

 

_. u A “ “ll/3.2% 38
e. n v m a _ o

79

1B (PTPase-1B; 13). The confirmation of PTPase activity in human CD45
(14) has prompted models for CD45 function in which the cytoplasmic region
is proposed to affect transmembrane signaling through the
dephosphorylation of critical tyrosyl residues on other membrane-
associated proteins (1, 2). The cytoplasmic region of CD45 appears to be
responsible for its biological function in modulating transmembrane
signaling for cellular activation and proliferation. It has been proposed
that the physical interaction of CD45 with other membrane-associated
proteins to facilitate CD45/PTPase-mediated dephosphorylation reactions is
determined by its extracellular N-terminal structure (15). This
hypothesis was supported by the evidence for cell type-specific
heterogeneity of CD45 expression (5, 7, 15, 16) and by evidence for
modulation of Ca2+ flux by cross-linking antibodies (1) . Further knowledge
of CD45 isoform expression in specified lymphocyte populations may provide
clues to the structure-function relationships in the CD45 glycoprotein
family.

Murine CD45 (Ly-5) alternate exon use in individual cell populations
has been studied using a limited number of characterized mAbs (17)
specific for peptides coded by Ex-4 (14.8) and Ex-5 (C363.16A). Human
CD45 has been studied using mAbs with specificity for certain of the
alternate exons, as well as a mAb specific for the CD45 isoform lacking
exons 4, 5 and 6 (UCHLl; 18). The use of specific mAbs has allowed
identification of cellular subsets expressing individual exons but is
limited in the recognition of CD45 isoforms containing multiple exons.
This limitation is also true for Northern hybridization analysis using
exon-specific probes. In addition, it is possible that variations in

glycosylation of some CD45 isoforms could prevent antibody binding, thus

80
rendering mAb reactivity analyses unreliable.

We recently showed RT-PCR to be a sensitive and reliable alternative
in the characterization of CD45 isoform expression in murine lymphoid cell
lines (9). In the same report, we observed amplified products whose sizes
on agarose electrophoresis suggested that additional alternately used
exons may exist in the CD45 glycoprotein family. In this report, we
present evidence that Ex-7 and Ex-8 are additional CD45 alternate exons,
and that lymphocytes of the T cell lineage generally include increasing
numbers of alternate exons during the processes of cellular maturation and
activation.

METHODS

Mgngglgng1_§n§1b2gie§. Monoclonal clonal antibodies used in this
study were 3.168 (anti-CD8; l9), AD4 (anti-CD8; 20), and RL172 (anti-CD4;
21).

W 702/3.12 cells were obtained from the American
Type Culture Collection.(Rockville, MD). 3B3 cells were Obtained from Dr.
R. Brooks (Michigan State University). subsets of thymocytes and
peripheral lymph node T cells were obtained from 8-10 week-old C57BL/6J
mice by a previously described method (22). Cells were at least 95% pure
with respect to a given phenotype. The alloreactive CD8+ CTL clone, 8.2.2,
was derived by limiting dilution from a mixed lymphocyte culture (MLC)
generated by reacting CS7BL/6J splenocytes with irradiated DBA/2
splenocytes. The alloreactive CD4*'T cell clone, BC-3, was Obtained from
an MLC containing Balb/c splenocytes and irradiated C57BL/6J splenocytes.
IEL*were isolated as previously described (23). Briefly, small intestines
from 4-8 C57BL/6J mice were cut into 5 mm pieces and stirred at 37’C in

HBSS containing 0.1 mM EDTA. Supernatant suspensions containing

81
lymphocytes and epithelial cells were then subjected to centrifugation
through a 44/67.5 percent Percoll gradient. Cells at the interface were
collected and further purified by panning with an anti-CD8 mAb to a final
purity of 88-95% C08+ cells.

T lymphocytes used for RT-PCR were washed twice in PBS, with 30,000
cells transferred into a 500 pl tube, pelleted and supernatants decanted.
Cell pellets were stored at -20°C and thawed just before adding the RT
reagents. Pellets thus prepared provided reproducible results for RT-PCR
for at least 9 months.

WM. The RT, PCR and sequencing primers, and
exon-specific oligonucleotide probes (Figure 1 and Table I) were made in
the Macromolecular Synthesis Facility at Michigan State University. For
5'-end labeling, 40 pmol of oligonucleotides were added to a final volume
of 50 p1 of 1X kinase buffer with 20 U T4 DNA polynucleotide kinase (New
England Biolab. , Beverly, MD) and 300 pCi of 1[32P]-ATP (ICN, Irvine, CA).
This mixture was incubated at 37°C for 45 min. The unincorporated
nucleotides were removed using a G-25 Sephadex (Pharmacia, Piscataway, NJ)
spun column.

WWW. The RT-PCR protocol
was performed as described previously (9) with modifications. Briefly,
cDNA was prepared by adding 200 units M-MLV reverse transcriptase (BRL,
Gaitherburg, MD) to a pellet of 30,000 cells in 20 pl of 1X RT buffer
containing 2% Triton X-100, 2 pg BSA (BRL), 200 pM spermidine (Sigma, St.
Louis, MO), 36 U RNasin (Promega, Madison, WI), 200 pg dNTPs (BMB,
Indianapolis, IN), and 0.2 pg unlabeled anti-sense PCR primer (P9) as the
RT primer. The mixture was incubated 60 min at 37'C. Approximately 5 #1

of the reaction mixture was used directly as template for amplification in

82

Table I. Oligonucleotides used for amplification of the alternate exon

region in CD45 cDNA and the probes used for Southern hybridization.

 

 

OLigonucleotide Orientation Sequence (base position)
P2 Sense CTTTGGATCCGCCCTTCTGGACACAGAAGT(157-186‘)
P9 Antisense GGCGGAATTCACACTAATGTTCCCAAACAT(769-740‘)
P3 Sense AACACCTACACCCAGTGATG(202-221‘)
J8-3 Antisense GCTTCGTTGTGCTAGCATCACTGGG(725-711+221-212‘)

 

‘Base position given according to cDNA clone p7OZ/3-3 (6).

83
a total volume of 50 pl containing 5' end-labeled primers (P2 and P9) in
addition to the regular PCR components (9). The reactionwwas performed.in
a DNA Thermal Cycler (Perkin Elmer Cetus, Norwalk, CT) for the number of
cycles specified. Each cycle consisted of 40 sec at 94'C for
denaturation, 15 sec at 55°C for annealing and 30 sec at 72°C for
elongation. The first cycle was preceded by a 5 min incubation at 94'C
and the last followed by 6 min at 72°C. Approximately one-fifth of the
total amount of each amplified. product was analyzed. by 8% native
polyacrylamide gel electrophoresis (PAGE) and 'visualized either by
autoradiography or ethidium bromide staining. For determination of
amplification efficiency, eight identical RT-PCR reactions were set up,
with one reaction stopped at the end of every 5 PCR cycles. After PAGE
and autoradiography, each band was excised and the radioactivity
determined by scintillation counting.

For isolation of RT-PCR fragments after autoradiography or direct
viewing under UV light, the desired fragment was excised and placed in
dialysis tubing for electroelution. The eluted DNA was further
concentrated using a Centricon-30 tube (Amicon, Danvers, MA). In some
cases, eluted DNA was used as template in subsequent PCR.

Eggyma;ig_§nalysi§‘ The exon composition of purified products was
identified by restriction enzyme digestion using NciI, Hian, and BanI as
previously described (9). After digestion, the fragments were separated
by 8% PAGE and visualized by autoradiography.

WW Purified DNA fragments were separated using
a 1.8% agarose gel and.blotted onto a nylon membrane (GeneScreen, DuPont,
Boston, MA) in 10X SSPE buffer. After baking under vacuum, the membrane

was pro-hybridized at 42°C for three hours in 6X SSPE buffer containing

84

10X Denhardt's solution, 1‘ SDS, and denatured salmon sperm DNA (Sigma) at
50 pg per ml. The membrane was then cut to generate two identical
membranes. One of the membranes was hybridized overnight with Ex-3-
specific probe (P3) , the other with Ex-8/Ex-3 junction-specific probe (J8-
3), both at 2x106 cpm/ml in 6X SSPE, 1% SDS at their respective
hybridization temperature (T3), 57’C and 73'C. The hybridized membranes
were washed twice in 2X SSPE, 1% SDS at room temperature for 20 min and
once in 1 X SSPE, 1% SDS at T5 for 5 min.

WW About 0.5 us of the DNA wee
sequenced according to the protocols by Dr. L. M. Latinwo (Michigan State
University, personal communication). All the sequencing reagents were
purchased from United States Biochemical Corporation (Cleveland, OH).
Briefly, template DNA with 1 pmol 32F end-labeled P9 was boiled 7 min in
10 pl 1X reaction buffer and chilled on ice 5 min. Then the following
were added: 1 p1 0.1M DTT, 2 p1 labeling mix, 1 pl Mn2+ buffer, and 2 pl
of Sequenase at 1:4 dilution. Immediately after mixing, 3.5 p1 aliquots
were pipetted into pro-warmed tubes containing 2.5 pl termination mixture.
Tubes were then incubated at 37’C for 5 min and the reaction stopped by
the addition of 4 p1 stop buffer. An aliquot of 3 p1 of the final product
was denatured at 85°C for 5 min, chilled on ice and electrophoresed in 6%
sequencing gel.

RESULTS

WWW. Previous studies
have provided evidence for CD45 alternate exon use involving only the
translated exons 4, 5 and 6 (4, 6), as well as the untranslated Ex-la and
lb (4, 24). However, other observations using RT-PCR have indicated the

presence of intermediate CD45 RT-PCR products (9) , raising the possibility

85

that additional alternate exons may exist. This possibility was addressed
by analysis of CD4+ lymph node T cells from C57BL/6J mice using RT-PCR,
Southern hybridization with exon- or junction-specific oligonucleotide
probes and direct sequencing of individual bands purified by PAGE.

RT-PCR analysis using P2 and P9 as primers (Figure 2) revealed a
prominent "zero" exon isoform as conventionally defined (i.e. , lacking Ex-
4, 5 and 6). CD4+ lymph node T cells also expressed very low level of one
exon isoform. Two additional products having sizes of about 124 and 100
bp were also apparent. The size of the 124 bp product was consistent with
a zero-exon isoform that additionally lacked Ex-7 [i.e., a ”minus-one"
exon isoform; Ex(-l)], while the 100 bp product was of a size consistent
with a "zero" exon isoform that also lacked both Ex-7 and Ex-8 [i.e., a
"minus-two" exon isoform; Ex(-2)]. Southern hybridization analysis of
gel-purified RT-PCR products using probes specific for Ex-3 and the Ex-
8/Ex-3 junction supported this interpretation (Figure 3) . The Ex-3-
specific probe confirmed the CD45 identity of RT-PCR products, since it
interacts with an exon.ef CD45 that is not subject to alternative splicing
and was always contained within the sequences amplified by the primers
used for these experiments (P2; P9). The Ex-3-specific probe reacteduwith
each amplified band (Figure 3B, lanes 1-3), but not with the negative
control sample (lane S), thereby confirming the CD45 origin of each RT-PCR
product. The Ex-8/Ex-3 junction-specific probe reacted only with the RT-
PCR product immediately below that of the size expected for the Ex(O)
isoform of CD45 (Figure 3C, lane 2). The size of this product (124 bp)
was as predicted for an isoform lacking Ex-4, 5, 6 and 7. This finding
strongly suggested that the product was amplified from a CD45 cDNA species

that excised all exons between Ex-3 and Ex-8, including Ex-7.

86

Figure 2. RT-PCR pattern of CD4+ LN T-cells showing the presence of the

124 bp and 100 bp isoforms.

Lane S, 123 marker; lane 1, CD4+ LN T cells.

 

 

 

.._.

87

Isol'orm

hp. '

 

369--

Ex.(l)
246 -.

l - Ex.(0)

123 -. - EX.("])
" Ex.(-2)

S I

Figure 2.

88

Figure 3. Southern hybridization analysis of the Ex(0) isoform, and the
124 bp and 100 bp CD45 isoforms of CD4+ lymph node T cells. Lane 1, 100
bp fragment; lane 2, 124 bp fragment; lane 3, Ex(0) form; and lane S, 123
size marker. (A). Electrophoresis of the re-amplified products, stained
with ethidium bromide. (B). Southern hybridization with the Ex-3-specific
probe, P3. (C). Southern hybridization with the Ex-8/Ex-3 junction-

specific probe, J8-3.

 

89

wmm

_

.m 9.3m;

mmm—

 

90

To further confirm these results, the 124 bp CD45 RT-PCR.product was
isolated by electroelution and subjected to direct sequence analysis
(Figure 4). The result confirmed that the last nucleotide of Ex-3
(nucleotide 221) was followed.immediately by the first nucleotide of Ex-8
(nucleotide 711). Thus, the 124 base-pair product resulted from direct
splicing of Ex-3 to Ex-8, thereby excluding Ex-7.

The size of the smallest RT-PCR fragment (100 bp) was as predicted
for a CD45 isoform lacking all the exons between Ex-3 and Ex-9. The
reactivity of this product with the Ex-3 specific probe confirmed.its CD45
identity, while its appropriate size and failure to bind the Ex-8/Ex-3
junction probe strongly suggested that the product represented an isoform
of CD45 that also lacks Ex-8. The very low level of expression of this
product in cells studied thus far has complicated efforts to further
confirm the absence of Ex-8 by direct sequence analysis.

QD45__amplifigatign__efifi1§1engxl Estimation of the relative
abundance of the 'predominant CD45 isoforms expressed. in. lymphocyte
populations was accomplished using radiolabeled.primers and comparison of
the radioactivity of products derived from RT-PCR. The reliability of
this approach required experimental conditions in which each CD45 isoform
could be shown to be amplified with comparable efficiency.

The relationship between CD45 cDNA template size and. RT-PCR
amplification efficiency was analyzed in C57BL/6J CD4+ thymocytes. Use of
32P-labeled primers enabled quantitation of RT-PCR products after every 5
cycles (Figures 5 and 6). Carrying out RT-PCR to 40 cycles resulted in
clear evidence of unequal amplification of the CD45 cDNA, with smaller
templates [Ex(-1) and Ex(0)] amplified approximately 10-fold more than

thelarger Ex(l) template. Amplification of this larger template appeared

91

Figure 4. Sequencing data showing the absence of exon 7 in the 124 bp

fragment isolated from CD4+ lymph node T cells. Exon 3 is fused directly

to exon 8. The nucleotide numbers are shown according to the sequence

published by Trowbridge, et al. (6).

92

Exon ACGT Nucleotide No.

 

 

-— 16

2
._ _ 197
*r— — 198

3
_221
— '— 731

8
— 745

 

Figure 4.

93

Figure 5. Autoradiography of the RT-PCR pattern of CD4+CD8' thymocytes at
5 PCR cycle intervals. RT-PCR was stopped at the end of 15th, 20th, 25th,

30th, 35th, and 40th cycle (lanes 1-6, respectively).

94

Isoiorm

Ex.(l)

—- - - - EX.(0)

- C . Ex.(-l)

Figure 5.

95

Figure 6. RT-PCR amplification curves for CD45 isoforms from CD4+CD8'
thymocytes. Radioactivity (cpm) of the RT-PCR products was plotted
against the cycle number at which the reaction was stopped. Ex(l) form,

circle; Ex(O) form, triangle; and Ex(-l) form, square.

96

 

1o , , g

Counts/ min

 

 

 

 

Number of PCR cycles

Figure 6.

97

to reach saturation at approximately 25 RT-PCR cycles. However, when PCR
was limited to 25 cycles or less, each isoform was amplified at comparable
rates independent of template size (Figure 6). As a result, RT-PCR was
limited to 24 cycles or less to allowrband.intensity to reflect the amount
of CD45 cDNA for each isoform present in the original cell population
studied. We observed similar results for several lymphoid cell lines
bearing all known CD45 alternate exon isoforms, and in PCR experiments
using mixtures of the plasmids p7OZ/3 (6) and pLy5-68 (5) as templates
(data not shown).

W. In
view of the evidence that CD45 alternate exon use differs in lymphocytes
of varying lineage (15, 22), and that such differences may affect the
function of this cell surface glycoprotein (25), we employed RT-PCR to
analyze CD45 isoform expression among T cells at different differentiation
stages and activation states. CS7BL/6J cell populations studied included:
1) immature thymocytes (Stage I: CD4‘CD8‘; Stage II: CD4*CD8+); 2) mature
thymocytes (Stage III: CD4ICD8', CD4'CD8"); 3) lymph node T cells. In
addition, alloreactive CD4*.and CD8*‘T-cell clones and IEL were examined.

Figure 7A compares CD45 alternate exon use in immature thymocyte
populations. Stage I (CD4'CD8’) thymocytes (lane 1) expressed only low
levels of Ex(-l) and Ex(-2) isoforms. Stage II (CD4ICD8*) thymocytes (lane
2) expressed much higher amounts of the Ex(-1) isoform, along with even
more predominant expression of the Ex(O) isoform and small amounts of Ex(-
2), Ex(l) and Ex(2) isoforms. Among mature (Stage III) thymocytes (Figure
7B, lanes 1 and 2), both CD4JCD8' and CD4'CD8+ thymocytes were comparable
in their expression of Ex(O) and Ex(-1) isoforms, with higher M, isoforms

not apparent. However, additional alternate exon use patterns were

98

Figure 7. CD45 RT-PCR pattern of T cell subsets and alloreactive T cell
clones. (A). Immature thymocytes: lane S, 123 size marker; lane 1, Stage
I (CD4'CD8') thymocytes; lane 2, Stage II (CD4+CD8+) thymocytes. (B).
Mature (Stage III) thymocytes and peripheral lymph node T cells: lane S,
123 size marker; lane 1, CD4+ thymocytes; lane 2, CD8+ thymocytes; lane 3,
CD4+ lymph node T cells, lane 4, CD81' lymph node T cells. (C).
alloreactive T cell clones: lane S, 123 size marker; lane 1, Th clone BC-3;
lane 2, T, clone 8.2.2. (D). IEL and control B cells: lane 1, 702/3; lane

2, 383; lane 3, IEL (CD8+).

 

 

 

 

99

K 050:...

 

 

 

 

A In a a F a _ m :15. . A.
an. A. - .x i
A—IV.XW A— v W '|l E-
3.. .im
oniw ' i :1 .im 1.. ‘i Swim --.Il -
u p w
. Sim Eu- Sim .1. . gleam A.
“Sim ’ I, - Elam i
.Zim ' AS. xm .
«Sim 1. III Anviw I! - onxw'
5.6 l 11 Ed - H Ema .#
mm aim
«a
e
.5862 . .5202 l... . Etoaom. .. . - . 8.20210. .1."
iii M.,.” ”dd“ 0
m m M O a O O O a m m
9 E Z 8 .7 8 .7 8 .7 .o .
.9 / 9 .9 .9 .9 .9 .9 o o
m 8 O 0
.1 8. 8.
0:05 Z._ 9: 2:

D
O
m
<

100

observed among CD4+CD8' and CD4'CD8" T cells recovered from lymph nodes.
CD4*CD8' lymph node T cells continued to express Ex(O) and Ex(-l) isoforms,
but also produced significant amounts of a Ex(l) isoform (Figure 7B, lane
3). CD4'ICD8+ lymph node T cells produced all of these same isoforms, but
additionally expressed an Ex(2) isoform.

RT-PCR analysis of the alloreactive T cell clones BC-3 (CD4*) and
8.2.2 (CD8*) , revealed additional significant differences in alternate exon
use (Figure 7C). CD4? BC-3 cells retained the CD45 phenotype of CD4FCD8’
lymph node T cells (Figure 7C, lane 1), with Ex(O) and Ex(-l) isoforms
predominating and a smaller amount of an Ex(l) isoform also present. In
contrast, CD8+ 8.2.2 cells expressed significant amounts of Ex(-l), Ex(0),
Ex(l), Ex(2) and.Ex(3) isoforms. CD45 alternate exon use among IEL (CD8*)
was also compared to that of the T cell populations described above. IEL
predominantly expressed Ex(0), Ex(l) and Ex(2) CD45 isoforms (Figure 7D,
lane 3), but also produced small amounts of Ex(-l) and Ex(3) isoforms. In
general, all T cells examined thus far have expressed.the Ex(-1) and Ex(-
2) isoforms, although the quantities have varied significantly.

MW In order to determine
which of the three previously identified alternate CD45 exons (exons 4, 5
and 6) is predominantly expressed in the Ex(l) isoforms identified in
these experiments, Ex(l) isoforms were purified from the RT-PCR.mmplified
products of CD4+ and CD8+ lymph node T cells, the CTL clone 8.2.2, and IEL,
followed by PCR re-amplification using 32P-labeled primers to yield
quantities necessary for restriction enzyme analysis. In these
experiments, P9 was labeled to a higher specific activity so that the
resultant fragments containing the P9 sequence could be identified by

their higher radioactivities following re-amplification and digestion.

101

Enzymes with unique CD45 alternate exon-specific digestion.patterns
(NciI, Hian, and BanI) were used to digest the Ex(l) isoforms from CD4+
lymph node T cells (Figure 8, lanes 1-3), CD8+ lymph node T cells (lanes
4-6), 8.2.2 cells (lanes 7-9), and IEL (data not shown). The digestion
pattern for all of these T cells was very similar. After NciI digestion,
the appearance of 156 bp and 187 bp bands (lanes 2, 5 and 8) indicated the
principal use of Ex-5. However, weaker bands of 249 bp and 88 bp, and.the
residual undigested fragment suggested the minor presence of Ex-6 and.Ex-
4, respectively. The dominant production of 237 bp and 106 bp fragments
following Hian digestion (lanes 3, 6 and 9), was also consistent with the
primary expression of Ex-5, with additional bands indicating the presence
of minor amounts of Ex(l) isoforms containing Ex-4 (95 bp; 130 bp) and.Ex-
6 (undigested). In order to determine if any of the undigested Ex(l)
forms (containing either Ex-4 or Ex-6) represented unknown isoforms or
non-specific products, we digested Ex(l) isoforms with BanI (cleaving at
a site created at the junctions of Ex-4/Ex-7, Ex-5/Ex-7 and Ex-6/Ex-7).
The results indicated no other isoforms or non-specific products, as all
fragment sizes were as predicted for Ex(l) isoforms containing Ex-4, Ex-S
or Ex-6.

DISCUSSION

The present studies established the existence of additional
alternately used exons in the murine CD45 glycoprotein family. From
earlier investigations, only exons 4, 5 and 6 (as well as the untranslated
exons, 1a and 1b) were known to be subject to alternate exon shuffling
during the production of CD45 mRNA transcripts in lymphocytes (4, 6). The
current data regarding the size of RT-PCR amplified products from murine

thymocytes and T cells, as well as their reactivities with exon-specific

 

102

Figure 8. Restriction enzyme analysis of CD45 Ex(l) isoforms purified
from CD4+ lymph node T cells (lanes 1-3), CD8+ lymph node T cells (lanes
4-6), and the T, clone 8.2.2 (lanes 7-9). Lane 8, 123 size marker; lanes
1, 4 and 7, undigested Ex(l) form (control); lanes 2, 5, and 8, NciI

digestion; lanes 3, 6, and 9, Hinfl digestion.

103

CDA'L N CDB‘L N CDB‘CIone

 

 

369 —‘
246 —

 

Figure 8.

104
oligonucleotide probes strongly suggested.that both.Ex-7 and.Ex-8 are also
alternately used. Direct sequencing data definitively confirmed the
alternate use of Ex-7 and the creation of a splice junction between Ex-3
and Ex-8. These data indicated that at least five alternate exons exist
in the CD45 glycoprotein family.

Analysis of CD45 isoform expression revealed a general trend toward
incorporation of additional alternate exons into the CD45 mRNA of T cells
as they matured and became more differentiated or activated (as summarized
in Figure 9). An exception to this tendency was apparent in a CD4+
allogenic Th clone, BC-3, which retained expression of the smaller CD45
isoforms after they were subject to activating stimuli (e.g., IL-2). CD4'
CD8' thymocytes (Stage I), expressed only low levels of Ex(-1) and Ex(-2)
isoforms. CD4+CD8+ thymocytes (Stage II), CD4+CD8' and CD4'CD8" thymocytes
(Stage III), showed much higher amounts of Ex(0) and Ex(-1) isoforms.

Additional but distinctive alternate exon inclusion patterns were
observed among CD4” and CD8+ lymph node T cells. While both cell
populations produced larger Ex(l) isoforms, CD8+ cells also expressed even
larger Ex(2) isoforms.

In activated T cell populations, further differences were observed.
Following allogenic stimulation, the CD8+ alloreactive CTL clone 8.2.2
expressed detectable quantities of isoforms containing from none [Ex(-2)]
to all five [Ex(3)] of the alternate exons recognized in these
experiments. Principal among these were the Ex(0), Ex(l) and Ex(2)
isoforms. In contrast, CD45 isoform expression in.the CD4J helper T cell
clone BC-3 revealed less of the Ex(l) isoform seen in CD4+ lymph node T
cells, and a preponderance of Ex(-1) and Ex(0) isoforms that were

prevalent in CD41 thymocytes. Hence, the pattern of CD45 alternate exon

105

Figure 9. Summary of developmental alterations of CD45 isoform expression
in murine T cells. The arrow indicates the progression of maturation or

activation.

106

lllllfllllll

._m_ a
mzoso op
+80.er

 

A~.v
..-e
no a
A .c
Au..
:3

 

 

A«.v
A P..

Ace
.pqu

 

 

 

mzomo E
.8930

 

._._mo.... 2..
+8930

3 ..

. I

3 .

. P ..

.~“%xm

 

 

3...
2....
.o.

111H|11111j

.p_.xm

 

 

 

some.» 2..
.8930

T

mica—Ext
+8930

a...
Z...
.0.
.pzm

1111]

 

 

.N..
2....
.o.
.p.au

llllll

 

 

 

nicest:
-8930

/
\

.m we:m_m

.N;
:1.

.o.

. —.
.mnqu

 

 

 

3.602;:
+3930

a;
.7qu

 

 

IL

mtooszt.
bop—«DO

107
use that emerges during activation of CD4+ and CD8+ T cells may be
divergent, with CD8+ T cells expressing additional CD45 alternate exons,
while CD4+ T cells may
instead delete certain alternate exons expressed more prominently in a
less activated state.

The observation that, at high numbers of PCR.cycles, smaller RT-PCR
products were amplified. more efficiently' than larger ones made it
important to identify conditions in which rates of amplification of each
product were similar; The achievement of this goal in the present studies
using radiolabeled primers and 24 cycles of PCR allowed estimation of the
relative amount of mRNA present for each of the CD45 isoforms observed in
a given cell population. While the Ex(-2) isoform of CD45 lacking all
known alternate exons (Ex-4 to Ex-8) never predominated, the Ex(-l)
isoform containing only Ex-8 was abundant in several of the cell
populations studied.

Understanding variations in CD45 alternate exon use in cells of
differing lineage, differentiation and activation states has taken on
additional significance with recent evidence indicating that the highly
conserved cytoplasmic portion of CD45 contains two PTPase domains (13,
26). .Along with evidence of physiologically relevant membrane-associated
phosphoprotein substrates for CD45 PTPase (e.g., p56“*; 27, 28), these
findings suggest that CD45 may play an.important role in the regulation of
lymphocyte activation and proliferation. The variably inhibitory and
stimulatory effects observed upon treatment of different lymphocyte
populations with antibodies directed against one or more epitopes of CD45
(29-32) underscore the complexity of the role of CD45 in lymphocyte

responses to external stimuli. The employment of different combinations

108

of CD45 alternate exons in the extracellular portion of this major
membrane glycoprotein may determine the ligand(s) with which CD45
interacts and the stimuli to which the lymphocyte responds. While
dephosphorylation of p561“ by CD45 could be expected to result in cellular
activation (27, 28), the PTPase activity of CD45 likely results in the
dephosphorylation of proteins phosphorylated by protein tyrosine kinases
and may thereby down-regulate these cellular activation signals.

RT-PCR data for CD45 isoform expression in lymphocyte populations in
this study generally agreed with earlier SDS-PAGE analyses of CD45
imunoprecipitates from identical cell populations in the same murine
strain (15, 22). Such analyses have demonstrated a 235 kDa glycoprotein
isoform of CD45 in CD81 lymph node T cells not present in CD8+ thymocytes.
Similarly, a new CD45 glycoprotein isoform of 240 kDa was noted in
activated CD8+ CTL clones; whereas comparing allogenically activated CD4+
clones to CD4+ lymph node T cells revealed decreased expression of the high
molecular weight isoform (220 kDa) . Although it is theoretically possible
that expression of higher or lower M, CD45 isoforms resulted from changes
in glycosylation rather than alternate exon use, the general correlation
of RT-PCR and SDS-PAGE data suggests that the pattern of alternate exon
use accurately predicts the Mr of CD45 isoforms observed on SDS-PAGE.
Correlation of CD45 mRNA sizes with the sizes of nascent proteins
determined by 35S-methionine pulse-chase and endo-H treatment methods also
supports the interpretation that increases in the size of CD45 mRNA
produced by the use of additional alternate exons accounts for the
emergence of higher M, glycoprotein isoforms (22, 33).

While others have also reported similar losses of high M, CD45

isoforms (CD45R) in CD4+ T cells after allogenic stimulation, this has also

109

been observed in CD8+ T cells (34). A similar finding has been made for
human lymph node T cells (35) . These disparate findings may be due to the
use of mAb to infer the presence or absence of CD45 alternate exons
without full knowledge of the epitopes recognized by these mAb. Further,
increased O-glycosylation associated with activation may block mAb binding
to alternate exon epitopes. Support for this possibility was demonstrated
in CTL expressing highly glycosylated CD45 isoforms (M, up to 260 kDa; 15,
36) , with associated unique, O-glycan-associated carbohydrate antigens
(CTl; CT2).

RT-PCR revealed differences in alternate exon use between Stage I
(CD4'CD8') and Stage II (CD4+CD8+) thymocytes not suggested by previous
SDS-PAGE analyses (15, 22). This discrepancy may be explained by
relatively minor differences in peptide length among the smallest CD45
alternate exon isoforms and the comparatively large effect of
glycosylation on apparent M, by SDS-PAGE.

Restriction enzyme analysis was used to determine whether any of the
previously recognized alternate exons (Ex-4, 5 or 6) predominated when a
single additional alternate exon was incorporated into the Ex(O) isoform.
Regardless of lymphocyte population studied, Ex(l) isoforms principally
involved insertion of Ex-S. Other Ex(l) isoforms containing Ex-4 or Ex-6
were present only in minor amounts. For lymph node T cells, this finding
was consistent with conclusions based on mAb labeling studies (34).

In addition to evidence for alternate use of Ex-7 and Ex-8 in the
CD45 glycoprotein family, our data indicated that a CD45 isoform lacking
all alternate exons from Ex-4 through Ex-7 is significantly expressed in
various members of the T lymphocyte lineage, including thymocytes, lymph

node T cells and CD4+ cloned helper T cells. However, only trace amounts

110

of this isoform were detected in the 8.2.2 CTL.clone, with no significant
expression in IEL. The as yet undetermined glycosylation.pattern of this
Ex(-l) isoform may eventually help to explain why it has not been
identified as a discrete isoform by SDS-PAGE analysis of CD45
immunoprecipitates.

While data were presented.that CD45 can.a1so alternately exclude Ex-
8, evidence of such exclusion was only seen in the context of exclusion of
all known alternate exons (Ex-4 to 8), and then only in extremely small
amounts. Alternate use of exons 7 and 8 may not have been previously
recognized in part because only a small number of CD45 cDNAs have been
directly sequenced. Further, other investigators using PCR to study CD45
isoform expression have selected primers in Ex-3 and Ex-7 (37), thereby
only allowing for amplification of sequences contained within these exons.

To-date, we have found.Ex-7 to be alternately excluded.on1y when.Ex-
4, 5 and 6 were also absent. No isoforms lacking Ex-7 but containing Ex-
4, 5 or 6 have yet been identified by Southern blotting or restriction
enzyme analysis. Even.considering the high sensitivity of the PCR method,
we cannot entirely preclude the possibility that other minor CD45 isoforms
may exist which alternately exclude Ex-7. However, the unique size of Ex-
7 (74 bases) and the lack of evidence of major, intermediate size PCR
products would seem to indicate that if any such isoforms exist, they are
likely to be present only in minute quantities. Nevertheless, evidence
exists to support the theoretical possibility that Ex-7 and Ex-8 may be
used independently, since a study of the expression of CD45 mini-gene
constructs revealed that the absence of Ex-3 and Ex-7 did not affect
tissue-specific splicing of exons 4-6 (38).

Acknowledgements. The authors wish to thank Drs. Latinwo and Jackson for

111

their help in DNA sequencing, and Dr. Schwartz for his precious advice.

LIST OF REFERENCES

7‘

10.

11.

LIST OF REFERENCES

Clark, E. A., and J. A. Ledbetter. 1989. Leukocyte cell surface
enzymology: CD45 (LCA, T200) is a protein tyrosine phosphatase.
Immunol. Today 10(7):225.

Hunter, T. 1989. Protein-tyrosine phosphatases: the other side of
the coin. Cell 58:1013.

Rudd, C. E., P. Anderson, C. Morimoto, M. Streuli, and S. F.
Schlossman. 1989. Molecular interactions, T-cell subsets and a role
of the CD4/CD8zp561ck complex in human T cell activation. Immunol.
Rev. 111:225.

Saga, Y., J. S. Tung, F. W. Shen, T. C. Pancoast, and E. A. Boyse.
1988. Organization of the Ly-5 gene. Molec. Cell. Biol. 8:4889.

Saga, Y., J. S. Tung, F.-W. Shen, and E. A. Boyse. 1987. Alternative
use of 5' exons in the specification of Ly-5 isoforms distinguishing
hematopoietic cell lineages. Proc. Natl. Acad. Sci. USA 84:5364.

Thomas, M. L., P. J. Reynolds, A. Chain, Y. Ben-Neriah, and I. S.
Trowbridge. 1987. B-cell variant of mouse T200 (Ly-5): evidence for
alternative mRNA splicing. Proc. Natl. Acad. Sci. USA 84:5360.

Streuli, M., L. R. Hall, Y. Saga, 8. F. Schlossman, and H. Saito.
1987. Differential usage of three exons generates at least five

different mRNAs encoding human leukocyte common antigens. J. Exp.
Med. 166:1548.

Tung, J. S., M. C. Deere, and E. A. Boyse. 1984. Evidence that Ly-5
product of T and B cells differ in protein structure. Immunogenetics
19:149.

Chang, H.-L., M. H. Zaroukian, and W. J. Esselman. 1989. T200
alternate exon use in murine lymphoid cells determined by reverse
transcription-polymerase chain reaction. J. Immunol. 143:315.

Gilbert, C. W., M. H. Zaroukian, and W. J. Esselman. 1988. Poly-N-
acetyllactosamine structures on murine cell surface T200

glycoprotein participate in NR cell binding to YAC-l targets. J.
Immunol. 140:2821.

Davis, C. C., A. Elhammer, D. W. Russell, W. J. Schneider, S.
Rornfield, M. S. Brown, and J. L. Goldstein. 1986. Deletion of

112

12.

13.

14.

15.

l6.

17.

18.

19.

20.

21.

22.

113

clustered O-linked carbohydrates does not impair function of low
density lipoprotein receptor in transfected fibroblasts. J. Biol.
Chem. 261:2828.

Barclay, A. N., D. I. Jackson, A. C. Willis, and A. F. Williams.
1987. Lymphocyte specific heterogeneity in the rat leucocyte common

antigen (T200) is due to differences in polypeptide sequences near
the NH2-terminus. EMBO J. 6:1259.

Charbonneau, H., N. R. Tonks, R. A. Walsh, and E. H. Fischer. 1988.
The leukocyte common antigen (CD45): a putative receptor-linked
protein tyrosine phosphatase. Proc. Natl. Acad. Sci. USA 85:7182.

Tonks, N. R., H. Charbonneau, C. D. Diltz, E. H. Fischer, and R. A.
Walsh. 1988. Demonstration that the leukocyte common antigen CD45 is
a protein tyrosine phosphatase. Biochemistry 27:8695.

Thomas, M. L., and L. Lefrancois. 1988. Differential expression of
the leucocyte-common antigen family. Immunol. Today 9(10):320.

Ralph, S. J. , M. L. Thomas, C. C. Morton, and I. S. Trowbridge.
1987 . Structural variants of human T200 glycoprote in
(leukocyte-common antigen). EMBO J. 6:1251.

Johnson, P., L. Greenbaum, R. Bottomly, and I. S. Trowbridge. 1989.
Identification of the alternatively spliced exons of murine CD45
(T200) required for reactivity with B220 and other T200-restricted
antibodies. J. Exp. Med. 169:1179.

Pulido, R., M. Cebrian, A. Acevedo, M. 0. de Landazuri, and F.
Sanchez-Madrid. 1988. Comparative biochemical and tissue
distribution study of four distinct CD45 antigen specificities. J.
Immunol. 140:3851.

Sarmiento, M., A. L. Glasebrook, and F. W. Fitch. 1980. IgG or IgM
monoclonal antibodies reactive with different determinants on the
molecular complex bearing Lyt 2 antigen block T cell-mediated
cytolysis in the absence of complement. J. Immunol. 125:2665.

Tung, J. S., M. P. Scheid, and M. A. Palladino. 1983. Different
forms of Ly-5 within the T-cell lineage. Immunogenetics 17:649.

Ceredig, R. , J. W. Lowenthal, M. Nabholz, and H. R. MacDonald. 1985.
Expression of interleukin-2 receptors as a differentiation marker on
intrathymic stem cells. Nature 314:98.

Lefrancois, L. , and T. Goodman. 1987. Developmental sequence of T200

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

114
antigen modifications in murine T cells. J. Immunol. 139:3718.

Goodman, T., and L. Lefrancois. 1988. Expression of the gamma-delta
T-cell receptor an intestinal CD8+ intraepithelial lymphocytes.
Nature 333:855.

Hall, L. R., M. Streuli, S. F. Schlossman, and H. Saito. 1988.
Complete exon- intron organization of the human leukocyte canon
antigen (CD45) gene. J. Immunol. 141:2781.

Lefrancois, L., L. Puddington, C. E. Machamer, and M. J. Bevan.
1985. Acquisition of cytotoxic T lymphocyte-specific carbohydrate
differentiation antigens. J. Exp. Med. 162:1275.

Maddox, J. F., C. R. Mackay, and M. R. Brandon. 1985. The sheep
analogue of leucocyte common antigen (LCA). Immunology 55:347.

Ostergaard, H. L., and I. S. Trowbridge. 1990. Coclustering CD45
with CD4 or CD8 alters the phosphorylation and kinase activity of
p561ck. J. Exp. Med. 172:347.

Ostergaard, H. L., D. A. Shackelford, T. R. Hurley, P. Johnson, R.
Hyman, B. M. Sefton, and I. S. Trowbridge. 1989. Expression of CD45
alters phosphorylation of the lck-encaded tyrosine protein kinase in
murine lymphoma T-cell lines. Proc. Natl. Acad. Sci. USA 86:8959.

Mittler, R. S., R. S. Greenfield, B. Z. Schacter, N. F. Richard, and
.M. R. Hoffman. 1987. Antibodies to the common leukocyte antigen
(T200) inhibit an early phase in the activation of resting human B
cells. J. Immunol. 138:3159.

Yakura, H., F. Shen, E. Bourcet, and E. A. Boyse. 1983. On the
function of Ly-S in the regulation of antigen-driven B cell
differentiation: comparison and contrast with Lyb-2. J. Exp. Med.
157:1077.

Nakayama, E. 1982. Blocking of effector cell cytotoxicity and T cell
proliferation by Lyt antisera. Immunol. Rev. 68:117.

Lefrancois, L., and M. J. Bevan. 1985. Functional modifications of
cytotoxic T-lymphocyte T200 glycoprotein recognized by monoclonal
antibodies . Nature 314 : 449 .

Lefrancois, L., M. L. Thomas, M. J. Bevan, and I. S. Trowbridge.
1986. Different classes of T lymphocytes have different mRNAs for
the leukocyte-common antigen, T200. J. Exp. Med. 163:1337.

34.

35.

36.

37.

38.

115

Birkeland, M. L., P. Johnson, I. S. Trowbridge, and E. Purze. 1989.
Changes in CD45 isoform expression accompany antigen- induced murine
T-cell activation. Proc. Natl. Acad. Sci. USA 86:6734.

Serra, H. M., M. J. Mant, B. A. Ruether, J. A. Ledbetter, and L. M.
Pilarski. 1988. Selective loss of CD4+ CD45R+ T cells in peripheral
blood of multiple myeloma patients. J. Clin. Immunol. 8:259.

Thomas, G. C., H. L. Chang, W. J. Esselman, R. LeCorre, and L.
Lefrancois. 1990. Characterization of the CD45 molecule on murine
intestinal intraepithelial lymphocytes. J. Immunol. (in press).

Saga, Y., J. S. Lee, C. Saraiya, and E. A. Boyse. 1990. Regulation
of alternative splicing in the generation of isoforms of the mouse
Ly-5 (CD45) glycoprotein. Proc. Natl. Acad. Sci. USA 87:3728.

Streuli, M. , and H. Saito. 1989. Regulation of tissue-specific
alternative splicing: exon- specific cis-elements govern the splicing
of leukocyte common antigen pre-mRNA. EMBO J. 8:787.

SUMMARY AND CONCLUSIONS

The method involving the combination of reverse transcription and
the polymerase chain reaction (RT-PCR) followed by analysis by restriction
enzymes, Southern blots and sequencing is the best strategy for studying
CD45 alternate exon shuffling. This approach is more sensitive, more
conclusive, and requires far less sample than any other assay. In
addition, it is the only practical assay available for the study of very
small numbers of T cell subsets and clones. The interpretation of the RT-
PCR data was taken with caution since target DNA templates of unequal
length were found to be amplified with different efficiencies under
different conditions. The use of radiolabeled primers in conjunction with
limiting the PCR reaction to the earliest possible cycle number was found
to overcome this problem.

The use of RT-PCR in the analysis of T cell subsets confirmed that
one cell type (or one cell line) expressed more than one CD45 mRNA
isoform. In fact, cells that expressed only one single CD45 isoform have
not been revealed by using RT-PCR in our study of CD45 expression. Murine
T-cells express CD45 mRNA isoforms in a tissue-specific pattern. In
general, the patterns observed by use of RT-PCR were in accord with the
patterns of CD45 expression obtained by SDS-PAGE analysis. T cells tended
to express more of the higher exon isoforms (two exon and three exon
forms) in addition to the common isoforms (mainly zero exon and minus one
exon isoforms) when they become more differentiated and more activated.

Despite the fact that CD4+ cells and CD8"’ cells expressed different levels

116

117
of the one exon isoform, the exon usage from both lineages appeared to be
identical and contained.mainly exon 5 with very little exon 4 and 6. Exons
7 and 8 are previously unrecognized alternate exons.

Since no other CD45 alternate exons have been discovered which could
account for a large increase in the size of the peptide backbone, the
appearance of the extraordinarily high molecular weight glycoprotein
isofonm (up to 260 kDa) of CD45 in IEL implies that there is tissue-
specific glycosylation. The glycosylation of the alternate exons is
likely to provide for 'highly unique characteristics 'when comparing
isoforms containing alternate exons to the isoform without alternate
exons. It is likely that CD45 molecular function involves interaction of
the glycosylated portion on the outside of the cell with other membrane
molecules. This interactiion would then bring the CD45 PTPase domains
into proximity of substrates in the cytoplasm of the cell. Knowledge of
the precise alternate exon use of different subsets of lymphoid and
myeloid cells will no doubt lead to an better understanding of this

function and will suggest future experiments.

MIC IB QRIES

 

mammmm»