2L" :

333‘};

2:2:

 
  
   

     

     
 
  
  
 
 
  

     

   

VVVVV 5’ V‘V'V ‘N fix“ 4‘53“?” 43,15

l
V, V
V V:

 

     
    

V’VVV'VHN'I H! «H..- :- .IV .
‘5’” ’5:"” .2” ‘5””"’:‘ N I! ‘R"V"’ A 55"“? " ,"Nmm "’V" 3va!” 451"?"
:V I7!-
‘ ‘I ».l
I55,” .3. V‘ "V'VVV‘, V! ! V

"1‘,"’”V"g}

”5,!
”I “‘59 V‘V‘ C5555
‘VVu : IVVV5‘5V5VJ‘1’5’A';!

 

 
 
 
     
   
   
     

   
 

«3,5,5, mi“

 

  
 
  
      
  
 

    

    
 
  
  
 
 
 

     
 
 
  

  

  
     
  

I ‘ V‘ 5
J ' 52‘V:‘ >.
V . ', . VV VVV l
V . 55:5. VV,~,..VV,,VV:“.V 5‘,5"“' .VVVVV C
N‘ . ‘ V ‘ I
,PV‘5. A'I‘ 5"." "5” “5:52! l,’\5'.V.,! U" 1.8535” ““1,
"11’ 5“ 4"” ‘1"5' " 5V’VV‘V‘
5V Vi’mVV VVIVVVVVVVMVV' 51(5"‘"51”“”I""’
V , f},
‘V'VVVV "V“ WV:

4‘

.-_.- ?;3’_: firs-(2.3?”
‘4 ‘ .‘Z-"é: graa‘i
.f‘ht- :1" .r.4.
fl - 33.32““

- Eggng
$25311" ,4, :.
-3‘r‘ -
'

$355. .'

“:6“. .
A
._ ;.
A

   
   
   
 

 

V5 .
: l V: V,‘ .5 ‘IV
“V“W‘”Www5355“
75’ ‘V‘VV‘1‘1’VVV’4 ”"’|.’V',V‘ In" ‘34?"
5' 1V VVVV’ (’555‘15‘3‘, V‘IVVVVJV“ i5 .

  

   
 
   
 

A ”SE—9;“
#:II E'-

   
  
    
  

   
 
 
 
  
 

    
    
 
  
  

  

  

   
 

     
 

 
  
  
 
      
 

   
   
  
 

  
   
  
 

   
  

  
       
 
 
  
  

      

         
     
     

   
     

  
  
  
 
   
  

       
 

   
  
  

 

 
 
 
 
 
  
 
  

  
 
  

 
            
  
   

    

  
  
   
   

 
 
   
 
  

 

 
 
 

    
   
 
   
  
   
  

   
 
   
  

     

   
        
 
     

     
   
 
       
   

  
   
  
   
    

 
   
  

  
   

   
    

 

 

 

  
        
  
     
    

  
  

  
  

  
 

  
  
   
  
 
    
 

   
   
   
 
  
 

 
  

       
 

         
   

      
   
    
   

 

  

      
     
    
  
 

  
   
  
   
 
   
     

     

 

 
 
 

  
 
      
      

    
    
   
     
    
      
    

     
  
 
 

      

 
 
    
  
  

   
   
    
    

  

     
    
   
    

    

 

    
   

      

      
  
   

  

  
 

       

 
  
   
 

 
  
   
  

     

    

 
 
   
   
   
  
   
       
     
 
 
  

       
            
 
  
 

  
  

      

    
   
 
   
  
 
    

    

 

      
   
 

 

   
 
 

   
  
 

     

   
 

  
    
    
   
   
    
   
  

     
  
    
   
    
   
     
  
 
  
  
    

   
     
      
    

 
 
  
 

  
 

  
   
    

 

 

 

 

     
  
  

 

 

 

   

 
 
 
 

   
  
  
 

 
 
 
 
     
    

             
 
  
  
   
  

      
   
   
         
  

 

     

  
  
  

  

      
 
 
  

     
   

 

     
     
 
 
 

    

    
   

 
  
         
       

   
 

  
   
   

     
  
  

    
 
    
      
 
     
       
  

    

 

 
 

 
       

  
       

 
  

”fig?“ "”5“" i‘

 

  
  

V
K. ' fVI .
‘5‘" ‘ VVV,,..VV V, gm. {.‘5 , 2‘1}! 5‘35,
. "“".".:u 5‘" V,,V.V‘:V .V 5555“ H .1, ‘55“ ,:V ,V,!', VyVV‘VVrVVf-‘V VVV‘VVI
‘V ‘V'V 21‘" :; :VVVVVVViVVV‘V‘V '5‘5VVV‘ ‘i‘V-VV~‘.V‘“ x’VV’VV‘VV ‘3‘“ V'VV.” :V'V‘V 4‘ ”‘5,” VVVVVVV‘V‘ I'VEVVVV-VVVVVV
I‘V‘I ‘V'V ,,,‘ V,‘ ,45VVV’ ’VI‘ V ,1, V ’5‘” :5:(:(V‘,V:5l,lg‘l“55‘4,g’1.l(‘I‘f."rh"y{r‘ ’ l V 5531”: .‘ H5 . .
:‘V. ””W%“‘ . VVV“MVV%MVVmV:fiVVVVWE%V- :

.\ ‘ V"""‘V‘V"’ I» w“: I A ".1", ‘. Pf“? ”I

V. V‘ :V"‘5‘V,5V,:: ” .‘V55VVV "‘1 ' ‘VJ VVV: "'1‘ :5), .'VV5 VS” 5'5‘ "5“ 554151, “#55 3551584“ ’15:“. I’V,,V,~V,V, ,,V 3,15“ 5214...: V. V'“ ..

‘5: 15’5"":‘V5"5 V . .VVVV’V5" l5‘5’5”‘ V" :55: '52'V‘m‘55 a“? ‘V"”V.~‘V"V I, 5f‘"”"”"”” W“! 51555555; , ‘5“"‘?‘“"‘5“V"’t’:‘3555‘“): ‘5“‘V‘5“$ ‘55’3 V5 ‘QLV‘g‘ ‘IEL‘QK’W ,. .
1 VV‘VV "WVVWVVVMVVVMVVVVVVWMMMHVMV .m. .
w: VVVVVVV‘”VVV”%VMNV*“VVVHVW Vk w -

"‘"V-V'-V “"V"V’"‘V . ’ 'VVV‘: V“ .V ‘ "‘V ,, ' :V" ,1: :‘V’ . ~ " WM : .
V”‘""5"’V‘ V" V’V.!’V ’4 ‘5"‘V,:V!’ ‘5‘“) ‘ 'V‘5’Vl ‘5‘5‘5‘5"V" ‘V,”‘ ' ' ‘ ’ " M . .
: V ,V' :‘V :VVV’ V4 .~.”‘VVVV:.,: VV V::.:V:.:‘: ‘VgV VVVVVV V‘“ ’V‘ V . > . V‘VV
)‘g’3““51’55’1””“5"5‘55’1’5‘55m5355’? .‘5.5"”’”" V ‘V5V. VVV‘.:, .01 ‘55 “VVVVV V ’V' . .
VVVVVVV‘: : V: VVVVV: VVV‘VV ::,.,,,,V,V, :13“:'""" V: .V .V :1... .
V’J IVVV'4". a‘V‘gV’ ,‘V V‘5V,‘. ’ 'V I ,V ‘V‘ ’., ‘i‘VW‘V .V' 5"“"353': .1‘2‘" 55"" :55, J: "‘ ‘ '-. . V .- '4:sz ' ‘V

"5“‘5V5‘V‘V’V’ 5‘5" (’1‘5‘5,1551'.N“55"VV“IVVVV.V‘V ‘V ’55 .VVVVVV, VVVVV 55"“ .VLVVV5VVV,V.VVL, ”’5, ”‘4‘”! iVVV, ‘HV‘J ‘V‘VV‘V‘ 5”," “1%“, (R .Q #31" 5d,. i§fifimx

.VV , ,V 555,5 j‘l’V'VL‘V’VV V ’V’VV5‘ V "5””VV5‘ 5'55” ‘VV’V’V“““"“VV5’5‘ 5’5"“‘fi’ 552$“? 555555555 15551.! ’555”,5$h"“351“§”L‘éi‘svfiglh ‘2? ()3);

V‘ "‘V,.VV!VV‘ ’.‘:,’ V, ‘ ' \ r-' 0'" 153
~WV ‘VVV HM”W:V WV“ V, 3% ‘ m:fi&_fifix§fi
.mV VNMWWMWW‘W‘ V'VV‘ VMVV VV”N%M*'“
V ‘5'.V:V ! . " .4: : ”in“ ‘1‘ V3! “1"‘V’ ’ "A‘ {“V’R' '

”NV“ ’55'515, V), H,’ 3,5 ’V’V"" 11V” ‘H’Vl :4)4~,:( I) ,V \ "V .fafiajta'1fi13m‘d‘gfi’
’V‘VV'TV ‘ “WWW-V 35“}: 5V «V :‘V. V‘V.‘ 55“”?! ‘V‘VI! “""’”V“ ‘VLV ‘VV‘ "I“: 1"“: N’ V’ ’7 ”.1. 35““: -
“:VVV: :V:,:::V .VV'V‘:V‘VV‘VVVV ‘VVIV‘VV" VVVVVVVVV " .V‘VVVV V“ ,.,*V':V » V: .
:‘I‘ "‘,;,“,‘ "”""5’I HI) L‘.‘ ‘ ;
. 5135me IV"? 3‘" 5"" H, .VV ‘VVTWV ,5.‘ "EV‘ V::.VV V M ‘51,“ 5‘5““ {,2‘ V 1,4 {‘55 53“," "V‘W’L
I"”' 5 ‘Via: "V' VcVV,V.VVVVV V5 “55“ 5555‘, 3’55" VVVV,V‘~VV‘ 5 5‘5" ,5, . V “59“5' V.£VV,..
. .VV V‘V“. V V VVVVVV‘!‘ :VVVVVVVVV V V‘VVVV V“ "" "V“VVV‘ ,V‘VVVVV :VI’ '2‘?» V.“ ""
,1V5f:_l Jig,“ "’V"‘ V, [5,5 ’V ,‘VVV, 5 '5” V} ,.5 V'VV ’KV’VV’ .V.
. ,VV,,,VVVV‘V,, 1,: 5"V‘V!,‘“A,‘ V’V‘ VVVV 1,54 N255 {I} V‘V‘I’ V
. 5‘5‘5\V'.’5‘:,5 V VVfllH'VV'IVVV ‘5“ 555" “V” “‘.55”’V’V" 5%:“15”;

‘_"<f> V 555V5, .5,‘, VI V‘VVV” .V 3"”55’5‘5 5"“ “5‘5” VVV‘VLVVVVVh;

VV:5VV 1.1: “51:3 ,VVVVVVV :VVV‘VV’ V: V’V‘ :VV! VVV'VVV'V VV‘VVZVVV.V‘ _

u: ‘5"IV,,‘V .‘5‘5: ‘4 ’ If’._'x-_I~’ .‘w

*V" L 5”” ,5555V’5 5,55} VV‘V‘V‘, V‘(’ V5 5 “55”" V1 "1‘ ‘5‘5’5,’ 5“", ‘g’h‘a‘kafl‘ . ' .‘ , 1 V",

w VVVVV .., :”a- :“Vfiwww -%w

»,;V " ”"515!”"35‘34’3 I”IV”V {5" ”’5’5‘:‘. 5 "“_‘37""’ :‘I v,— ‘.- 9"“ ’H “.7 ”‘5" ; H .' "1&1,

'. ” ‘5V5‘5’5” ’55V ‘5‘M1,""I ‘5‘5 {“5' 5’5“ "”“’I1' 5V .3’ .455”- . ' :5 , 5 .’ '. «final: ‘5}"5' -'V 11,“. :

I,“ V 55.535“! ‘VVVV: VVV “H.555 VVVV‘VVVV'..:,VV' “5"“ 'V:.V.VVV~ V2,, .:. -V V. . quVV, ‘1‘}; ‘5, "a ,5

55‘ 5V“ ’V 5 VS‘f‘" 5’5“ ‘.1”V"" 5““ 5M ”455:" ‘: V5?! ‘ V'V ‘ VVVgé' . 1,V 1.. $159 *4," N '5, 1,. ,ng?’ 'l,

‘ 5%??? VPI" "" ‘5,“(‘55 ’ ‘5" 52 Illl‘a ("(5, ‘55 l, " ‘ f]. V‘, . . V . V 5‘.’ng‘$ "‘d“ .H‘I', l:‘I}0:V c"’l-IV ""55’8’!
VVV ,wuw VWflfl w :ww' u. , :VWVVVfixMVWMwMV,
.. :VVVV.VVVV V“: V... V V: , . ~V: . :VV‘ .., :. :V-«IV: :

'55?" V, ’55“ 5V: :5 V'.‘VVVV 5,. “'V . ‘VV‘V‘V- ’ V‘VV‘ :‘VVV ' 7‘51 5 5“ “5"" ““1” “’2’" 2,... ‘2‘1’3 "

. ' .,1.--‘-V.V‘ . ., “V :1: ' : . ,V. . I . a ’2.
"H, ",5" ‘V‘,"V, I515.“l V} ”5155‘ 3’55”" 5 1-,IV . V , V1: V1 , P ‘0” ’-’-, ‘ " .3555“V'5‘.'..

VWINWMVV.V VV‘” V .WVV‘ . mm“:

" ’ ’ "”’ ‘ V ‘ " ‘:‘ ‘ "’1 ’ I“ 33:.

- V‘Vf VVVVV VV‘VVVV‘V .‘VVVVIV VIV‘VZVVV' . .V :V '5' . V V‘ .VVV‘VVVVVVV V "V V Q“.
‘lV"’I" "”55“ (55" ‘5“ "5“" "”1” 5‘15 "“1”"!ill 5‘ "I", :5" I ‘5" ”"!“5“"" "‘VV ’ VV’I" A
":V' ‘5‘!" "5‘! "’2‘“ "'2" 52‘V‘VV“ VV' '- “V -.‘ ' ”V" 5"” ‘fl‘L‘V .. $44??-
9’55‘5’55V’V‘V’ (55‘3er "‘53 5.‘ :‘5‘VV-VJ V \‘VVVVVVVVV ,(VV..V.V -V. (53V, .5 5451. "g" V ’. V ‘3 V s ‘. ,5 ._ 5H1
V ’r‘"5“"‘ MV’5’V‘5“~ ”N.“ “’V’V‘.""5V‘ ‘ 5 "51"5 ‘5“) V 3 555;! 5,. ‘5"‘ 75‘4"! ‘_ ‘, ,1”, Winn". AI
'-'~.'.V V} 5H5. V ""J-V V," 1’" P,‘ 155““ 5 .V5’ ,5, .‘5‘ “IV-VI 5‘ 5- I .' V . :sl‘fiufi‘u .

H‘V‘V' 555V’55V'4"’ ‘3 ‘V"’VI‘V‘“V ” .555‘5 5‘5, ’VV ’53:“le . ‘E‘VV V5,,VV595V5 "5.“1‘5,“ I. ' , V ‘V‘JVT 14’1"!
“31" MMVHVMVV.WF ‘"%MVVV “fi' ‘MW
V. “T” .‘V!“V1“V‘=V§“:‘VV.““VVV5“"’ ‘5‘ ‘VVVVVV? .VVVVVVVQVV” . :2, - ,
5’ ’ ‘:“‘:‘VV"VVV ‘..' "5V5’V5VV’VV(‘V:5V5VV‘V’:V“ 53,5! 3’" " ‘ VVVVVV‘VV'VVV’V" . ‘
‘V.. . : >V V:':V' .-V'V;‘! :‘V' ' ‘. w.
VV‘ ‘1: “‘V“‘“VVV,,,V’1VVVVV.‘:‘VVV!VV,’,V‘:V ,, :VVVVVV VVfiVVVVV
‘v‘,! '2‘ ,VV‘V.:.:,VHV 'V . M‘V‘ “, 2 ._“: 5,3,. --
V "’ “V55 “5““; ""5‘fl’l . ‘1, $5,315
',-.V'V 55:, ,VVVV“ ‘VV’VVVV "Vi“ 5““V.V 1,, Qua? :, .
VVVV.VV 1us V'NV“ V:,:: w: V. :5? _
-hth ,, w V V»:
V 7.: W V W
V :VV‘ .:, ’4‘ ‘VV': ' .. VVVVVV
' ‘ V V3 w ‘ ‘ ‘ -' .VV'VVVV; . ,
:VVVV‘ ‘ : ,,-V:'.: "3“! .2 -,
( ‘53.: :355‘" ‘ . ‘ ‘ , I
V“ - .V . 5‘. ‘V ’V : ‘ 1.4:: ' "5'2'WV’3‘V
. ' ‘ IV!“ 5"“‘5' ‘: 5"“~V"555‘5§"55‘V V. -. ~"- ‘« 'I5‘, : 3’ .VJVGE EH":
, ‘l‘1‘5'5VVIVC :VV:,,5:TV5VVVV',,.!.VVV5: 4V, -_ V VV55‘, 7'24. . . I..\ .
V. V5, V‘L“ “Ht“! V}! A‘VVSVV 'V,‘ 1} .4305?
V: V: .
:V 5

    
 

VVI H}; «’5‘ 5; -'V .
J5"

es“; .5 . ugl’é V:

1Michégan State
University

THESIS

 

This is to certify that the
d?ssertation entitled
IDENTIFICATION AND CHARACTERIZATION

OF THE NUCLEAR RNA TRAFFICKING PATTERN
IN NORMAL AND ADENOVIRUS-INFECTED CELLS

presented by
Roger Martin Denome

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Genetics

@km

MU grotessor

Date ¢//%/£

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

MSU

LIBRARIES
m

\—

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

IDENTIFICATION AND CHARACTERIZATION OF THE NUCLEAR RNA

TRAFFICKING PATTERN IN NORMAL AND ADENOVIRUS-INFECTED CELLS

By

Roger Martin Denome

A DISSERTATION
Submitted to
Michigan State University
in partial Fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Genetics Program

1985

ABSTRACT

IDENTIFICATION AND CHARACTERIZATION OF THE NUCLEAR RNA
TRAFFICKING PATTERN IN NORMAL AND ADENOVIRUS-INFECTED CELLS

By

Roger Martin Denome

The study of nuclear partitioning of RNA has received very little
attention as a possible area of regulation of gene activity. The
physical location of RNA transcription and processing has been
ascribed to the nuclear matrix. which is a salt-insoluble
proteinaceous network that fills the nuclear space and is contiguous
with the lamina and pore complexes. Described here are experiments
that determine the fate of nuclear RNA after it has completed these
matrix—associated processing steps. Pulse-chase and continuous label
experiments indicate that after this RNA is processed it changes its
state of attachment in the nucleus such that it is now removed from
the nucleus in the high salt extraction step of matrix isolation. It
is this salt-extractable RNA that will be transported to the
cytoplasm. Late in adenovirus infection. when transport (but not
transcription or processing) of cellular sequences is decreased.
these sequences do not make the transition from the matrix-associated
to the salt-extractable nuclear pool. The implication of these data
on the regulation of gene expression in both virus infected and

normal cells is discussed.

ACKNOWLEDGEMENTS

I would like to thank Dr. Ronald J. Patterson for taking a chance on
a student of dubious descent. Without his help and encouragement
this work would not have been started. I would also like to thank
Dr. Elisabeth A. Werner and Nancy Dwan. both of whom supplied

technical and philosophical assistance that made this research both

possible and enjoyable.

ll

TABLE OF CONTENTS

Page
LIST OF TABLES ................................. . ........... iv
LIST OF FIGURES ............................................ v
INTRODUCTION ............................................... I
LITERATURE REVIEW........ .................................. 3
RNA processing ....... . .............................. 4
Nuclear architecture ................. ... ............ 7
RNA transport.... ...... . ............................ 12
Conciusion..... ............... .... ...... ............ 20
Bibliography ........................................ 22
INTRANUCLEAR RNA TRAFFICKING ............................... 31
Introduction.. ...................................... 31
Materials and Methods ............................... 34
Data 0000000000000 000.000 0000000000000000000 O ........ 38
Discussion ................................ . ......... 60
Bibliography ........................................ 68
APPENDIXOOOOOOOOO 0000000000 .00 000000000 OOOOOOOOOOOOOOOOOOOO 73
Introduction ..... . ....... . ............ . ....... . ..... 73
Thy-I cDNA sequence suggests
a novel regulatory mechanism... ................ 74
A hydrophobic transmembrane segment
at the carboxyl terminus of Thy-l .............. 77
Structural organization
of the rat Thy-I gene .......................... 80

iii

Table

LIST OF TABLES

RNA recovery after cellular fractionation....

Relative distribution of introns and

exons in nuclear RNA pools.............. .....

Levels of cellular and viral RNAs

late in Ad-2 infection........................

Levels of cellular RNAs at intermediate

times in Ad-Z infection.. ...... ...... ........

Page

....... 39

Figure

LIST OF FIGURES

Page

Dot blot analysis of salt-extractable and
matrix-associated RNA. RNA isolated from Hela
cells mixed with cytoplasm from an equivalent
number of myeloma cells (lanes 1 and 3). or
from myeloma cells alone (lanes 2 and 4) was
diluted serially (two-fold) (rows A-G) and
applied to nitrocellulose. Row H contains
loug yeast tRNA. The probe was
nick-translated pLZl-l DNA. Lanes 1 and 2 are
salt-extractable RNA; lanes 3 and 4 are
matrix-associated RNA..... ........ ...... ............ . 42

Size analysis of HeLa and myeloma RNA
fractions. HeLa and myeloma cells were
fractionated into cytoplasmic,
salt-extractable. and matrix-associated RNA,
and the RNA was separated on 1% agarose-
formaldehyde gels. Gels were stained (panel
A). or RNA was transferred to nitrocellulose
and probed with nick-translated perNA (panel
B) or pL21-l (panel C) DNA. Lanes: 3. HeLa
cytoplasmic RNA; b. HeLa salt-extractable RNA;
c. HeLa matrix-associated RNA: d. myeloma
cytoplasmic RNA; e. myeloma salt-extractable
RNA; f. myeloma matrix-associated RNA.. .............. 45

Continuous label kinetics of RNA populations.
Eyeloma cells were labelled continuously with
H-uridine, aliquots were fractionated into
cytoplasmic. salt-extractable and
matrix-associated RNA at various times. and
specific activities determined. O——C.
cytoplasmic RNA; C>——-<). salt-extractable
RNA: I—I . matrix-associated RNA.. ................ 50

 

Pulse-chase kinetics of labeling of RNA
pulations. Myeloma cells were pulsed with
H-uridine for 8 minutes and chased as
described in the text. Cells were fractionated
at various times after the initiation of the
chase. and the RNA was used for TCA
precipitation (for total counts. panel A) and
hybridization to pLZl-l DNA immobilized on
nitrocellulose filters (panel B). 0—0,
cytoplasmic RNA;C>--<D. salt-extractable
RNA; H. matrix-associated RNA..... .............. 52

vi

INTRODUCTION

RNA metabolism has received a great deal of attention. Studies
investigating transcription, various processing steps. and
translation have revealed the complexity in these processes and have
advanced our understanding of the basis of gene regulation to a point
where a great many regulatory phenomena can be understood on the
molecular level. One area of RNA metabolism that has not received
much attention is the actual movement of RNA within the cell and the
effect of this movement on gene expression. Described here is a set
of experiments designed to determine if there is indeed some
organized partitioning of RNA within the nucleus of a eukaryotic
cell. and if so. whether this partitioning is used as an additional

level for regulation of RNA metabolism.

The idea of gene regulation at the level of RNA partitioning is not
new. Messenger RNAs are sequestered in the cytoplasm of amphibian
oocytes. These RNAs are not available to the translational
machinery. thus allowing for the storage of the large amount of mRNA

required for the early stages of development of the organism.

The work described here addresses two questions that have not been

addressed directly before:

I. Is there some physical partitioning of nuclear RNA that
separates this RNA into functionally different groups?

2. If there is some physical partitioning of RNA within the
nucleus. is this partitioning used as a means of regulating

gene expression?

The importance of this work. regardless of its conclusions. in the
advancement of our knowledge of gene regulation cannot be estimated.
This study attempts to determine whether the potential for gene
regulation at this level exists. This work examines the possible
role of regulation at this level in the massive effects on cellular
RNA metabolism caused by adenovirus infection. Further. more subtle
uses of these mechanisms for regulation of cellular phenotype and
differentiation await both more sensitive techniques for the analysis
of nuclear RNA populations and a great deal more basic research into
the structure of the nucleus and its relationship to RNA metabolism.
From this point of view. this research is just a first step into a
field that holds great potential for the understanding of gene

regulation and cellular differentiation.

LITERATURE REVIEW

Intranuclear RNA movement is a subject about which almost nothing is
known. A great deal of work has been done on characterization of RNA
transcription. processing. and translation; the association of RNA
with other macromolecules during these processes has also been
described in some detail. RNA transport from the nucleus has
received less attention. although some advances have been made in the
area of defining in vivo and in vitro model systems that can be

used to study this phenomenon. Information from studies in all of

these areas bears on the subject of intranuclear RNA trafficking.

This review is an attempt to pull together all of the pertinent data
bearing on the subject of intranuclear RNA partitioning and movement
from all of these fields. RNA processing. nuclear architecture. and
nucleocytoplasmic RNA transport will be reviewed here. This is not
an attempt to cover these subjects in depth; rather the information
in these general areas that bears upon intranuclear RNA movement will
be described. Where possible. major reviews of the general subject
discussed will be cited, since much of this work has been reviewed in

much more depth than will be used here.

 

RNAgprocessing

Eukaryotic organisms usually synthesize a primary RNA transcript that
requires several processing steps to produce a functional RNA. These
modifications vary from transcript to transcript. and in fact may be
different for the same primary transcript made at different times or
in different cell types. For example. primary transcripts of the
immunoglobulin u heavy chain genes are processed differently
depending on whether the protein produced is to be membrane bound or
secreted (4, 49. 67). The murine a~amylase mRNA exhibits
tissue-specific polyadenylation, which may result in either a
differential stability or differential affinity for translational

machinery (95).

Primary transcripts produced by a given RNA polymerase seem to share
common processing pathways. Ribosomal RNA undergoes a series of
modifications that produces three mature rRNAs (5.88. 188. and 285)
from a 458 primary transcript (31). This processing involves the
removal of external and internal transcribed spacers from this

primary transcript, plus internal methylation.

Messenger RNAs are produced by a more variable processing pathway.
Primary transcripts are "capped" by the addition of a methyl
guanosine in the structure m7G(S')ppp(5')X where "X" is the 5' most

nucleotide of the transcript (reviewed by Shatkin. 75). This

 

structure. or a modification of it. has been found on every mRNA
examined from eukaryotic cells. Capping occurs extremely quickly
after the initiation of RNA transcription. and probably occurs on the

nascent chain (7, 69).

Precursors to mRNAs also undergo endonucleolytic cleavage at their 3'
end. followed by the addition of polyadenylic acid at a point near
this incision (reviewed by Brawerman. 21). This processing step does
not seem to be as universal as the production of the S'cap structure
and the frequency of polyadenylation seems to be both species and RNA
sequence dependent. For example, histone mRNAs are rarely
polyadenylated; a large minority of mRNA in Drosophila (about 252) is
not polyadenylated. whereas all yeast messages examined are
polyadenylated. The length of the polyadenylic acid (poly(A)) tail

also varies from RNA to RNA.

Precursors to mRNAs are also spliced to remove nucleotide sequences
that are not present in the mature mRNA (reviewed by Breathnach and
Chambon. 22). The function(s) of these "introns" that are removed is

not known. and not all primary transcripts contain them.

Finally. mRNAs are methylated at internal nucleotides. usually at
adenosines in the sequences GpmGApC or Apm6ApC. This methylation
usually occurs in the nucleus and is conserved during the processing

of precursor RNA (43). Inhibition of methylation in SV40-infected

 

 

cells with cycloleucine inhibits the production of mRNA, but not the
synthesis of hnRNA (32). Similar results suggesting that methylation
is important for processing and nucleocytoplasmic RNA transport of

cellular mRNA have been reported (25).

RNA polymerase 111 products (small nuclear RNAs (snRNA). tRNAs) are
also processed from primary transcripts that are longer than the
mature RNA. The processing of the snRNAs will be discussed in the
section on RNA transport since their processing occurs after the RNA
is transported to the cytoplasm. Transfer RNAs are produced by
endonucleolytic scission of long primary transcripts to produce a
mature-sized tRNA. Further base modification of this product is

required before the RNA is transported and functional (45. 97).

All of the modifications described above, with the exception of those
occuring on snRNAs. occur in the nucleus and are presumably required
for normal RNA transport and function. There are some RNA species
that are restricted to the nucleus (see 73 and 74 for examples); the
function of these is unknown. although some snRNAs have been
implicated as structural elements in the nuclear matrix (86. see
below). There does not seem to be any consistent differences between
transported RNAs and nuclear restricted ones. nor does there seem to
be a common biochemical characteristic of all RNAs that are

transported.

 

Nuclear architecture

The study of intranuclear RNA movement must take into account
information supplied by ultrastructural studies of the nucleus.
Without some physical framework to accompany any biochemical
descriptions of RNA processing and transport. the description of RNA
traffic within the nucleus will never be complete. Fortunately. the
picture that has emerged from a broad range of studies on nuclei from
many species is one of suprising consistency in nuclear structure.
Because of this consistency. generalizations using data obtained from

species in a broad phylogenetic spectrum can be considered.

If the nucleus of a cell is stripped of its membranes. digested with
DNase. and extracted with high salt concentrations. the resulting
insoluble structure looks very much the same as the original nucleus
(13. 76). It retains the same shape. size. and organization of
nucleoli as are observed in the original nucleus. This DNA-depleted
salt-stable nuclear structure. the nuclear matrix. has been found in
rat liver (48). Tetrahymena macronucleus (91). Friend erythroleukemia
cells (51). HeLa cells (17. 84), BSC—l cells (12). chicken liver (10.
20) and ventral prostate (10). murine 3T3 cells (40). chicken oviduct
cells (28). and chicken erythroblasts and erythrocytes (68). Its
structural constituents after DNase digestion are mainly protein and
RNA. Extraction of membrane-free DNA—free nuclei with high salt

concentrations removes a large portion of the RNA but leaves the

 

structure morphologically unchanged. Digestion of the matrix with
protease destroys it morphologically. Digestion of the matrix with
RNase has left the matrix structurally intact in most cases. There
is one report of RNase digestion destroying the integrity of the
matrix (26); whether this was an artifact of the system being used or
whether there are in fact some structural RNAs in the matrix is
unknown. There is also one report suggesting that a small nuclear
ribonucleoprotein particle (snRNP) containing snRNA is a structural
constituent of the matrix. This class of particles was detected
using immunofluorescense techniques and antibodies specific for a
subgroup of the snRNPs (88). Again. it is impossible to confirm or

deny this from the available literature.

The matrix structural proteins fall into two classes. The lamina
proteins comprise about 40-60% of the matrix proteins (13. 38). The
remaining proteins are of heterogeneous size. Actin and vimentin are
present. probably as components of the matrix that correspond to thin
and intermediate cytoskeletal elements (26). Electronmicroscopic
analysis of the matrix shows a network of heterogeneous-sized fibers
filling the nuclear space and contiguous with the peripheral lamina.
The nucleolus is seen as a darkly-staining body imbedded within this

network.

A number of properties have been attributed to the matrix. All are

consistent with the idea that the matrix is used as a "workbench" for

 

nuclear functions. DNA binding has been ascribed to the matrix. as a
way of holding chromatin in the 40-60 kilobase pair loops that are
found in eukaryotic nuclei (57). as a site of DNA replication (63).
as a method of aligning interphase chromatin (2). and as the site of
transcription (see below). The site of attachment of the DNA to the
matrix of murine 3T3 cells (40) and Drosophila Kc cells (57) is DNA
sequence dependent. and the proteins involved in the attachment of
HeLa DNA to the matrix seem to share extensive homology with other
such proteins found in a wide variety of organisms (17. 60).

Electron microscopic autoradiography of nuclei labelled for 1 minute
with 3H—thymidine in vivo shows that 85% of the newly-labelled

DNA is associated with the interior of the nucleus. Similarly
labelled nuclei that were fractionated into either matrix or lamina
suggested a similar distribution of newly-synthesized DNA. Assays of
the nuclear matrix and lamina for DNA polymerase activity found DNA
polymerase a activity specifically associated with the interior of
the matrix. consistent with its role in normal cellular DNA
replication. DNA polymerase B activity copurified with the lamina
portion of the nucleus. DNA polymerasee B activity. which is
involved in repair synthesis. has been localized to the perilaminar
nuclear space by autoradiography (77). The genomes of SV40 (1).
adenovirus (94). and influenza (42. 44) viruses are matrix-associated
in infected cells. as are newly—synthesized RNAs from adenovirus (54.
55). SV40 (12). influenza (44) and host cell (86) genomes. Estradiol

and dihydrotestosterone bind specifically to the nuclear matrix of

10

liver and prostate of chickens competent to bind these hormones.
These hormones did not bind to tissues that were not normally target
tissues for these hormones. Isolated nuclear matrix from these
target tissues was also able to specifically bind these hormones
(10). This indicates that the matrix may be the point of action for!
their modulation of transcription. Indirect immunofluorescense using
antibodies specific for adenovirus E1Aa protein (30) and SV40 T
antigen (98). which are know to modulate transcription of cellular
and viral sequences (18. 30. 37. 46. 62. 87. 98). has localized these
proteins to the matrix in infected cells. There is some disagreement
in the literature as to whether or not actively transcribed DNA is
preferentially associated with and protected from nuclease digestion
by the matrix. Robinson et a1. (66) found that the chicken ovalbumin
gene was closely associated with the matrix after DNase I digestion
in oviduct cells but not in chicken liver. suggesting that the matrix
association does indeed offer some nuclease protection to actively
transcribed genes. However. Ross et a1. (68) found no such

protection of the B-globin gene in a similar study in erythroblasts.

High molecular weight hnRNA is associated with the matrix (55. 86).
and when individual transcripts are examined. both mature- and
precursor-sized molecules are found in association with the matrix.
This has been shown in chicken oviduct cells using probes for

ovomucoid and ovalbumin mRNA (28). chicken erythrocytes and

 

ll

erythroblasts using probes for globin mRNA (68). and Tetrahymena

macronuclei using probes for rRNA and its precursors (41).

All of these data argue that the matrix plays a central role in
nuclear macromolecular metabolism. What other structure(s) may be
involved in RNA transport is unknown. The fact that most of the
studies on matrices involve extraction of DNA-depleted nuclei with
high salt concentrations makes analysis of other less stable nuclear
structures difficult. There is some controversy over the exact in
3129 disposition of the matrix. (See references 20 and 96 for
example.) Some of this uncertainty may be the result of the
sensitivity of the matrix to variations in isolation procedure.
Changing the order of DNase and salt washes or leaving out protease
inhibitors during isolations may have lead to poor matrix isolation
that in turn leads to a variety of misinterpretations (48). One
example of this is probably the work by Berezney and Coffey (13) on
protein components of the matrix. They found that the matrix was
composed of three proteins of 60-70 kilodaltons. These are in fact
the sizes of the lamina proteins (38). indicating that the

preparation of matrices was not performed correctly.

Only two publications examine the RNA extracted by the high salt
washes. Ciejek et al. (28). working with chicken oviduct nuclei.
separated this RNA electrophoretically on denaturing agarose gels.

and then transfered it to nitrocellulose. This was then probed with

 

 

 

 

 

 

 

12

nick—translated DNA probes for ovalbumin and ovomucoid mRNA. Only
mature-sized RNA was found. A similar study by Ross et al. (68)
found equivalent results using chicken erythroblasts and erythrocytes
and probes for globin RNA. These data. along with the fact that
unprocessed precursors to mRNA and rRNA are always associated with

the matrix. argue for some key role for the matrix in RNA processing.

RNA transport

RNA transport from the nucleus has been studied using a variety of
approaches. all of whichhave their disadvantages. Several groups
have worked with isolated nuclei in attempts to develop in vitro
systems. These have met with only dubious success. mainly because of
the lack of proper controls on the types of RNA transported. One of
the main conclusions to be drawn from the data resulting from several
of these papers is that RNA is in fact transported from the nucleus.
and that the double membrane surrounding the nucleus has little to do
with this process. Immunoglobulin K light chain mRNA (82) and rRNA
(39) are transported with the same efficiency from nuclei. regardless
of whether they are membrane bound. In Tetrahymena. the amount of
rRNA transport is directly proportional to the temperature of the
medium in which the cells are grown (39. 89). This regulation has
been shown to be a property of some intranuclear structure. and is
believed to involve some "temperature sensitive domain of the nuclear

matrix” (90).

 

 

13

Others have worked with developmental systems where expression of a
given RNA is controlled over time by some internal or environmental
stimulus. The major system for this work has been the modulation of
transport of cellular RNAs in adenovirus infection. with some
additional information coming from work on the "heat shock response”
in Drosophila. CHO and HeLa cells. A third area of study is the
characterization of RNA-associated macromolecules. Work on both the
nuclear and cytoplasmic RNA-associated proteins and RNAs has led to
the description of a large group of associations which may or may not

have something to do with RNA transport.

Infection of HeLa cells with adenovirus brings about a set of massive
changes in cellular metabolism. These changes all result in the
eventual takeover of the cellular machinery to produce virus.
Adenoviral infection is divided into early and late periods. with the
delineation between the two being the onset of viral DNA

replication. Early in infection. cellular metabolism is effected
only slightly (27). Cellular RNAs are synthesized and transported to
the cytoplasm at normal rates. and translation of cellular mRNA
occurs normally. During this time. the expression of heat shock (46.
62) and B—tubulin (79)genes is increased by the protein coded for by
early region 1A (ElA) of the virus. Whether this altered gene
expression is responsible for events seen later in the infection is
unknown. although the transcription of heat shock genes is back to

preinfection levels by the time viral DNA synthesis begins (62).

 

 

 

 

 

14

Late in infection. transcription from the late viral promoter results
in a large increase in the amount of viral RNA in the nucleus and
cytoplasm (3. II. 27). The processing of this adenovirus-encoded RNA
seems to follow the same pathway as cellular RNAs (3. 7. 14. 53. SS.
61. 98). The transport of cellular polyadenylated RNA (6. 11) and
rRNA (27) to the cytoplasm is reduced to almost undetectable levels.
even though the transcription (6) and processing (34) of these
sequences seems to occur at normal rates. Very late in infection
transcription and processing of ribosomal RNA is inhibited (27).
although this may just be due to the lack of ribosomal proteins that

are required for transport of rRNA from the nucleus.

Several investigators have looked for the block to cellular RNA
transport. either by looking at total cellular hnRNA and rRNA. or at
specific cellular and viral RNA sequences late in infection. Nuclear
viral RNAs seem to be associated with the same proteins as cellular
RNA molecules (16. 59. 83). Both total cellular poly-A-containing
RNA (11) and rRNA (27) transport is reduced to almost zero. although
transport of rRNA continues later in infection than that of mRNA
(27). Examination of individual RNA sequences late in infection has
revealed a slightly more complex picture. The transport of some
cellular sequences. such as the mRNA coding for dihydrofolate
reductase (DHFR). is inhibited only 4 fold (93). Furthermore. about
half of the reduction seen in DHFR mRNA transport can be accounted

for by a reduction in the nuclear stability. but not the

 

 

 

 

IS

transcription rate. of the transcripts late in infection (92).
(Adenovirus causes an analogous loss of stability of the cytoplasmic
heat shock RNAs late in infection (62).) Other RNAs are not
transported at all in late infection. and the transcription of
histone mRNA is inhibited late in infection (35). These
investigations have not yielded either a consistent cause for the
loss of cellular RNA transport nor have they localized the block in
RNA transport to some subnuclear processing step. The inhibition of
cellular RNA transport is not due simply to a "swamping" of the
transport system by large amounts of adenoviral RNA. since about 50%
of the RNA in the nucleus late in infection is cellular in origin
(11). Very little is known about the viral functions that cause the
loss in cellular RNA transport. Babiss and Ginsberg (8) have
described an adenovirus mutant that is missing the protein coded by
the early region 1b gene and does not inhibit host cell RNA transport
late in infection. Similar results with a deletion mutation
affecting the smallest early region 4 protein product were found by
Sarnow et al. (70). Interestingly. this protein is normally

associated with the matrix late in infection.

The linkage of the template for a given RNA seems to have a some
effect on whether or not that RNA will be transported late in

infection. When the entire human a-globin gene was inserted into
the adenovirus genome. it was transcribed and transported to the

cytoplasm late in infection (71). These data indicate that physical

 

 

 

 

 

 

I6

linkage of the template for transcription to the viral genome is
enough to allow RNA transport. Consistent with this hypothesis is a
report that influenza virus transcripts are transported in cells
doubly infected with adenovirus and influenza (47). These data are
at odds with the information on DHFR mRNA presented above. since
linkage of that gene to the cellular genome does not seem to result
in inhibition of RNA transport. In an experiment analogous to the
work on a-globin described above. 293 cells. which have a copy of
the adenovirus El gene integrated into the cellular genome. were
infected with an adenovirus mutant deleted for the E1 cistron (78).
Late in infection. the cellular copy of the El gene was transcribed
and transported to the cytoplasm. indicating that something beyond

simple linkage is controlling RNA transport.

Work with heat-shocked Drosophila. CHO. and HeLa cells has led to
contradictory information on the effects of physiological stress on
RNA transport. In Drosophila. heat shock results in the inhibition
of normal cellular RNA synthesis (33). the inhibition of rRNA
processing. and the enhanced transcription and translation of a set
of RNAs coding for "heat shock proteins" (56). The amount of protein
associated with the nuclear RNA after heat shock is reduced by a
factor of 20 (56). However. in these experiments there was no
preferential association of the heat shock RNA with the protein that
is still associated with RNA. Heat shock of CHO cells results in

essentially the same effect as heat shock in Drosophila (19).

 

 

 

 

17

Furthermore. processing of rRNA is aberrant and the resulting
"misprocessed" rRNA is never transported to the cytoplasm. even after
normal RNA processing and transport is resumed at normal
temperatures. In HeLa cells. the amount of protein associated with
nuclear RNA during heat shock is reduced compared to normal levels
(56). similar to what is seen in Drosophila. Hela cells continue

to transcribe RNA coding for non-heat shock proteins during heat
shock. but most of the RNA that is transported to the cytoplasm is
that coding for heat shock proteins. The association of heat shock
with the loss of RNA transport from the nucleus is a tenuous one. As
mentioned above. adenovirus infection induces the production of heat
shock protein early in infection. During this period cellular RNAs

are transcribed and transported at normal rates.

It has been suggested that the association of various macromolecules
with RNA in the nucleus and cytoplasm could be used as signals or
mechanisms for RNA transport. The proteins that are associated with
the RNA at various times during the transcription and transport
process have not been assigned any functions. Nuclear RNA seems to
become associated with proteins over its entire length while it is
being transcribed (15). and these proteins seem to have some sequence
specificity. Electronmicrographs of transcription complexes show a
regular pattern of proteins associated along the length of the RNA.
Nuclease digestion studies of polyoma virus RNA suggest that the late

polyoma virus RNAs are also completely covered with proteins; some

 

 

 

18

sequences seem to be extremely resistant to RNase digestion (80).
Similar studies on nuclear adenovirus RNA came to the same conslusion
(59. 81). Studies by Augenlicht (5) have found that the
oligonucleotides protected by protein associations do not have a
unique sequence but do seem to be enriched for guanosine and cytosine
residues and the sequences AGC. AGGC. and GAGC. There is a 78
kilodalton poly-A binding protein that is associated with the
poly-A-tails of nuclear RNA (72). A 60 kilodalton protein seems to
be associated at regular intervals along the poly—A tails of
cytoplasmic RNA (9. 14. 72). Nuclear ribonucleoprotein particles
have been isolated. There are 6 proteins (28-40 kilodaltons)
associated with the 305 "monoparticles" and the multimeric forms of
this basic particle (16. 50. 59. 64. 65. 79). In contrast. three
proteins (120. 76. and 52 kilodalton) are bound to cytoplasmic
polysomal poly-A-containing RNA (50). In addition. most of the
investigations have described extra. higher molecular weight
RNA-associated proteins that are present in low abundance. Several
groups have looked for differences between the RNA-associated
proteins in normal and adenovirus-infected HeLa cells (16. 83. 86).
No major differences have been found. although there is one report of
a protein that is preferentially. but not exclusively. associated

with adenovirus mRNA in the nucleus late in infection (85).

 

 

 

 

19

It has been suggested that snRNAs might play a role similar to the
one suggested for RNA-associated proteins. There is good evidence
for a role for the snRNAs in RNA splicing (23. 24. 36. 58); whether
these or some other RNA(s) are used as a signal for transport is not

known.

The processing and transport of tRNA has been studied in some
detail. The final RNA is clipped from a large precursor (45) and
transported to the cytoplasm by what is probably a carrier mediated
transport system (97). Inhibition of the initial processing events

inhibits both base modification and transport.

The nucleocytoplasmic transport of snRNA has been studied and can be
used as an example of the variety that may exist in mechanisms of RNA
transport. These RNAs are processsed from a large transcript and
transported to the cytoplasm. There they become associated with a
specific set of proteins. and an additional 1-8 nucleotides are
removed from the 3’ end of the molecules. The RNA-protein complex is
then transported back to the nucleus where additional nucleotides may
be removed from the 3’ end (52). This processing pathway must be
vastly different than that used by other cellular RNAs. but it serves
as a reminder that the "RNA transport system" may not be one or even

a few systems.

 

 

20

Conclusion

Several conclusions can be drawn regarding RNA transport from the
nucleus and the types of processing that must take place before
transport can occur. Both mRNA and rRNA must be completely processed
before they can be transported to the cytoplasm; intact
precursor-sized molecules are not found in the cytoplasm. although
there are reports of free introns (29) and external transcribed
spacers from rRNA primary transcripts (31) in the cytoplasm.

Although nuclear RNA stability may be used as a means of controlling
transport in some cases (as in the case of DHFR mRNA). this does not
seem to be a general phenomenon. The same can be said about the use
of transcriptional modulation as a means of controlling transport.
The site of RNA transcription and processing is almost certainly the
nuclear matrix; other areas of nuclear architecture involved in RNA
trafficking and transport have not been described. Because some RNAs
are nuclear restricted. there must be some form of sequence-specific
RNA transport regulation. but to what extent this and other factors
control transport is unknown. Adenovirus infection obviously
presents an excellent system for perturbing nuclear trafficking of
transcripts; it also has led to a genetic system that might be useful
in describing both the factors that are normally involved in this
process and how this process might be taken over during adenovirus

infection. Although the snRNAs and tRNAs obviously have a mechanism

21

of processing different from that of mRNA and rRNA. the fact that
mRNA and rRNA seem to react similarly to adenovirus infection
suggests that. although there is certainly not a single RNA transport

system. there may be only a few such systems.

 

 

 

 

 

 

 

BIBLIOGRAPHY

 

 

 

 

 

 

10.

ll.

BIBLIOGRAPHY

Abulafia. R.. A. Ben-Ze’ev. N. Hay. and Y. Aloni. 1984.

Control of late simian virus 40 transcription by the attenuation
mechanism and transcriptionally active ternary complexes are
associated with the nuclear matrix. J. Mol. Biol. 112:467-487

Agard. D.A.. and J.W. Sedat. 1983. Three-dimensional
architecture of a polytene nucleus. Nature 39;:676-681

Akusjarvi. G.. and H. Persson. 1981. Controls of RNA splicing
and termination in the late adenovirus transcription unit.
Nature 2222420-426

Alt. F.W.. A.L.M. Bothwell. M. Knapp. E. Siden. E. Mather. M.
Koshland. and 0. Baltimore. 1980. Synthesis of secreted and
membrane-bound immunoglobulin Mu heavy chains is directed by
mRNAs that differ at their 3' ends. Cell 29:293-301

Augenlicht. L.H. 1979. Protected nucleotide sequences in
nuclear ribonucleoprotein. Biochem. 18:3780-3786

Babich. A.. L.T. Feldman. J.R. Nevins. J.E. Darnell. Jr.. and C.
Weinberger. 1983. Effect of adenovirus on metabolism of
specific host mRNAs: transport control and specific
translational discrimination. Mol. Cell. Biol. 3:1212-1221

Babich. A.. J.R. Nevins. and J.E. Darnell Jr. 1980. Early
capping of transcripts from the adenovirus major late
transcription unit. Nature 281:246-248

Babiss. L.E.. and H.S. Ginsberg. 1984. Adenovirus type 5 early
region 1b gene product is required for efficient shut off of
host protein synthesis. J. Virol. 59:202-212

Baer. B.W.. and R.D. Kornberg. 1980. Repeating structure of
cytoplasmic poly(A)-ribonucleoprotein. Proc. Nat. Acad. Sci. USA
1:1890-1892

Barrack. E.R.. and 0.8. Coffey. 1980. The specific binding of
estrogens and androgens to the nuclear matrix of sex hormone
responsive tissues. J. Biol Chem. 255:7265-7275

Beltz. G.A.. and S.J. Flint. 1979. Inhibition of HeLa cell
protein synthesis during adenovirus infection. J. Mol. Biol.
131:353-373

22

 

12.

13.

15.

16.

17.

I9.

20.

21.

22.

23.

24.

23

Ben-Ze’ev. A.. and Y. Aloni. 1983. Processing of SV40 RNA is
associated with the nuclear matrix and is not followed by the
accumulation of low-molecular-weight RNA products. Virology
12;:475-479

Berezney. R.. and 0.5. Coffey. 1974. Identification of a
nuclear protein matrix. Biochem. BiOphy. Res. Commun.
gg:1410—1416

Berget. S.M.. C. Moore. and P.A. Sharp. 1977. Spliced segments
at the 5’ terminus of adenovirus 2 late mRNA. Proc. Nat. Acad.
Sci USA 15:3171-3175

Beyer. A.L.. 0.L. Miller Jr.. and S.L. McKnight. 1980.
Ribonucleoprotein structure in nascent hnRNA is nonrandom and
sequence-dependent. Cell 29:75-84

Blanchard. J.M.. and J. Weber. 1981. Nuclear ribonucleoprotein
particles from adenovirus infected HeLa cells. Molec. Biol. Rep.
1:107-113

Bodnar. J.W.. C.J. Jones. D.H. Coombs. G.D. Pearson. and D.C.
Ward. 1983. Proteins tightly bound to HeLa cell DNA at nuclear
matrix attachment sites. Mol. Cell. Biol. 3:1576-1579

Borelli. E.. R. Hen. and P. Chambon. 1984. Adenovirus-2 EIA
products repress enhancer-induced stimulation of transcription.
Nature 312:608-612

Bouche. G.. F. Raynal. F. Amalric. and J.P. Zalta. 1981.
Unusual processing of nucleolar RNA synthesized during a heat
shock in CHO cells. Molec. Biol. Rep. 1:253-258

Brasch. K. 1982. Fine structure and localization of the
nuclear matrix in situ. Exp. Cell Res. 159:161—171

Brawerman. G. 1981. The role of the poly(A) sequence in
mammalian messenger RNA. Crit. Rev. Biochem. 19:1-38

Breathnach. R.. and P. Chambon. 1981. Organisation and
expression of eucaryotic split genes coding for proteins. Annu.
Rev. Biochem. 59:349-383

Calvet. J.P.. L.M. Meyer. and T. Pederson. 1982. Small nuclear
RNA U2 is base-paired to heterogeneous nuclear RNA. Science
211:456-458

Calvet. J.P.. and T. Pederson. 1981. Base-pairing interactions
between small nuclear RNAs and nuclear RNA precursors as
revealed by psoralen cross—linking in vivo. Cell 26:363-370

 

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

24

Camper. S.A.. R.J. Albers. J.K. Coward. and F.M. Rottman.
1984. Effect of undermethylation on mRNA cytoplasmic appearance
and half-life. Mol. Cell. Biol. 4:538-543

Capco. D.G.. K.M. Wan. and S. Penman. 1982. The nuclear
matrix: three-dimensional architecture and protein composition.
Cell 22:847-858

Castiglia. C.L.. and S.J. Flint. 1983. Effects of adenovirus
infection on rRNA synthesis and maturation in HeLa cells. Mol.
Cell. Biol. 3:662-671

Ciejek. E.M.. J.L. Nordstrom. M.-J. Tsai. and B.W. O'Malley.
1982. Ribonucleic acid precursors are associated with the chick
oviduct nuclear matrix. Biochem. 21:4945-4953

Coleclough. C.. and D. Wood. 1984. Introns excised from
immunoglobulin pre-RNA exist as discrete species. Mol. Cell.
Biol. 4:2017-2022

Feldman. L.T.. and J.R. Nevins. 1983. Localization of the
adenovirus Ela protein. a positive—acting transcriptional
factor. in infgcted cells. Mol. Cell. 8101. 3:829-838

Fetherston. J.D.. E.A. Werner. and R.J. Patterson. 1984.
Processing of external transcribed spacer of murine rRNA and

site of action of actinomycin D. Nucleic Acids Res.
1;;7187-7198

Finkel. D.. and Y. Groner. 1983. Methylations of adenosine
residues (m A) in pre-mRNA are important for formation of late
simian virus 40 mRNAs. Virol. 131:409-425

Findly. R.C.. and T. Pederson. 1981. Regulated transcription
of the genes for actin and heat-shock proteins in cultured
Drosophila cells. J. Cell Biol. 88:323-328

Flint. S.J.. G.A. Beltz. and D.I.H. Linzer. 1983. Synthesis
and processing of simian virus 40-specific RNA in
adenovirus—infected. simian virus 40-transformed human cells.
J.Mol. 8101. 161:335-359

Flint. S.J.. M.A. Plumb. U.-C. Yang. G.S. Stein. and J.L.
Stein. 1984. Effect of adenovirus infection on expression of
human histone genes. Mol. Cell. Biol. 5:1363-1371

Flytanis. C.. A. Alonso. C. Louis. L. Krieg. and C.E. Sekeris.
1978. Association of small nuclear RNA with hnRNA isolated from
nuclear RNP complexes carrying hnRNA. FEBS Letters 26:201-206

 

 

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

25

Gaynor. R.B.. D. Hillman. and A.J. Berk. 1984. Adenovirus
early region 1A protein activates transcription of a nonviral
gene introduced into mammalian cells by infection or
transfection. Proc. Nat. Acad. Sci. USA 81:1193-1197

Gerace. L.. and G. Blobel. 1980. The nuclear envelope lamina
is reversibly depolymerized during mitosis. Cell 12:277-287

Giese. G.. and F. Wunderlich. 1982. In vitro ribosomal
ribonucleoprotein transport. J. Biol. Chem. 258:131-135

Goldberg. 6.1.. I. Collier. and A. Cassel. 1983. Specific DNA
sequences associated with the nuclear matrix in synchronized
mouse 3T3 cells. Proc. Nat. Acad. Sci. USA 89:6887-6891

Herlan. G.. W.A. Eckert. W. Kaffenberger. and F. Wunderlich.
1979. Isolation and characterization of an RNA-containing
nuclear matrix from Tetrahymena macronuclei. Biochem.
18:1782-1788

Herz. C.. E. Stavnezer. R.M. Krug. and T. Gurney. Jr. 1981.
Influenza virus. an RNA virus. synthesizes its messenger RNA in
the nucleus of infected cells. Cell 26:391—400

Horowitz. S..A. Horowitz. T.W Nilsen. T.W. Munns. and F.M.
Rottman. 1984. Mapping of N ~methyladenosine residues in
bovine prolactin mRNA. Proc. Nat. Acad. Sci. USA 81:5667-5671

Jackson. D.A.. A.J. Caton. S.J. McCready. and P.R. Cook. 1982.
Influenza virus RNA is synthesized at fixed sites in the
nucleus. Nature 226:366-368

Johnson. P.F.. and J. Abelson. 1983. The yeast tRNAtyr gene
intron is essential for correct modification of its tRNA
product. Nature 39;:681—685

Kao. H.-T.. and J.R. Nevins. 1983. Transcriptional activation
and subsequent control of the human heat shock during adenovirus
infection. Mol. Cell. Biol. 3:2058-2065

Katze. M.G.. Y.-T. Chen. and R.M. Krug. 1984.
Nuclear-cytoolasmic transport and VAI RNA-independent
translation of influenza viral messenger RNAs in late
adenovirus-infected cells. Cell 31:483-490

Kaufmann. S.H.. D.S. Coffey. and J.H. Shaper. 1981.
Considerations in the isolation of rat liver nuclear matrix.
nuclear envelope. and pore complex lamina. Exp. Cell Res.
132:105-123

 

 

 

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

26

Kemp. D.J.. G. Morahan. A.F. Cowman. and A.W. Harris. 1983.
Production of RNA for secreted immunoglobulin u chains does not

require transcriptional termination 5’ to the unexons. Nature
391:84-86

Kumar. A.. and T. Pederson. 1975. Comparison of proteins bound
to heterogeneous nuclear RNA and messenger RNA in HeLa cells. J.
Mol. Biol. 26:353-365

Long. B.H.. C.—Y. Huang. and A.O. Pogo. 1979. Isolation and
characterization of the nuclear matrix in Friend erythroleukemia
cells: chromatin and hnRNA interactions with the nuclear matrix.
Cell 18:1079-1090

Madore. S.J.. E.D. Wieben. and T. Pederson. 1984.
Intracellular site of U1 small nuclear RNA procession and
ribonucleoprotein assembly. J. Cell Biol. 28:188-192

Manley. J.L.. P.A. Sharp. and M.L. Gefter. 1982. RNA synthesis
in isolated nuclei; processing of adenovirus serotype 2
messenger RNA precursors. J. Mol. 8101. 159:581-599

Mariman. E.. A.-M. Hagebols. and W. van Venrooij. 1982. On the
localization and transport of specific adenoviral mRNA-sequences
in the late infected HeLa cells. Nucleic Acids Res.

1Q;6131—6145

Mariman. E.C.. C.A.G. van Eekelen. R.J. Reinders. A.J.M. Berns.
and W.J. van Venrooij. 1982. Adenovirus heterogeneous nuclear
RNA is associated with the host nuclear matrix during splicing.
J. Mol. Biol. 154:103-119

Mayrand. S.. and T. Pederson. 1981. Nuclear ribonucleoprotein
particles probed in living cells. Proc. Nat. Acad. Sci. USA
18:2208-2212

Mirkovitch. J.. M.-E. Mirault. and U.K. Laemmli. 1984.
Organization of the higher-order chromatin 100p: specific DNA
attachment sites on nuclear scaffold. Cell 32:223-232

Mount. S.M.. 1. Patterson. M. Hinterberger. A. Karmas. and J.A.
Steitz. 1983. The U1 small nuclear RNA-protein complex
selectively binds a 5’ splice site in vitro. Cell 3;:509-518

Munroe, S.H. 1982. Ribonucleoprotein structure of adenovirus
nuclear RNA probed by nuclease digestion. J. Mol. Biol.
16;:586-606

 

 

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

27

Neuer. 8.. and D. Werner. 1985. Screening of isolated DNA for
sequences released from anchorage site in nuclear matrix. J.
Mol. Biol. 181:15-25

Nevins. J.R. 1979. Processing of late adenovirus nuclear RNA
to mRNA. J. Mol. Biol. 130:493-506

Nevins. J.R. 1982. Induction of the synthesis of a 70.000
dalton mammalian heat shock protein by the adenovirus EIA gene
product. Cell 22:913-919

Pardoll. D.M.. B. Vogelstein. and 0.5. Coffey. 1980. A fixed
site of DNA replication in eucaryotic cells. Cell 12:527-536

Pederson. T. 1974. Proteins associated with heterogeneous
nuclear RNA in eukaryotic cells. J. Mol. Biol. 83:163-183

Pederson. T. 1981. Messenger RNA biosynthesis and nuclear
structure. Amer. Sci. 62:76-84

Robinson. 5.1.. 8.0. Nelkin. and B. Vogelstein. 1982. The
ovalbumin gene is associated with the nuclear matrix of chicken
oviduct cells. Cell 28:99-106

Rogers. J.. P. Early. C. Carter. K. Calame. M. Bond. L. Hood.
and R. Wall. 1980. Two mRNAs with different 3' ends encode
membrane—bound and secreted forms of immunoglobulin u chain.
Cell 29:303-312

Ross. D.A.. R.-W. Yen. and C.-B. Chae. 1982. Association of
globin ribonucleic acid and its precursors with the chicken
erythroblast nuclear matrix. Biochem. 21:764-771

Salditt-Georgieff. M.. M. Harpoid. S. Chen-Kiang. and J.E.
Darnell. Jr. 1980. The addition of 5' cap structures occurs
early in hnRNA synthesis and prematurely terminated molecules
are capped. Cell 12:69-78

Sarnow. P.. P. Hearing. C.W. Anderson. N. Reich. and A.J.
Levine. 1982. Identification and characterization of an
immunologically conserved adenovirus early region 11.000 M
protein and its association with the nuclear matrix. J. Mo .
Biol. 16;:565-583

Schneider. R.J.. C. weinberger. and T. Shenk. 1984.
Translational regulation by adenovirus VAI RNA: analysis of a
human alpha-globin gene in Va1+/- backgrounds. Abstr. I984 tumor
virus meetings on SV40. Polyoma. and Adenovirus. Cold Spring
Harbor. New York.

 

 

 

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

28

Setyono. 8.. and J.R. Greenberg. 1981. Proteins associated
with poly(A) and other regions of mRNA and hnRNA molecules as
investigated by crosslinking. Cell 24:775-783

Shafit-Zagardo. 8.. F.L. Brown. J.L. Halo. and J.W. Adams.
1982. KpnI families of long interspersed repetitive DNAs
associated with the B-globin gene cluster. Gene 20:397—407

Shafit—Zagardo. 8.. F.L. Brown. P.J. Zavodny. and J.J. Maio.
1983. Transcription of the Kpnl families of long interspersed
DNAs in human cells. Nature 394:277-280

Shatkin. A.J. 1976. Capping of eucaryotic mRNAs. Cell
2:645-653

Smetana. K.. W.J. Steele. and H. Busch. 1963. A nuclear
ribonucleOprotein network. Exp. Cell Res. 31:198-201

Smith. R.C.. E. Puvion. L.A. Buchholtz. and R. Berezney. 1984.
Spatial distribution of DNA loop attachment and replication
sites in the nuclear matrix. J. Cell Biol. 29:1794-1802

Spector. D.J.. D.N. Halbert. and H.J. Raskas. 1980. Regulation
of integrated adenovirus sequences during adenovirus infection
of transformed cells. J. Virol. 36:860-871

Stein. R.. and E.B. Ziff. 1984. HeLa cell B-tubulin gene
transcription is stimulated by adenovirus 5 in parallel with
viral early genes by an Ela-dependent mechanism. Mol. Cell.
Biol. 4:2792-2801

Steitz. J.A.. and R. Kamen. 1981. Arrangement of 305
heterogeneous nuclear ribonucleoprotein on polyoma virus late
nuclear transcripts. Mol. Cell. Biol. 1:21-34

Stevenin. J.. R. Gattoni. P. Keohavong. and M. Jacob. 1982.
Mild nuclease treatment as a probe for a non-random distribution
of adenovirus-specific RNA sequences and of cellular RNA in
nuclear ribonucleoprotein fibrils. J.Mol. Biol. 155:185-205

Stuart. S.E.. G.A. Clawson. F.M. Rottman. and R.J. Patterson.
1977. RNA transport in isolated myeloma nuclei; transport from
membrane-denuded nuclei. J. Cell Biol. 12:57-66

van Eekelen. C.A.G.. E.C.M. Mariman. R.J. Reinders. and W.J. van
Venrooij. 1981. Adenoviral heterogeneous nuclear RNA is
associated with host cell proteins. Eur. J. Biochem.

112:461-467

 

 

 

 

 

 

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

29

van Eekelen. C.A.G.. M.L.H. Salden. W.J.A. Habets. L.8.A. van de
Putte. and W.J. van Venrooij. 1982. On the existence of an
internal nuclear protein structure in HeLa cells. Exp. Cell Res.
141:181-190

van Venrooij. W.J.. T. Riemen. and C.A.G. van Eekelen. 1982.
Host proteins are associated with adenovirus specific mRNA in
the cytoplasm. FEBS letters 145:62-66

van Eekelen. C.A.G.. and W.J. van Venrooij. 1981. hnRNA and
its attachment to the nuclear protein matrix. J. Cell Biol.
§§:554-563

Velcich. A.. and E. Ziff. 1985. Adenovirus Ela proteins
repress transcription from the SV40 early promoter. Cell
50:705-716

Vogelstein. 8.. and B.F. Hunt. 1982. A subset of small nuclear
ribonucleOprotein particle antigens is a component of the
nuclear matrix. Biochem. Biophys. Res. Commun. 105:1224-1232

Wunderlich. F.. G. Giese. and G. Herlan. 1984. Thermal
down-regulation of exportable rRNA in nuclei. J. Cell. Phys.
120:211-218

Wunderlich. F.. G. Giese. and V. Speth. 1984. Thermal
diminution and augmentation of the retention of transportable

rRNA in nuclear envelope-free nuclei. Biochim. Biophy. Acta.
182:187-194

Wunderlich. F.. and G. Herlan. 1977. A reversibly contractile
nuclear matrix. J. Cell Biol. 13:271-278

Yoder. 5.8.. and S.M. Berget. 1985. Posttranscriptional
control of DHFR gene expression during adenovirus 2 infection.
J. Virol. 54:72-77

Yoder. S.S.. B.L. Robberson. E.J. Leys. A.G. Hook. M. Al-Ubaidi.
C.-Y. Yeung. R.E. Kellems. and S.M. Berget. 1983. Control of
cellular gene expression during adenovirus infection: Induction
and shut-off of dihydrofolate reductase gene expression by
adenovirus type 2. Mol. Cell. Biol. 3:819-828

Younghusband. H.B.. and K. Maundrell. 1982. Adenovirus DNA is
associated with the nuclear matrix of infected cells. J. Virol.
5;:705-713

Young. R.A.. A. Hagenbuchle. and U. Schibler. 1981. A single
mouse a-amylase gene specifies two different tissue-specific
mRNAs. Cell 23:451-458

96.

97.

98.

30

Zakian. V.A. 1985. Taken with a grain of salt. Nature
313:223-224

Zasloff. M. 1983. tRNA transport from the nucleus in a
eukaryotic cell: carrier-mediated translocation process. Proc.
Nat. Acad. Sci. USA 89:6436-6440

Ziff. E.B. I980. Transcription and RNA processing by DNA
tumour viruses. Nature 281:491-499

 

 

INTRANUCLEAR RNA TRAFFICKING

INTRODUCTION

Maturation of eukaryotic ribosomal and messenger RNA involves a
variety of processing steps. Production of a cap structure (41).
endonucleolytic cleavage and addition of polyadenylic acid (9).
internal methylation (24). and removal of intervening sequences
from mRNA precursors (10) or transcribed spacers from rRNA primary
transcripts (15) are required for the production of a functional
RNA. All of these processes are intranuclear; obviously. the
additional step of transporting the RNA out of the nucleus is
required before the mRNA or rRNA can be functional in vivo. This
process of nucleocytoplasmic RNA transport is poorly understood.
The completion of the above mentioned maturational events is almost
certainly not the rate-limiting step in RNA transport. since in
many instances (11. 16. 20. 23. 36) the amount of mature-sized RNA
present in the nucleus far exceeds the amount of precursors for
that RNA. This indicates that something other than the completion

of RNA processing is required for transport.

To attack the problem of what. besides "normal" RNA processing. is

required for RNA transport. we have used HeLa cells infected with

31

 

 

 

 

32

adenovirus as a model system in which RNA transport has been
perturbed. Late in adenovirus infection. host cell-specific RNA
species are transcribed (2. 16). polyadenylated and processed
normally (16). However. these sequences are not transported to the
cytoplasm (2. 4. 11. 16). Adenovirus-encoded RNA. which undergoes
essentially the same processing steps as cellular RNA (6. 30. 35.

48). is transported in large quantities at this time (4).

Where is this block to cellular RNA transport? 15 it at the point
of exit from the nucleus. or is there a block at some other step
between the completion of processing and the final departure from
the nucleus? Obviously. an understanding of the physical and
functional relationships between precursor and mature RNAs within
the nucleus is required before a complete understanding of the

transport process is possible.

The physical location of transcription of both rRNA and mRNA
(cellular (12. 38. 39) and adenoviral (31)) has been ascribed to an
intranuclear proteinaceous network called the nuclear matrix. The
matrix and its associated RNA can be isolated from
membrane-denuded. DNA-depleted nuclei by extraction with high salt
concentrations (5. 42). The RNA in the salt-insoluble fraction
(i.e. matrix-associated RNA) consists of both mature and
precursor-sized molecules for mRNA (12. 31. 39. 49) and rRNA (23).

The salt-soluble fraction (i.e. salt-extractable RNA) has been

 

 

'.

 

33

analyzed less thoroughly. 1n the salt—extractable RNA from chicken
oviduct nuclei. Ciejek et al. (12) found 185 and 285 rRNA and
mature-sized mRNA hybridizing to probes specific for ovomucoid and
ovalbumin RNA but no precursor-sized molecules for these species.
Similarly. Ross et al. (39) found only mature-sized globin RNA when
the salt-extracted material from erythroblast and erythrocyte
nuclei was probed with a globin cDNA. These observations suggest

that RNA processing occurs in association with the matrix.

In this report we describe experiments designed to:
1. determine the functional relationship between the

salt-extractable and matrix-associated nuclear RNA pools. and

2. identify the major block to the transport of cellular RNAs

late in adenovirus infecton.

We will show that mRNA and rRNA precursors are processed to
mature-sized molecules in association with the nuclear matrix.
This RNA then changes its state of attachment in the nucleus such
that it can now be removed with 0.5 M NaCl. It is the RNA in this
salt-extractable pool that will be transported to the cytoplasm.
Furthermore. we will show that the block to the transport of
cellular RNA late in adenovirus infection is at or before the
movement of cellular sequences from the matrix-associated to the

salt-extractable pool.

 

 

MATERIALS AND METHODS

Cellggrowth and labelling conditions P K myeloma cells were

3
grown in Dulbecco’s minimal essential medium plus antibiotics and
10% horse serum. HeLa cells were grown in Eagle's minimal
essential medium plus 2 mM glutamine. non-essential amino acids.
antibiotics. and 10% newborn calf serum. Both cell types were
grown in spinner flasks at 37°C in a 5% C02 humidified

incubator. For labelling. cells were concentrated 10 fold (to 5-10
X 106 cells/mL) and incubated with 100—200 uCi/mL 3H-uridine

in normal growth medium for the indicated times.

Cell fractionation and RNA isolation Cell lysis and nuclear
fractionation were essentially the same as those of Ciejek et al.
(12). Cells were pelleted and washed 1-2 times in R58 (10 mM NaCl.
10 mM Tris HCl. pH 7.4. 3 mM MgCl. 50 ug/mL phenylmethylsulfonyl
fluoride. Sigma). lysed either by Dounce homogenization or
treatment with 1% Nonidet P-40 (NP-40). and the nuclei were
pelleted at 700 X g for 4 minutes. The supernatant was treated as
total cytoplasm. Nuclei were washed in RSB. stripped of their
membranes with 1% NP-40 if isolated by Dounce homogenization. and
then treated with 0.2 mg/mL RNase-free DNase I (32) for 30 minutes
at 0°C. Nuclei were pelleted as above. resuspended in R58. and
brought to 0.5 M NaCl with the gradual addition of 5 M NaCl while

vortexing slowly. Nuclei were then incubated at 0°C for 15 minutes

34

35

and pelleted. The supernatant was treated as the salt-extractable
pool. The pellet (i.e. the matrix fraction) was resuspended in
R58. and cytoplasmic. salt-extractable. and matrix pools were
brought to 51 sarkosyl. RNA was isolated as described by

Fetherston et al. (15).

Adenovirus isolationvgnd infection conditions Adenovirus was
isolated using the procedure of Munroe (34). Cells were infected
essentially as described by Beltz and Flint (4). Briefly. cells
were pelleted. and resuspended in medium at 1 X 107 cells/ml.
Adenovirus was added and cells were incubated at 37° C for 1 hour
with stirring. Cells were diluted to 5-10 X 105 cells/mL and
harvested at times indicated. Adenovirus infections were performed
at virus concentrations that gave complete inhibition of actin RNA

transport to the cytoplasm 20 hours post-infection.

Pulse-chase conditions Pulse-label conditions were essentially
those of Beltz and Flint (4). Briefly. cells were pretreated in
medium with 15 mM glucosamine for 45 minutes. and 15 mM glucosamine
plus 0.08 ug/mL actinomycin D for 15 minutes. Cells were

pelleted. resuspended at 1-2 X 107 cells/mL in medium plus 0.08
ug/mL actinomycin D and 0.2-0.5 mCi/mL 3H-uridine. At the end

of the pulse. the cells were added to 5 volumes of cold medium.
pelleted. and resuspended at l X 106 cells/mL in warm medium plus

15 mM glucosamine. 0.08 uQ/ml actinomycin D. 10 mM uridine and

 

 

36

cytidine. Cells were removed at indicated times. added to 2-3

volumes ice cold R58. and pelleted.

_l§smid preparation and filter hybridizations Plasmid DNAs were
isolated by the method of Garger et al. (18). using a 50 Ti Beckman
rotor at 40.000 rpm at 15°C for 15 hr. DNA was immobilized on
filters and pretreated as described by Fetherston et al. (15).
Hybridizations were in 4 X SSC. 0.1% SDS. 10 X Denhardt’s. 10

ug/mL yeast tRNA at 68°C for 48 hours. Filters were washed.
treated with RNase T1 and RNase A and processed as described by
Schibler et al. (40). Total counts in each hybridization were
determined by TCA precipitation of an aliquot of the hybridization
mix. Specific RNA hybridized is expressed as parts per million

(ppm) of this total. Hybridizations over a broad range of RNA

concentrations were performed to insure probe (DNA) excess.

Northern hybridizations. dot blots.g§nd nick translations
Formaldehyde agarose gel electrophoresis was performed as described
by Maniatis et al. (29). and blots were transferred and probed as
described by Thomas (46). Dot blots were performed using a Beckman
dot blot "minifold" and hybridizations were performed as described

by Thomas (46).

 

 

37

Probes The following probes were used in northern blot. dot blot

and filter hybridizations.

i)

ii)

iii)

iv)

v)

pL21-I: cDNA clone specific for immunoglobulin kappa
light chain mRNA (mRNA‘. 45. obtained from W. Salzer.
U.C.L.A.).

p15 a subclone we constructed of pEC‘ (13).

6:
provided by B. van Ness (Fox Chase Cancer Center). The

0.95 kbp BglII-Hindlll fragment of pEC . which is part

k
of the 3.3 kbp intron separating the variable and
constant regions of the mRNAK precursor. was inserted
into the BamHI-HindIII sites of pBR322.

pArDNA: pBR322 clone derived by inserting the 6.7 kbp
murine rDNA in AgtWES MR100 (22. provided by P. Leder.
Harvard University) into the EcoRI site of pBR322.
pXrDNA contains the 3’ terminal 181 bp of 185 rDNA.
both internal transcribed spacers. 5.85 and 285 rDNA
(47).

pilBS: a subclone we constructed from thWES MR100.
The insert in pBR322 contains the 3’-terminal 181 bp of
185 rDNA. internal transcribed spacerl. 5.85 rDNA and
the 5' terminal half of internal transcribed spacerz.
pS’Sal: almost the entire 5’ external transcribed
spacer from the murine rDNA cloned into the Sail site of

pBR322 (33. provided by R. Reeder. Fred Hutchinson

Cancer Research Center).

 

 

38

vi) pAdcos: a cosmid clone of all but the left 101 of the
Ad-2 genome constructed by E. Werner (manuscript in
preparation).

vii) pActin: pBR322 subclone of LK121 clone (supplied by L.
Kedes. Stanford University). which is a human actin
pseudogene clone (14). pActin consists of the 2.0kbp
BglII fragment of LK121 inserted into the BamHI site of
pBR322.

viii) pHe7: cDNA clone of an poly-A containing RNA of unknown
specificity. isolated from a HeLa cDNA library (2.

supplied by A. Babich. Rockefeller University).

RAIL“.

Nuclear Salt—extractable RNA represents an in vivo Nuclear RNA
9991; Nuclei from HeLa or myeloma cells were isolated either by
Dounce homogenization or treatment with 1% NP-40 in RSB. Nuclei
were washed. stripped of membranes with NP-40 if isolated by
homogenization. and then treated with 0.2 mg/mL DNase I for 30 min
at 0°C. The resulting membrane-free. DNA-depleted nuclei were then
extracted with 0.5 M NaCl in RS8. The salt-extractable fraction.
the insoluble matrix fraction which remained. and the cytoplasm
were brought to 51 sarkosyl and layered onto 5.7 M CsCl cushions.
The RNA was then isolated as described by Glisin et al. (21). The

average yields for this fractionation are given in table 1. RNA

 

39

Table 1. RNA recovery after cellular fractionationa

RNA recovered (HQ/106 cells)

Cytoplasm 5.61
MOCK

Salt—Extr 1.05
INFECTED

Matrix 2.00

Cytoplasm 13.6
AD-2

Salt-Extr 2.17
INFECTED

Matrix 1.49

a Average recovery of RNA (in Lag/106 cells) from
three separate experiments. Cytoplasmic RNA (cytoplasm).
salt-extractable RNA (Salt-Extr). and matrix-associated RNA
(Matrix) was isolated from mock- and adenovirus-infected cells 20
hours after infection.

 

 

 

40

removed during DNase treatment and the following wash amounted to
less than 11 of the total and was not analyzed further.
Fractionations performed on cells prelabelled with 3H-thymidine
for at least two cell generations indicate that less than 0.5% of
the DNA copurifies with the RNA in this procedure. Use of
non-RNase-free DNase. the lack of protease inhibitor during RNA
isolation. or any change in the order of individual steps within
this protocol resulted in RNA fractions that had characteristics
similar to total nuclear RNA. as has been described by Kaufman et

al. (27).

One of the obvious caveats that must be considered when separating
components of a biological system is that the components must
represent biologically significant classes. To insure that the
salt-extractable RNA was present as an in vivo nuclear pool and
was not just the result of cytoplasmic contamination that occured
during fractionation. mixing experiments were performed. Myeloma
cells were labeled with 3H-uridine for 4 hours and the cytoplasm
was isolated as described in methods. The labelled cytoplasm was
then mixed with an equivalent number of HeLa cells. which were then
fractionated into cytoplasmic. salt—extractable and
matrix-associated RNA. Aliquots were removed and TCA-precipitable
counts were determined for the three fractions. The remaining RNA
was used in dot blots. starting with equal cell equivalents of RNA

at the highest concentration followed by serial two-fold

41

dilutions. Blots containing equivalent amounts of nuclear
fractions isolated from myeloma cells were used for comparison.
These blots were then probed with nick—translated pLZl-l DNA. which
is specific for immunoglobulin kappa light chain RNA (mRNAK).

The autoradiograph of this dot blot is shown in figure 1. There is
at least a 64-fold difference in the hybridization intensity of the
myeloma fractions compared to the HeLa fractions. indicating I-Zt
contamination of the nuclear fractions with added myeloma
cytoplasm. Contamination of the Hela nuclear fractions by the
myeloma cytoplasm. as estimated by the amount of 3H recovered. is

about 11..

A second. more direct way of determining the amount of cytoplasmic
contamination of the nuclear RNA fractions is to determine the
amount of "cap-2" structures in the nuclear fractions. These
structures are only found in cytoplasmic RNA (41). Using the same
isolation procedure described here. Mary Edmonds (personal
comunication) has found no cap-2 structures in the
poly-A-containing RNA from either of the nuclear populations. The
level of sensitivity of this assay is on the order of 51. These
data argue strongly for the nuclear fractions being essentially

free of cytoplasmic contamination.

Nuclear Salt-extractable pool contains only mature RNA We have

performed experiments to analyze size and sequence specificity of

42

Figure 1. Dot blot analysis of salt-extractable and
matrix—associated RNA. RNA isolated from Hela cells mixed with
cytoplasm from an equivalent number of myeloma cells (lanes 1 and
3). or from myeloma cells alone (lanes 2 and 4) was diluted
serially (two-fold) (rows A—G) and applied to nitrocellulose. Row
H contains 10ug yeast tRNA. The probe was nick-translated pL21-1
DNA. Lanes 1 and 2 are salt-extractable RNA; lanes 3 and 4 are
matrix-associated RNA.

 

43

 

ABCDEFGH

Figure 1

44

the RNA in the various fractions. Hela and myeloma RNA from
cytoplasmic. salt-extractable and matrix-associated pools was
electrophoretically separated on 11 agarose-formaldehyde gels and
stained with ethidium bromide (figure 2a). The RNA from both cell
types shows 285 and 185 rRNA in all three fractions. Only the
matrix-associated pool contains prominent bands with mobilities
slower than 285 rRNA. Similar gels were run. blotted and
hybridized to nick-translated probes specific for rRNA (figure
2b). Only matrix-associated RNA contained hybridizing bands with
relative mobilities of 455. 375. 345. and 325. the mobilities of
the known precursors to rRNA (15). Similar blots of myeloma RNA
probed with nick-translated pL21—1 DNA revealed precursor-sized
molecules for mRNAK only in the matrix-associated RNA (figure

2c).

Because northern blots give no accurate measure of the amount of a
given RNA species. quantitative dot blots were performed using
serial dilutions of salt-extractable and matrix-associated RNA from
myeloma cells. These were probed with nick-translated DNA specific
for intron (p156) or exon (pLZl-l) sequences from the
immunoglobulin kappa light chain gene. or for the external
transcribed spacer (p5'5al) or 185 and 5.85 (pAIBS) sequences from
the rRNA clstrons. Densitometric scans of the autoradiograms of
these dot blots (table 2) indicate that greater than 941 of the

p15 -specific sequences and 951 of the p5'5al-specific sequences

6

45

Figure 2. Size analysis of HeLa and myeloma RNA fractions. HeLa
and myeloma cells were fractionated into cytoplasmic.
salt—extractable. and matrix-associated RNA. and the RNA was
separated on 11 agarose—formaldehyde gels. Gels were stained

(panel A). or RNA was transferred to nitrocellulose and probed with
nick-translated pArDNA (panel B) or pL21-l (panel C) DNA. Lanes:
3. HeLa cytoplasmic RNA; b. HeLa salt-extractable RNA; c. HeLa
matrix-associated RNA; d. myeloma cytoplasmic RNA; e. myeloma
salt-extractable RNA; f. myeloma matrix-associated RNA.

 

46

 

6

2345

6

5

234

 

 

47

Table 2. Relative distribution of introns and exons
in nuclear RNA pools

Probe DNA
p5'Sal pAIBS pIS6 pL21-l
Salt-extr 4.91 32.51 5.91 50.71
Matrix 95.11 67.51 94.1% 49.3%

a Relative distributions of intron and
exon sequences in the nuclear pools were determined
from densitometry scans of autoradiographs of dot
blots. RNA from the salt-extractable (Salt—extr) and
matrix pools from myeloma cells was dot blotted and
probed with pS’Sal (specific for 5’ external
transcribed spacer from rRNA primary transcript).
pXIBS (specific for internal transcribed spacers. 185
and 5.85 rRNA). p15 (specific for intron from mRNA
precursor). and pLZI-l (specific for mRNA exons).
Values given are in percent of the total huclear
hybridization.

48

are found associated with the matrix. Work with chicken oviduct
cells using probes for ovomucoid and ovalbumin mRNAs and rRNA (12).
with erythroblasts and erythrocytes using probes for globin mRNA
(39). and with adenovirus-infected HeLa cells and probes for
adenovirus-encoded RNA (31) found precursors for these RNAs only in
association with the matrix. Ciejek et al. (12) examined the
salt-extractable RNA pool of oviduct cells and found only
mature-sized RNA. These findings. in conjunction with our
observations. suggest that precursors to mRNAs and rRNAs are found
only in association with the matrix. Mature-sized RNAs are found

in both nuclear fractions.

Nuclear Sajt-extrgctable RNA behaves with kinetics expected for

an intermediate in RNA transport. If the salt-extractable RNA is
the pool from which RNA is transported to the cytoplasm. cells
continuously labelled with 3H-uridine should show distinct
labelling kinetics for cytoplasmic. salt—extractable and
matrix-associated RNA pools. Matrix-associated RNA should be
labelled rapidly and should be the first of the three pools to
reach a steady-state specific activity. Salt-extractable RNA
should become labelled after a short lag period (equivalent to the
briefest processing time for matrix-associated RNAs). Cytoplasmic
RNA should be labelled only after a longer lag. corresponding to
nuclear RNA processing time. In addition. the specific activity of

the salt-extractable RNA should reach the same level as the matrix-

49

associated RNA. whereas the cytoplasmic RNA should not reach this

level in any reasonable time because of the presence of long-lived
species such as rRNA. The data from two such continuous labelling
experiment are shown in figure 3; the parallel between the data and

the scheme described above is obvious.

Pulse-chase experiments were also performed to insure that there
was in fact a one-directional flow of RNA from the matrix to the
salt-extractable pool. then to the cytoplasm. If there is a
"precursor-product" relationship between these fractions,
newly-labelled RNA should be found on the matrix at short times. in
the salt-extractable pool at ”intermediate" times. and in the
cytoplasm at long times. Myeloma cells were pretreated with
glucosamine (15 mM) and actinomysin D (0.08 ug/ml) and labelled in
the presence of actinomycin D for 8 minutes. They were then chased
for 2 hours in the presence of actinomycin D. glucosamine. and
excess cold uridine and cytidine. At various times. 6 X 107

cells were removed and fractionated. Aliquots of the isolated RNA
were removed for TCA-precipitation. and the remaining RNA was
hybridized to pL21—1 DNA immobilized on nitrocellulose filters.
Figure 4a shows the data for total counts from one such

experiment. Matrix—associated counts decreased with time.
Salt-extractable counts increased for a period of about 30 minutes
and then decreased. Cytoplasmic counts increased slowly. Figure

4b shows the data from hybridizations of the RNA in figure 4a to

50

Figure 3. Continuous label kinetics 05 RNA populations. Myeloma
cells were labelled continuously with Heuridine. aliquots were
fractionated into cytoplasmic. salt-extractable and
matrix-associated RNA at various times. and specific activities
determined. O———O . cytoplasmic RNA;O—-O. salt-extractable
RNA;H . matrix-associated RNA.

 

51

1.0-

b b b

e e 4

0 O O
>._._>_.—.O< 0.“:0mam w>_.—.<AU¢

0.2-

200

100
MINUTES

 

Figure 3

 

-v we— .4.-

 

52

Figure 4. Pulse-chase kineticsaof labeling of RNA populations.
Myeloma cells were pulsed with H-uridine for 8 minutes and
chased as described in the text. Cells were fractionated at
various times after the initiation of the chase. and the RNA was
used for TCA precipitation (for total counts. panel A) and
hybridization to pL21—l DNA immobilized on nitrocellulose filters
(panel B). 0—0 . cytoplasmic RNA; O—O. salt—extractable
RNA; IF-—-4I. matrix—associated RNA.

15-
10-
e
a.
0
G)
2
5..

 

 

 

MINUTES

Figure 4
Panel A

oHYBRIDIZATION TO pL21-1 (com)

 

54

160*

140 -

120'-

\9\
\FC

80-

40-

 

20_ W

 

0 1 I
O 60 120

MINUTES

 

Figure 4
Panel B

55

pL21-I DNA. This probe is specific for exon regions of the kappa
light chain gene. and because the filters were treated with RNase
after hybridization. only exon-specific sequences are detected in
this analysis. The data resemble those in figure 4a. Matrix
counts decreased with time. while salt-extractable counts increased
and then decreased. After a 60 min lag. the cytoplasmic counts
increased. The total counts hybridized was relatively constant
throughout the chase. indicating that most of the matrix-associated
exon sequences present at the earliest time point were eventually
transported to the cytoplasm. This is in agreement with
Gilmore-Hebert and Wall (20). who found there was no major loss of
kappa light chain exons during pulse-chase experiments in which

myeloma cells were fractionated into nuclear and cytoplasmic RNA.

These data. along with our observation that the salt-extractable
pool contains only mature. non-cytoplasmic RNA. point to the
salt-extractable RNA pool as the intermediate between the

matrix-associated RNA and the cytoplasmic RNA.

Adenovirus infection inhibits the movement of cellular sequences
from the matrix-associated pool to the saltfigxtractablegpool. It
has been reported that late in adenovirus infection. cellular
sequences are not transported from the nucleus to the cytoplasm.
whereas adenovirus mRNA is transported (4). Late in Ad-2
infection. HeLa cells were labelled with 3H-uridine for 2 hours

and fractionated into cytoplasmic. salt—extractable and

11
1|

56

matrix-associated RNA. These fractions (and analogous fractions
from mock-infected cells) were hybridized to filter-immobilized
probes specific for cellular mRNAs (actin and pHe7 mRNA). rRNA. and
adenovirus RNA. The data from these hybridization are given in
Table 3. Late in infection. transport of newly-synthesized actin
and pHe7 RNAs to the cytoplasm was greatly reduced. These RNAs
were detectable at near normal levels in the matrix-associated
fraction. indicating that transcription rates for these species are
not vastly changed from normal levels. as described by Babich et
al. (2). Ribosomal RNA transport was also reduced by a factor of
10. as described by Castiglia and Flint (11). As expected.
adenovirus RNA was transported in large quantities. When the
salt-extractable RNA from these cells was hybridized to these
probes. the amount of newly—labelled cellular RNA detected was

reduced by 8-20 fold relative to control values.

To insure that the loss of cellular sequences in the
salt-extractable pool was not due to some secondary effect of
infection. cells at intermediate times in infection were subjected
to the same experimental procedures. At the intermediate times
examined. the amount of actin RNA transported to the cytoplasm was
reduced 2-3 fold while pHe7 and ribosomal RNA transport was
unaffected. If movement of RNA from the matrix to the
salt-extractable pool is the point of inhibition to transport

during infection. then at intermediate times only sequences with

57

Table 3. Levels of cellular and viral RNAs late in Ad-2 infectiona

RNA HYBRIDIZED

 

Actin pHe7 rRNA AD—2
pm pm ppm ppm
RNA source hybridized hybridized hybridized hybridized
Cytoplasm 146 60 202.000
MOCK
Salt-extr 72 23 221.000
INFECTED
Matrix 8.3 5.3 213.000
Cytoplasm 0 13 18.100 479.000
AD-2
Salt-extr 3.6 2.8 16.900 432.000
INFECTED
Matrix 4.6 4.6 76.100 366.000
CYTOPLASM AD-Zb o 0.22 0.09
CYTOPLASM MOCK
SALT-EXTR A0-2c 0.05 0.12 0.08

 

SALT-EXTR MOCK

a In vivo labelled RNA from mock- and
adenovirus-infected HeLa cells was hybridized to DNA probes
immobilized on nitrocellulose filters. Values for hybridization to
pActin (Actin) and pHe7 are averages of duplicate hybridizations from
two independent experiments. Values for hybridization to pArDNA
(Ribosomal RNA) and pAdcos (AD-2) represent values from a single
experiment where a wide range of RNA concentrations was used for each
probe to insure probe excess. In both cases. similar experiments
gave qualitatively similar results.

bCYTOPLASM AD-Z represents the of amount of a given sequence
CYTOPLASM MOCK transported late in infection relative to
mock values. and indicates the loss in
nucleocytoplasmic transport of that

 

sequence.
CSglt-extr AD-2 represents the of amount of a given sequence
Salt-extr MOCK present in salt-extractable nuclear pool

in infected cells relative to mock
infected cells.

58

reduced transport will show lower levels of that RNA in the
salt-extractable pool. Data from these experiments (table 4) show

exactly this trend.

These data indicate that the inhibition of cellular RNA transport
by adenovirus infection occurs at or before the movement of these
sequences from the matrix-associated pool to the salt-extractable
pool. Furthermore. these results argue convincingly that the
movement of RNA from the matrix to the salt-extractable pool

reflects the i vivo intranuclear trafficking. since loss of

 

transport is always accompanied by loss of movement between nuclear

pools.

59

Table 4. Levels of cellular RNAs at intermediate times
in Ad-2 infection

RNA Hybridized

 

 

Actin pHe7 Ribosomal RNA
PPM ppm ppm
RNA source hybridized hybridized hybridized
Cytoplasm 176 62 58.400
MOCK
Salt-extr 38 13 96.000
INFECTED
Matrix 8.4 3.4 145.800
Cytoplasm 63 60 44.000
AD-2
Salt-extr 11 11 86.900
INFECTED
Matrix 12 11 167.400
CYTOPLASM AD-2 0.36 0.97 0.75
CYTOPLASM MOCK
SALT-EXTR AD-Z 0.29 0.85 0.90

 

SALT-EXTR MOCK

a Levels of cellular RNAs at intermediate times in
Ad-2 infection were determined as in Table 3. except that cells
were harvested at a time when actin RNA hybridization in the
cytoplasm from adenovirus infected cells was reduced 3 fold
relative to controls. Probes and conditions are described in the
legend to Table 3. Values represent those of a single experiment.
Similar experiments gave analogous results.

 

DISCUSSION

In this report. we describe experiments designed to characterize
the nature and function of two nuclear RNA populations that are
separable by salt-extraction of membrane-free. DNA-depleted
nuclei. Further. we have examined the effect of adenovirus
infection on the movement of RNA between these pools and from the

nucleus to the cytOplasm. Two major conclusions can be drawn.

1. RNA is transcribed and processed in association with the
salt-insoluble (i.e. matrix-associated) nuclear pool. Mature
RNA then moves to the salt-extractable pool. and subsequently
is transported to the cytoplasm.

2. Late in adenovirus infection. the movement of cellular
sequences is inhibited at or before the movement of this RNA
from the matrix-associated pool to the salt-extractable pool.
This block is sufficient to explain the inhibition of
nucleocytoplasmic transport of cellular sequences late in

adenovirus infection.

Nuclear RNA partitioning The RNAs associated with the two
nuclear fractions described are different in sequence composition.
size. and labelling kinetics. These differences all argue for the

first conclusion stated above. The fractionation procedure used

60

 

61

here results in very little contamination of nuclear RNA fractions
with cytoplasmic RNA. This was assessed using mixing experiments
and by analysis of the cap structure of the RNA fractions. Both
analyses yielded estimates of contamination of less than 51. All
of this contamination could be explained by the presence of mitotic
cells (which should be present in logarithmically growing myeloma
and HeLa cells at a frequency of about 51) or by a low level of
nuclear disruption during fractionation (28). Furthermore. the
contamination of the salt—extractable pool by the matrix-associated
RNA is minimal (see below). Regardless of its origin. the slight

contamination observed does not alter the basic conclusions.

The composition of the two nuclear populations is markedly
different. The matrix-associated transcripts consist of precursors
and mature species of the RNAs examined. In addition.
hybridizations using probes specific for an intron from the
immunoglobulin kappa light chain gene and the transcribed spacer
from rRNA detected these sequences almost exclusively in the
matrix-associated fraction. In contrast. the salt-extractable
fraction contained only mature-sized RNAs for all of the sequences

examined by us or others (12. 39).

The labeling kinetics of these nuclear populations are different
and are consistent with a precursor-product relationship between

the two. In continuous label experiments. the specific activity of

 

62

the RNA in the two pools reached the same level at steady-state.
Labelling of the salt-extractable pool showed a slight lag period
before beginning to be labelled. consistent with the idea that this
pool receives its RNA. after processing. from some other source.
Labelling of the matrix-associated pool showed almost no lag
period. consistent with previous reports that transcription occurs
in association with the matrix. Results of pulse-chase experiments
extend these observations. Data for total newly-synthesized RNA
and RNA of an individual sequence (kappa light chain exon) showed
that there is a flow of RNA from the matrix-associated pool to the
salt-extractable pool. then to the cytoplasm. Although the total
amount of newly labelled RNA decreases with time. the amount of

exon mRNAK stays reasonably constant.

Throughout this report we have used the term "matrix-associated
RNA" with a meaning equivalent to the term "salt-insoluble nuclear
RNA". There is some degree of uncertainty in the literature as to
the extent. nature and function of the nuclear matrix (see (8) for
instance). The fact that the only way to isolate the nuclear
matrix is by extraction with high salt concentrations makes it
difficult to determine how much of the structure seen after
extraction represents in vivo architecture. Along with
transcription and RNA splicing. the sites of DNA replication (37).
steroid hormone binding (3). and DNA attachment (7. 43) have been

ascribed to the matrix. Further evidence for the existence of some

63

internal nuclear structure comes from the regular spacial
arrangment of polytene chromosomes in Drosophila salivary gland
nuclei during interphase (1). This nuclear RNA fractionation
procedure results in functionally distinct and biologically
significant RNA populations. Since this is essentially the same
fractionation used to isolate matrices. this offers further

indication that these structures are present in vivo.

The matrix-associated and salt-extractable populations both contain
mature-sized RNAs. This does not preclude the possiblity that
there are some other less obvious biochemical differences between
these two pools. It is possible that the matrix-associated
mature-sized RNA still retains some covalently linked remnants of
the splicing process. If these remnants were relatively small. our
analysis would not have detected them. Regardless of these
possibilities. our data define functionally distinct subpopulations

of nuclear RNA.

Adenovirus-induced modulation of intranuclear trafficking The
effect of adenovirus infection on the transport of cellular RNA has
been described in detail (2. 4. 11. 16). Late in infection. the
transport of cellular RNA to the cytoplasm is essentially stopped.
while adenoviral RNA is transported in large quantities. We have
found that at the time cellular sequences are not transported to

the cytoplasm. they are not moved from the matrix—associated to the

64

salt-extractable pool. Transcription of cellular sequences at this
time occurs at near normal levels (2). eliminating this as a
possible cause for the loss of transport. Furthermore. at
intermediate times in infection. when some cellular RNA sequences
are transported and some are not. only those RNAs that are being
transported are found in large amounts in the salt-extractable
pool. Preliminary evidence from several experiments in which the
transport of various RNAs was inhibited to varying degrees suggests
that there is a linear correlation between the amount of
nucleocytoplasmic transport and the amount of RNA in the
salt-extractable pool (data not shown). This is further evidence
that the salt-extractable pool is indeed the intermediate between
the matrix—associated pool and the cytoplasm. It also suggests
that the block to cellular RNA transport takes place coincidentally
with or before this movement takes place. It is possible that the
cellular RNA makes this transition between the nuclear pools and is
degraded quickly after entering the salt-extractable pool. There
are two reports of a decrease in the half-life (by a factor of 2-4)
of cellular sequences in the nucleus late in infection (16. 50).
The authors of these reports acknowledge that this change in half
life is not sufficient to explain the reduction in cellular RNA
transport and suggest that there is some other specific block to
the transport of these sequences. We believe that we have

described this block.

 

65

Different types of RNA are affected differently by adenovirus
infection. Transcription of certain cellular sequences increases
early in infection. although their transcription rates are reduced
to pre-infection levels by late times (19). Histone RNA
transcription is repressed late in infection. consistent with its
cell cycle dependency (17). Similarly. the transport of different
RNA species to the cytoplasm is inhibited at different rates. The
transport of poly—A containing RNA is reduced before an effect on
rRNA is observed (4). Our data suggest that there are even some
differences in the rate of inhibition within the poly-A containing
RNA. since actin mRNA transport to the cytoplasm is reduced 3 fold
before the transport of pHe7 RNA to the cytoplasm is affected. The
reasons for these differences are unknown; however. the effect on

the salt-extractable pool always mirrors the change in transport.

RNA intranuclear trafficking_§nd transport to the cytgplasm The
mechanism that selectively inhibits the intranuclear movement of
cellular sequences while allowing adenoviral RNAs to make this
change remains enigmatic. There seem to be no structural
differences between the adenoviral and cellular RNAs. In fact. RNA
transcribed from an entire o—globin gene inserted into the
adenovirus genome is treated as an "adenovirus message" and is
transported late in infection (Schneider. R.J.. C. Weinberger. and
T. Shenk. Abstr. 1984 Tumor Virus Meeting on SV40. Polyoma. and

Adenoviruses. p. 125). This suggests that the linkage of the

66

sequences coding for the RNA in question is more important than the
sequence of that RNA. RNA need not be transcribed from the
adenovirus genome to be transported late in infection. Transcripts
from the influenza genome (which are made in association with the
nuclear matrix (25. 26)) are transported late in adenovirus
infection (26). Work by Spector et al. (44). who found that RNA
transcribed from the E1 gene of Adenovirus type 5 integrated into
the genome of 293 cells was transported to the cytoplasm late in
adenovirus infection. suggests that RNA sequence may also be

involved.

One possible explanation for this selective transport is that
adenoviral transcription is physically associated with the pool of
a "matrix release factor" that is required for movement of the RNA
between the two nuclear pools. Because of this "position effect".
adenovirus transcripts are preferentially associated with this
signal to the exclusion of all other RNA. Potential "matrix
release factors" are the nuclear RNA-associated proteins and the
small nuclear RNAs. Another possibility is that the
salt-extractable RNA is that which is "in transit" between the site
of processing and the nuclear pore. Obviously. some physical
movement is necessary to accomplish this. The matrix-associated
RNA would then represent a "traffic jam" of RNA waiting to use a
"transportation system" of limited capacity. Adenovirus RNA would

then inhibit intranuclear movement of cellular RNA by usurping the

67

transport system. These models require that a great deal more be

understood about transport in general before they can be tested.

We have characterized a previously overlooked intranuclear RNA
trafficking pathway that explains normal RNA transport and its
disruption by adenovirus infection. The existence of the
salt-extractable nuclear pool as an intermediate between the
matrix-associated and the cytoplasmic RNA pools adds another level
of complexity to nuclear RNA processing. and offers possibilities
for previously unappreciatedlevels of gene regulation. Only when
the biochemical entities involved in this intranuclear movement are
elucidated will the extent of gene regulation at this level be

understood.

BIBLIOGRAPHY

 

10.

11.

BIBLIOGRAPHY

Agard. D.A.. and J.W. Sedat. 1983. Three-dimensional
architecture of a polytene nucleus. Nature 392:676-681

Babich. A.. L.T. Feldman. J.R. Nevins. J.E. Darnell. Jr.. and C.
Weinberger. 1983. Effect of adenovirus on metabolism of
specific host mRNAs: transport control and specific
translational discrimination. Mol. Cell. Biol. 3:1212-1221

Barrack. E.R.. and 0.5. Coffey. 1980. The specific binding of
estrogens and androgens to the nuclear matrix of sex hormone
responsive tissues. J. Biol Chem. 255:7265-7275

Beltz. G.A.. and S.J. Flint. 1979. Inhibition of HeLa cell
protein synthesis during adenovirus infection. J. Mol. Biol.
131:353-373

Berezney. R.. and 0.5. Coffey. 1974. Identification of a
nuclear protein matrix. Biochem. Biophy. Res. Commun.
69:1410-1416

Berget. S.M.. C. Moore. and P.A. Sharp. 1977. Spliced segments
at the 5’ terminus of adenovirus 2 late mRNA. Proc. Nat. Acad.
Sci USA 15:3171-3175

Bodnar. J.W.. C.J. Jones. D.H. Coombs. G.D. Pearson. and D.C.
Ward. 1983. Proteins tightly bound to HeLa cell DNA at nuclear
matrix attachmnet sites. Mol. Cell. Biol. 3:1576—1579

Brasch. K. 1982. Fine structure and localization of the
nuclear matrix in situ. Exp. Cell Res. 149:161-171

Brawerman. G. 1981. The role of the poly(A) sequence in
mammalian messenger RNA. Crit. Rev. Biochem. 19:1-38

Breathnach. R.. and P. Chambon. 1981. Organisation and
expression of eucaryotic split genes coding for proteins. Annu.
Rev. Biochem. 50:349-383

Castiglia. C.L.. and S.J. Flint. 1983. Effects of adenovirus

infection on rRNA synthesis and maturation in HeLa cells. Mol.
Cell. Biol. 3:662-671

68

12.

13.

14.

15.

16.

I7.

18.

19.

20.

21.

22.

69

Ciejek. E.M.. J.L. Nordstrom. M.-J. Tsai. and B.W. O’Malley.
1982. Ribonucleic acid precursors are associated with the chick
oviduct nuclear matrix. Biochem. 21:4945-4953

Coleclough. C.. R.P. Perry. K. Karjalainen. and M. Weigert.
I981. Aberrant rearrangements contribute significantly to the
allelic exclusion of immunoglobulin gene expression. Nature
229:372-378

Engel. J.N.. P.W. Gunning. and L. Kedes. 1981. Isolation and
characterization of human actin genes. Proc. Nat. Acad. Sci. USA
1§:4674-4678

Fetherston. J.D.. E.A. Werner. and R.J. Patterson. 1984.
Processing of external transcribed spacer of murine rRNA and
site of action of actinomycin 0. Nucleic Acids Res.
12:7187-7198

Flint. S.J.. G.A. Beltz. and D.I.H. Linzer. 1983. Synthesis
and processing of simian virus 40-specific RNA in
adenovirus-infected. simian virus 40-transformed human cells.
J.Mol. Biol. 161:335-359

Flint. S.J.. M.A. Plumb. U.-C. Yang. G.S. Stein. and J.L.
Stein. 1984. Effect of adenovirus infection on expression of
human histone genes. Mol. Cell. Biol. 4:1363-1371

Garger. S.J.. O.M. Griffith. and L.K. Grill. 1983. Rapid
purification of plasmid DNA by a single centrifugation in a
two-step cesium chloride-ethidium bromide gradient. Biochem.
Biophys. Res. Commun. 111:835-842

Gaynor. R.B.. D. Hillman. and A.J. Berk. 1984. Adenovirus
early region 1A protein activates transcription of a nonviral
gene introduced into mammalian cells by infection or
transfection. Proc. Nat. Acad. Sci. USA 81:1193-1197

Gilmore-Hebert. M.. and R. Wall. 1979. Nuclear RNA precursors
in the processing pathway to MOPC 21 K light chain messenger
RNA. J. Mol. Biol. 135:879-891

Glisin. V.. R. Crkrenjakov. and C. 8yus. 1974. Ribonucleic
acid isolation by cesium chloride centrifugation. Biochem.
13:2633-2637

Grummt. I. 1981. Specific transcription of mouse ribosomal DNA
in a cell-free system that mimics control in vivo. Proc. Nat.
Acad. Sci. USA 18:727-731

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

70

Herlan. G.. W.A. Eckert. W. Kaffenberger. and F. Wunderlich.
1979. Isolation and characterization of an RNA-containing
nuclear matrix from Tetrahymena macronuclei. Biochem.
18:1782-1788

Horowitz. S..A. Horowitz. T.W Nilsen. T.W. Munns. and F.M.
Rottman. 1984. Mapping of N -methyladenosine residues in
bovine prolactin mRNA. Proc. Nat. Acad. Sci. USA 81:5667-5671

Jackson. D.A.. A.J. Caton. S.J. McCready. and P.R. Cook. 1982.
Influenza virus RNA is synthesized at fixed sites in the
nucleus. Nature 226:366-368

Katze. M.G.. Y.-T. Chen. and R.M. Krug. 1984.
Nuclear-cytoplasmic transport and VAI RNA—independent
translation of influenza viral messenger RNAs in late
adenovirus-infected cells. Cell 31:483-490

 

Kaufmann. S.H.. D.S. Coffey. and J.H. Shaper. 1981.
Considerations in the isolation of rat liver nuclear matrix.
nuclear envelope. and pore complex lamina. Exp. Cell Res.
132:105-123

Long. B.H.. C.—Y. Huang. and A.O. Pogo. 1979. Isolation and
characterization of the nuclear matrix in Friend erythroleukemia
cells: chromatin and hnRNA interactions with the nuclear matrix.
Cell 18:1079-1090

Maniatis. T.. E.F. Fritsch. and J. Sambrook. 1982. Molecular
cloning (a laboratory manual) Cold Spring Harbor laboratory.
Cold Spring Harbor. New York

Mariman. E.. A.-M. Hagebols. and W. van Venrooij. 1982. On the
localization and transport of specific adenoviral mRNA-sequences
in the late infected HeLa cells. Nucleic Acids Res.

19:6131-6145

Mariman. E.C.. C.A.G. van Eekelen. R.J. Reinders. A.J.M. Berns.
and W.J. van Venrooij. I982. Adenovirus heterogeneous nuclear
RNA is associated with the host nuclear matrix during splicing.
J. Mol. Biol. 154:103-119

Maxwell. I.H.. F. Maxwell. and W.E. Hahn. 1977. Removal of
RNase activity from DNase by affinity chromatography on
agarose-coupled aminophenylphosphorly—uridine-2’(3’)phosphate.
Nucleic Acids Res. 5:241-246

Miller. K.G.. and B. Sollner-Webb. 1981. Transcription of
mouse rRNA genes by RNA polymerase I: in vitro and in vivo
initiation and processing sites. Cell 21:165-174

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

71

Munroe, S.H. 1982. Ribonucleoprotein structure of adenovirus
nuclear RNA probed by nuclease digestion. J. Mol. Biol.
16;:586-606

Nevins. J.R. 1979. Processing of late adenovirus nuclear RNA
to mRNA. J. Mol. Biol. 130:493-506

Nevins. J.R. 1982. Induction of the synthesis of a 70.000
dalton mammalian heat shock protein by the adenovirus EIA gene
product. Cell 22:913-919

Pardoll. D.M.. B. Vogelstein. and 0.5. Coffey. 1980. A fixed
site of DNA replication in eucaryotic cells. Cell 12:527-536

Robinson. 5.1.. 8.0. Nelkin. and B. Vogelstein. 1982. The
ovalbumin gene is associated with the nuclear matrix of chicken
oviduct cells. Cell 28:99-106

Ross. D.A.. R.-W. Yen. and C.—8. Chae. 1982. Association of
globin ribonucleic acid and its precursors with the chicken
erythroblast nuclear matrix. Biochem. 21:764-771

Schibler. U.. K.8. Marcu. and R.P. Perry. 1978. The synthesis
and processing of the messenger RNAs specifying heavy and light
chain immunoglobulin in MPC-II cells. Cell 15:1495-1509

Shatkin. A.J. 1976. Capping of eucaryotic mRNAs. Cell
2:645-653

Smetana. K.. W.J. Steele. and H. Busch. 1963. A nuclear
ribonucleoprotein network. Exp. Cell Res. 31:198-201

Smith. H.C.. E. Puvion. L.A. Buchholtz. and R. Berezney. 1984.
Spatial distribution of DNA loop attachment and replication
sites in the nuclear matrix. J. Cell Biol. 22:1794-1802

Spector. D.J.. D.N. Halbert. and H.J. Raskas. 1980. Regulation
of integrated adenovirus sequences during adenovirus infection
of transformed cells. J. Virol. 36:860-871

Strathearn. M.D.. G.E. Strathearn. P. Akopiantz. A.Y. Liu. G.V.
Paddock. and W. Salser. 1978. Characterization of an
immunoglobulin cDNA clone containing the variable and constant
regions for MOPC 21 kappa light chain. Nucleic Acids Res.
5:3101-3111

Thomas. P.S. 1980. Hybridization of denatured RNA and small
DNA fragments transferred to nitrocellulose. Proc. Nat Acad.
Sci. USA 11:5201-5205

47.

48.

49.

50.

72

Tiemeier. D.C.. S.M. Tilghman. and P. Leder. 1977.
Purification and cloning of a mouse ribosomal gene fragment.
Gene 2:173-199

van Eekelen. C.A.G.. E.C.M. Mariman. R.J. Reinders. and W.J. van
Venrooij. 1981. Adenoviral heterogeneous nuclear RNA is
associated with host cell proteins. Eur. J. Biochem.

112:461-467

van Eekelen. C.A.G.. and W.J. van Venrooij. 1981. hnRNA and
its attachment to the nuclear protein matrix. J. Cell Biol.
88:554-563

Yoder. 5.5.. and S.M. Berget. 1985. Posttranscriptional
control of DHFR gene expression during adenovirus 2 infection.
J. Virol. 54:72-77

 

APPENDIX

 

APPENDIX

INTRODUCTION

The following material represents work performed in the laboratory
of Dr. Jack Silver at Michigan State University between January.
1983 and January. 1984. The three sections have been published as
individual reports in the journals NATURE and Science. and I would
like to thank the editors of those journals for allowing me to
reprint these articles here. I would also like to thank my
coauthors on these papers. and especially Dr. Silver for giving me

the opportunity to do this research.

The following were reprinted by permission from Nature.

Thy-1 cDNA sequence suggests novel regulatory mechanism.
Moriuchi. T.. H.-C. Chang. R.M. Denome. and J. Silver. Nature
391:80 Copyright (c) 1983 Macmillan Journals Limited.

Structural organization of the rat Thy-1 gene. Seki. T.. T.
Moriuchi. H.-C. Chang. R.M. Denome. and J. Silver. Nature
31;:485 Copyright (c) 1985 Macmillan Journals Limited.

The following was reprinted by permission from Science.
A hydrophobic transmembrane segment at the carboxyterminus of
Thy-1. Seki. T.. H.-C. Chang. T. Moriuchi. R.M. Denome. H.

Ploegh. and J. Silver. Science 221:649 Copyright (c) 1985
by the AAAS.

73

74

—_

Thy-1 cDNA sequence
suggests a novel regulatory mechanism

g

Teisuya Morluchi, Hsui-Ching Chang. Roger Denome
& Jack Silver

Department oi Microbiology and Public Health.
Michigan State University. East Lansing. Michigan 48824. USA

 

Thy-l was originally defined in mice as a cell-surface alloaurlgen
of thymus and brain with two allelic forms. Thy-1.1 andThy-IJ
(ref. 1). Subsequently, the Thy-1.1 alloantigenic determinant
was identified in rals'. In both species. Thy-1 ls present in
hrge amounts on thymus and brain cells’ and in smaller an-
fldes on ibroblasis‘. epidermal cells’. mammary glands and
Immature skeletal mascle’. In many of these tissues the level
of Thy-1 expression changes dramatically during cell dilerenil-
atlou. The molecules expressing the Thy-1 antigenic deter-
minant have been Isolated from rat and mouse brain cells and
have been shown to have a molecular weight oi 17,500 (ref.
8). One-third oi the Thy-1 molecule is carbohydrate and the
remainder is a polypeptide oi 111 amino acids whose sequence
has been fully deiermined’. We report here the isolation and
characterization oi a cDNA clone encoding the rat thymus
Thy-l antigen bar lad that the DNA sequence ends pre-
maturely at a position corresponding to amino add 103. It
appears to he a complete transcript, however. as the last codon
is followed directly by a poly(A) tract.

Poly(A) containing RNA was isolated from W/Fu rat thy-
mocytes. and cDNA was synthesized by reverse transcription
using oligoidT).,-.. primer. Double-stranded cDNA was
synthesized as described previously”. A cloned cDNA h'brary
was constructed by inserting the total cDNA population into

MXIIJ/OIWJMI N

NATL” VCR ”I 6 IWAIT I”)

 

 

 

 

2 3' '3' '3' ‘1'
6 7- ‘ 1 fl 6 A8 3-
;r r l 3 3 ”a r
A v 4' 4A .. fi

 

a ‘3

' In. 1 Partial restriction map and strategy for wound»; Thy-1
cDNA clone. pT64. The entire cDNA clone was cleaved at selec-
and sites with restriction endonucleases and fragments correspond-
ing to the insert were purified by acrylnmide gel electrophoresis
and e1ecuoelution. Fragments were treated with call intestinal or
bacterial alkaline phosphatase and ”P-labelled a: both 5' ends
with T4 polynucleotide kinase as described”. Alternatively. some
rnpnenu were ”Pelabelled at both 3' ends with DNA poly-
merase”. Double-end-labelled fragments were separated by
aerylarnide gel electrophoresis. clean-eluted and subjected to
partial chemical de lion sequence analysis as described by
hfaxarn and Gilbert . Partially cleaved fragments were separated
on 8. 15 and 20% acrylamide gels. The extent of sequence
deierminedirorneachiragmentisindicaiedbythelengthdihc
arrow.1heclcsedcirciesandverticallinesaioneendofeach
arrow indicate labelling at the 3' end or 5' end. respectively. The
sequencing strategy used here allowed us to sequence both DNA

strands completely.

the Psi! sire oi pBR322 using poly(dG).poly(dC) homopoly-
rneric extensions ’. Colonies containing cDNA were screened‘2
with a synthetic ”P-Iabelled oligodeoxynucleotide (17-rner)
mixture composed of all 32 possible sequence permutations
corresponding to amino acids 82-87 oi Thy-l. Two hybridiz-
ation-positive clones were isolated from ~10,000 colonies.
One clone (pT64) contained cDNA encoding the Thy-1 antigen
and its entire sequence was determined by the procedure of
Maxarn and Gilbert". The overall structure oi the cDNA clone
pT64 and the sequencing strategy used are shown in Fig. l.

The DNA sequence contains a 4S-base S'-untrans1ated region
followed by 57 bases coding for a presumptive leader peptide
of 19 amino acids starting at the first methionine codon (num-
bered -19 in Fig. 2). The leader sequence exhibits several
features characteristic of leader peptides present at the amino
terminus of membrane bound proteins"; it contains many
hydrophobic amino acids (11 nompolar residues) and termin-
ates in a residue with a small neutral side chain (glycine). The
remainder of the predicted amino acid sequence is in agreement
with the published amino acid sequence for rat brain Thy-1’.
There are. however. several unusual features at the 3' end of
the 'Ihy-I cDNA clone. The DNA sequence ends prematurely
at a position corresponding to amino acid 103. 8 amino acids
earlier than predicted from the protein sequence. but isiollowed
directly by a poly(A) tract. There is. however. no termination
codon and, furthermore. a presumptive polyadenylation signal.
$'-AATA.AA-3' (ref. 15). which is pan of the coding sequence,
is found 12 nucleotides upsieam from the poly(A) tract. Two
alternative explanations for these unusual features at the 3' end
can be proposed: (1) a deletion of the sequence encoding the
C-ierminal peptide occurred during the cloning procedure while
retaining the sequence corresponding to the poly(A) tract. or
(2) the AATAAA sequence at positions 293-298 was recog-
hired as a polyadenylation signal in vice and position 310.
which is 12 nucleotides downstream from the AATAAA
sequence. served as a polyadenylation site.

The first explanation cannot be excluded although such a
cloning aneiaci has not been described before. The second
explanation is a very attractive model for a mechanism regulat-
ing rapid changes oi Thy-1 gene expression during cell
diflerenuation and is similar to that observed in other systems.
For example. during differentiation, B lymphocytes undergo a
shift from expression of membrane-bound lgM to secreted IgM.
It has been shown that the transcription unit for secreted and
membrane-bound lgM heavy chains contains two separate
poly(A) addition sites. The basis for selection between these

QICSle-nfiw

7

NAME V“. ”I h JANUAIY I.)

-I9

mm

5

 

twogeneproductsapparentlyresidesinthechoiceoipoly'
denylation site at the precursor RNA followed by diderential
‘mlicing out' of introns"". Expression oi the late adenovirus
transa'iption unit is also controlled through diderential poly(A)

t as l

Wmmu fee as; 23 3c site selection". A similar mechanism may be responsible for
-. t regulating Thy-l gene expression. Ditlerential poly(A) site
Ila out its the lee In ten see val leu gin let an org gly gla ”mm my mm an “FM .d a” C..:mmu adj”
areaecarcscr crccremrcacrerrcucsre rccccwcetcsc fiWflLMlS'PP‘WWY‘MRMNWJ with
to ior membrane integration’ and. in the presence of diiierenti-
anon-inducing tactors such as thymic hormones. complete
:2 3:32;; 33:: 22;; a; 3:3; 82:; $33 :2 cxprcssionoiceil~surface1hy-lmaybeinduced.l(omuroand
Boyse demonstrated that Thy-l negative precursor cells in the
u, m '2': m s! ”I u. u“ ”a u ”o u. :0 u ‘ eh- bonemarrow and spleens of micecanbeinducedtodiflerentiate
" “ 3' "" inviteintoThyelpositivelemphoeyteswithinZhoiincuba-
csc rec ccr at me “7 sac ace sac m on: are me a, “6 rrc tion with crude thymus extract". Our hypothesis might explain
so the mechanism responsible for this rapid induction oi Thy-l

our Ion the erg glu In Iys In his val Ieu ser gly thr Iou giy on the cell guflgce,

acccrcacctaa

$0
walpxogluhta
cr'rccetacose

60
actuccccrcccccc'rcuccrrrrcscr

70 IO

phn 110 Iya val

cscueucucuccrccrcrucccscccrcmc

the trr erg see an val asn Ieu phe see asp neg
use one

In tbs Ian ale asn pha the ehr lys asp qu eg

DNA and amino acid sequence analyses have revealed
homology among the major histocompatibility antigens. fl,-
microglobulin and immunoglobulins. and suggests that all three
have evolved from a common ancestral gene encoding a primi
tive dornainn'”. Homology between Thy-1. which has a
domainolikc structure including a disulphide loop of the

m are “c etc cr'r “'1 cu ccc “5 17c sec see “a“! as: one appropriate size. and immunoglobulins has been noted at the
so amino acid sequence level”. We have compared the nucleotide

ue trr -t m sin in are u! an ab sin In P" thr on Mr sequences oi Thy-l cDNA and the mouse A, light chain variable
N W m “3 W m “A m m “C “c “T “I m ”c ”C region" which was shown to have the highest degree of amino
“,0 to: acid sequence homology to Thy-l”. When the sequences are

nan lye thr tie on val aligned as shown in Fig. 3. overall homology between the

an m set are Mr crc Wu

In. 2 Primary structure and predicted amino acid sequence ior

pT64. Amino acids are numbered relative to the aminooterrninus

oi the protein sequence as determined by Campbell er cl... The

IO-amino acid leader sequence 5' to the codon for the first amino

add is numbered -19 to -l. The hexanucleotide sequence.

AATAAA. at the 3' end is underlined. A 24-base poly(A) tail
directly iollows codon 103.

h. 3 Comparison oithe DNA

aqua-rues of mouse Vat (MOPC Sln Ale Val Ial

nucleotide sequence oi Thy-l and the variable region is 36%.
This sequence homol0gy supports the hypothesis that Thy-l
and immunoglobulin have evolved from a common ancestral
gene. Furthermore. our observations concerning the Thy-l
cDNA clone. pT64, suggests that the Thy-l gene may even
have inherited a remnant of a mechanism involved in regulating
immunoglobulin gene expression. namely diilcrential poly-
adenylation. Additional experiments aimed at defining the
genetic organization and regulating elements of Thy-l
expression are in progress. Thus Thy-l. which has long been
used primarily as a marker for T lymphocyte diilerentiation”.
may itself represent an intriguing system tor the study of gene
expression.

an Gin Clo Sr Ala leu Y» t» See he an Glu it: Val lie ten the

AU us w 7:! cut?! act ACA m ccr an an Act etc ram lC‘l

l ' ' 5 l ii I I ' l

ABC C16 ACA Git I“ (in 616 Mt -

Ser Lev The no Us in Va) Asn
1i)

Cslro
Titre:
« l lt' ::
- (All 56-! CW CUCICHC ICC CB!
Sin Asn Leo hgmAspm g
20

if?)

 

1045) and m Thy-l. DNA "W" "l R? “' Fr. ‘"
sequences were aligned with the lat lay-l CA5 066 676 Ht
amino acid sequence alignment of ii! *9 IL' ”0
Williams er al.” .except for certain l

regions, indicated by parentheses.

that were realigned to increase Set Ser 1hr Sly
homology. Identical amino add "0““ "1 '0‘ '5' ‘5? “G

l
wetsuits:

residues are underlined and ... 7,0,1

nucleotide sequence homologies lits slo Asn ash
are indicated by vertical lines.

Gaps inserted to maximise

homology are indicated by dashes.

The hypervariable regions (HVl. he 1hr at Lee

mmnvnmhmo abovethc lbw "1 VIC t? I. C"
able-constant (V-C) junction U ‘ y u“ s" 21 my

region) '- indicated by a dotted
fine. it 'n interesting that although
the overall nucleotide sequence

homologybetween Thy-l and V.. my in Leu lie

5 36%. COmpgnso' n3 0‘ small seg- but! VAI 565 :i ET A?
ments reveal conssde' table vari- ' . '

. . ltln-lflc citrc
atronmthedegrceoihomology. ‘ ’ mmmm
Thebomologyisn‘leinpcsitions so—
I-ZO. 19% in positions 21-34
(sequences corresponding to the
MI hypervlriablt region ll. ‘l'k n. 1 pm 5 g
'm positions 3545. 38% in pos- no." va‘ mrfr in rir
mouse—ozmvanodzmm l . -;'
poem ' 93—103 (sequences cor. ‘“ "tr-1 it: i: 2:5 EV;

responding to the I region).

Ale Val The in Ser - Asa lyr Ala Asn lrp Val Gin Glu L M Asp Iris Leo
‘P SllACAIl A61 AAClAl GCCMCTEGUCCAA'UI CU SHUT!“

i l ’ i I ll: i H . '! '
ICC Mt m CCC Alt (‘6 CAT W "C ASE C16 ACC C0 06 W MS MS CAL €76

tnr Asn Les 'ro lle tln NU Glu Me'Ser ten in Arg Glu 32 in Lys Hts Vol
I) _ 60 '—

 

an
lle Cl, Gly fhr Asn As» #9 Ala "a Cl lal’ro Ale - -(Ar 'heSe'
streetcaracrmaxcuscrtcalfimm“: tin: its
a ll itll‘ ! ‘ it
. . . . . . . .(ctsicscsrrctcucatnmcéc tcsc
milligfl‘“”“"'lrlis""9
fily Asp Us Ala) - . Ale Lev futile In! Gly Ila thrush. Asp Clu Ala
calmness! sccmm'ncnetsssccrttssl'rwwmatt
I 5 l iii ' ' :13! 7;"
strutscrrrstcussrc militia acutnc sec muimkc
SerAsu

he he lle Lys Val Leu Ti Lev Ala Asa Me "or 1: us is: ‘lu Cl,
‘76 'll

"'3 .-----l‘Eli‘l‘B“ ..........

Ale Lev lru Yyr Ser Asa ms M: m he Sly Gly GI, tr: m Ltu TM rs‘
atmrcsucltuccn resets": ssrsuouuwcrstcrm

‘l . 3 9 ‘i , ' "
Estat m can an 1:6 as: as w ccc ACA act rcc w in in air but sir.
Slu Lev kg Isl % Sly Gin Asa Pro it: See see an US it Its Ash is!

- w —

76

 

'1
W1! “.mnOn-unuz.

I‘LA E..I Aha. 1 MY I on Had I”. IIHJJIINL

MT C. l. u. M ”I. IOM-IMZIImI

M.LEI AICIJ M. V. M’JDIIWI

Sam. P. L Nun K 76-7. H.730

MIL“. ELCMLE LIN. Ll luv “mow-43mm»

ulna-J! A.ngu.ll OM. I has n- Anl In USA 13.09140"
(IflII,

W.J.F.‘ U..-AI. A Man MICLIISII'”)

MP0..MD..NG-IU¢D7.A WWI-xhll“.
Ill-I" ("7”.

WILD 0.6”. “I N IVMI F lush-“l Inls-JOIIUIL

J..
MA..fl-mf C. In- A ll Illa-nu. T. CoITJTO-IIIIIV‘MI.
VIII-Km. L «at but... And in U.$.A 7!. ”27-1111 (107')
..w-nuIJ. m.1..mEH-Ih-knlhxun
.A IUIHUHNII
GO“! .W M Earn U. ‘99- SUIIUO)
Hue [OI-Ls OCmL In. I-uh- 1.33%)?! (INOI.
. 1.. Im let-.6 G Nmuxlll-IIHII‘MI.
Inner-.1 out an N. 101-1”: (I900)
any. I cuLCou n. Ill-JIHIOIOI

MILAN» MN C le‘lD-IIIIIIII

III-urn. K.. I Ionc. E A land I. 140—70 ”’73)

Putnam! A.W. I. A. Ian“! cud-ax. I! ~ an“
in USA a. 1607- 170! mm

Inn-cu. H «cl C4! 1‘. lIS—D‘ ("7!)

tart-mu. D (4 .1 hot. nu; And. Sn USA. 19. 1637-3“! mm.

VIII-u. A. F.. I Guam.) Sana 116. 696-703 IIIII

Ins-M.G.. m. N I Tap-m S CA! ll. 1133-1164091”.

III. M C (m [9. IJ7¢$0III7OI

Milt. PM! F AM. I. MobnnlquII-‘IISICaUSu-‘Hm
Duncan. ~0th 1'32)

 

77

A Hydrophobic Transmembrane Segment at the
Carboxyl Terminus of Thy-l

Abstract. The mode of integration of the glycoprotein thy~l within the cell
membrane has been controversial due to an apparent lack of a transmembrane
hydrophobic segment. Rat and mouse complementary DNA and genomic clones
encoding the thy-l molecule have been isolated and sequenced. These studies have
enabled us to determine the intron-exon organization of the thy-I gene. Furthermore.
they have revealed the existence of a sequence which would encode an extra segment
(3! amino acids) at the carboxyl terminus of the thy-l molecule. These extra amino
acids include a Ill-amino acid hydrophobic segment which may be responsible for
integration of thy.l within the plasma membrane.

ety of III and ”2 amino acids. respec-
tively (9. IO). These sequences were
of thymocytes and brain cells (I. 2). lackingahydrophobic segment.which is
Originally identified in mice. proteins necessary for integration of thy-l within
similar to thy-l are present in many the lipid bilayer of the membrane. This
species (3-6) although the distribution of has prompted speculation that thy-l is
thy-l among hematopoietic cells seems covalently linked to some hydrOphobic
to vary (7. 8). Thy-l proteins isolated component such as glycolipid which an-
from rat and mouse brains have been chors the thy-l molecule to the mem-
sequenced and consist of a protein moi- brane. The DNA sequence analyses re-

Thy-l is a membrane glycoprotein
found predominantly on the cell surface

The nucleotide se- ‘ ‘C

Gin Are Ya: In Ear tau Thr an in Les Va: Ian Gin Aan
as ISO 615 "C AG: CTG AC1 5C5 WC ETC an MC Cit: lac

F3. l.
quence of the cDNA insert of .
the thy-l cDNA clone pT86

and the predict atnino acid 1° to

of med l 1 l "I L“ “D C” In Its Glu tan tan The Ian tau Pro 11a Cite Itta
sequence e ”'"p‘ ‘ “ cu etc on: rat en en as m is: see it: etc etc it: cu: cn
thon antigen. The DNA of

P sdcuf IthRSIC. Cl Fnserta‘l'hrlrgci L Lmt It VltaSeGl Th
- u e u u ya ya ya a a u r y r
M“ "m““cm‘” (BILL ”“d etc it: no: an set can at: as us it: at 6?: crc m one it:
fragments corresponding to

the insert were purified by so so
' Leu 63] Val Pro Glu Ila 7hr Tyr In finer Ir. Val Ian tau FM Ser
””2: wm'mk egrétclfo or: act. on ccc csc cu m it: one it: cc: ct: nc :11 t1: m

“.0:
CT?

Purified fraunents were treat-
ed with call intestinal alkaline
phosphatase (Boehnnget). la-
beled at both 5' ends with T4

ynucleotide kinase plus [7-

lATP (II). and cleaved sec-
ondarily to generate sublrag~
ments with only one labeled
end. The restriction enzyme
sites shown and both Pat 1
sites are those that were la-
beledattheS'eod.'l'hesub-
fragments were separated by
acrylamide gel electrophore-
sis. electroeluted. and subject-

cDNA were sequenced. The
hydrophobic ZO-atnino acid
aeuneot is underlined and the
termination eodoo is indicated
by an asterisk.

10
no Ar; he 2“ Ln Val tau 1hr tau Ila tar. he m- Thr Ln
GAC CGC Tfl aft 1 ACT Ch EEC as: $72 It: ICC as:
Ice 11

00 90

Chi my lap Tyr In Cya Clu tau In lat Ser 617 Gin Aan Pro

NCGGC cat: TIC 1151673161me can or ccc
ha 1

I00 110
Ser- ar Aan tn the lie Ian m lie In lap tn tau Ia) Ln
CC“? at It? at Mt Inca: aaccrccrc us

in I In I I

130
Ct, CU He Ser tau tau Va) Gin nan Tr See Tra tau tau tau
ocraccanacccmmmcuncacrtcc .

"0

‘30
taulauSertauSermtauclnna‘ll'hapheneaertau
.- a-

DD
GIT

Thr
aca

the
to:

lav

a
m

can?” “WIN IOWMICC W emission enact-re
u 961

TC meson mt ccucaccrt tannin mam “an
Ito!

no:

I FEBRUAIY IIS “9

ported here have revealed a h

region (amino acids 124 to “3) which
probably represents the transmembrane
segment responsible for anchor-in thy-l
so the cell. ‘

We previously isolated a rat comple-
mentary DNA (cDNA) clone. pTot.
which encodes pan of the thy-l molecule
(II). Analysis ofthis clone allowed us to
defineasignalpeptideot'lilaminoacids
aswellaspanot'theS'untranslated
region of thy-l. Unfortunately. this clone
terminated prematurely at amino acid
103 which prevented us from defining the
carboxyl end of the molecule. After
screening a second rat thymocyte cDNA
library using the entire Pst linsert ot'
pT64 as a probe. we obtained an addi-
tional thy-l cDNA clone. p136. The se-
quence of the insert from this clone (Fig.
I) startedataminoacid l andwasin
agreement with the DNA sequence of
pT64. Furthermore. it encoded the entire
sequence of lll amino acids ot'the thy-l
molecule which was previously obtained

-19
net

tan Pro his He Ser Isl
scc'rt'rcccuccscscssrccsmmcsscrcrrcccscc src MC cci GCC ATC scc CT: on CTC crc crc res c crscrccccucccrcscccc

78

by conventional protein sequencing
methods. The reading frame encoding
this sequence continued on for an addi-
tional 3t amino acids before a termina-
tion codon. TGA (T. thymine; G. gua-
m'ne; A. adenine). was encountered.
This codon was followed by a 3' untrans-
latedregionol'll9nucleotidesbeforethe
deoxyguanylate-deoxycytidylate (d6-
dC) homopolymeric tail was observed.
Within these 3! amino acids there was an
extremely hydrophobic stretch of 20
amino acids (including six consecutive
leucine residues). which strongly resem-
bled the transmembrane segments found
it other membrane proteins. In order to
determine whether these extra 3! amino
acids were also present in the normal
thy-l gene. rat and mouse genomic Ii-
braries were screened with the thy-l
cDNA clone. and positive clones were
analyzed in terms of restriction enzyme
digestion products and DNA sequence.
The coding sequence of the mouse thy-l
gene (Fig. 2) was distributed among

-0
his tau tau lau See I

three exonszthehrstoneshowntactual-
lythesecondexonolthegenesincean
istronispresent withintheS'untranslat-
edpartol'thegene)encodedpartofthe
5' untranslated region and the first 12
aminoacidsol'thesignalpeptide.This

.waslollowedbyanintronofSNnucleo-

tidesand then by another exon encod-
iig the remainder ol the signal peptide
andamindacids l to l060t'themature'
thy-l protein. After this exon. there was
an additional intron of 386 nucleotides
and an additional exon encoding amino
acids 107 to I43 plus the termination
codon. TGA. A polyadenylation signal.
AATAAA. is located lllO nucleotides
downstream from the termination codon.
The organization ot'the rat thy-l gene is
like that of the mouse gene. including the
presence of an extra 3! antino acids in
the third coding exon (12).
Hybridization analyses of rat and
mouse DNA and RNA ruled out the
possibility of a second thy-l gene encod-
ing a molecule of ll: amino acids or.

 

WWIIEMWRJ 7e.

CCTCICCIGTWCTTTGICITIT‘TIGIIACTCCITMMITCTIGGMmmmwmmlCICICCfiTMTCTCIGCICTCIKIWCIC

in".

TMLTLTI’M‘.‘ IIUIL ILIWUCIWTWI [WWIW IWU

"- I this i LMWIOGIGCIMTTCCIWICIGCCIGIGCTITTCTCIIGITIGICIITCCCTTTCTTCIIIMICCITTTI IIIICII

IIICIIMWCICIWWCWIICICIWICCCIGIIUTCICTI6ITNIWICTTWIOTCTf-f

 

TICCTWGZCSICTGCIAICCCCICITCCCCCCIGCTCAIGCTCIurn ru. run-aw

-1 -1 l

r mt crututc ILCLLIIIquCIImCImwc

““TCII

IWIPFJI

a1 tau Gin Vat Ier Arg Gly Gin Lys Val The Sar tau The Ila Cya tau 'al Ann Gin tan tau lrg tau lab Cys Arg Itta Glu
TCTTGCLGCTGTCCCGAWCIG MOGTGACCAGCCTCICIGCCTGCCTGGTGMCCM MCCTTCGCCTGGACTBCWCCATGAG

Ian Ian 1hr Ln asp sen Ior n. Gln Its Gtu For In tau The Arg Glu Lye Lr'g Ln Its Val tau Irr- 619 The tau 61y Ito
urncsccssccstucrccsrccsccsrcscrrcscccrcsccccscicucsocuccscmcrcrcttzc scccrrccctn

MClultaThrTyrmarArg'alhrtausarsanmnho‘ryrnetys'altauanIaullsdanFneTrir‘l’hrLyabp
wccaccscsccrsccccrccasccrcscccrcrccuccsccccnrsrcssscrccrticccncccascrrcsccsecssocn

I06
GlucuhefifFIeQactuheuma'atwalruslanholtstururunusur near-Val ‘ryrAr-ga
ascmccsctscmrcrcsccrrcucrcrccccccccurcccsrcsccrccursstscrtr:sc'rcrcrsrscsccrcsascr

mmwaflmmmnImmImlTlllllIIIIIIIIIIIIIMMIIMIIICINCLCCTCCITTICCCTTCCCCTMW
I’ h ‘ ram ".11 r I

.acaflamtmrr - i "‘ “ “Anarrcsccssscnurcccsuascscrscscrccr

 

rmaatza“ ‘ “FMMWCWCICWWW

“II3I w ILIUIIIL-m-‘n
m
as Ln tau Isl Ln we 617 as, Its See tau tau Isl Clo nan hr Ier Trs tit lau

rcrccccrccrccmccsccccrrcrcnrccscscscssccrccrcsscrcrwcmcsrtscccrccrccncscucscsrccmsrccrc

"2

1|]
laulautaulauIertauIar-tautauwnstaumhenaaertau 0
mmmmrecacrccmcrccuccccrcucncsnmmmwm ‘‘‘‘

nan-“-
mu m; ll

 

 

' I thnasuwu

ml n [1“ [1 rllh‘l’l’lmt-n“ llz-ll-

WIRmSTBMWImT‘n—

 

MIMI“ - m u l nu. u; u; ljjl‘l u mmmmxpnguwnfilllWIILRWIIIIIIIILIWWTI

 

SWUIIIIWIIIUI ".51qu ‘ mIImIIPJYIIfl'A m. “ “ “1‘“ "AIMF‘J ama.uc

 

 

“AIL- ‘

 

nun» IVIWLWl-Cfi‘m-T-r 7n n_. at

ocrscecmcrmscesrccssrcsmsrcaaccr‘tscsrrcrsccsocrcsrrcrescccsccsacscuccccrercccsreacrcsscssccccccumcccnccc

 

ccsmccscscrcscruscsrccsocrrcscrcm n.- Kamemmuaarammgm

 

 

A“ ...
an“!

- mu-u.m-.mm.mm.mimmcmmumrccrscuecsrccocnccscccsocrncc
usccscscccrrrsrrmmrcccrcraccccrcrcrcccscccccrcrrcercrcccrcsrccccscscscsccscicrcnccustacccrcrrccsscsccrccrn
rerun-r rm- mmwrmrmcflcscmnurwcmcccssmncmmmmcccircsccsrc'rsrur
sccaccscrccucsccsrerciccuacssscunircciwscuiumnmmmmmm

735

II?

929

10 lb

12..

$35.

"53

'5"
I“)

"It

2‘ ‘9
2133
2!"
8M
257$
590

Fig. 2. Nucleotide se-
quence of the mouse thy-
lgene. Agenomic library
was prepared in the cos-
mid vector cZRB ()6)
with mouse (C578U6)
DNA pat'uall' y digested
with San “M. The geno-
mic h'brary was screened
with a nick-translated
thy-l cDNA probe and
two thy-I genomic clones
were obtained. The thy-l
pne was subsequently
subclohed into the
Eco RI site of pBR322
and sequenced (15). The
terminatlbo codon is indi«
cated by an asterisk.

SCIENCE. VOL. 227

alternatively dilerential processing of a
thy-l nuclear ving rise toa
second messenger RNA (mRNA) encod-
flathy- l-nleculeofllZarninoacids.
Onlyasinglethy-lgenewasobservedin
boththeratndrnousegenomeandonly
a single thyvl mRNA species of approxi-
antely (.85 kilobases (lib) in brain and
thymus tissue was detected ([2). This
mRNA hybridized to a nick-translated
DNA fragment corresponding to amino
acids 107 to 143 of the thy-l molecule
([2) and thus 'ncludcd the extra 3| amino
acids observed in the cDNA and geno-
rnicclooes.1hereisnoevidencefora
second smaller mRNA encoding a
shorter thy-l molecule.

The discrepancy between the DNA
sequence data and the protein sequences
of Williams and Gagnon (IO) may be
explained in either of two ways. One
possibility is that. although the purified
thy-l molecule is I43 amino acids long
the hydrophobic peptide containing the
extra 3| amino acids was lost during the
preparation and purification of the pep-
tide fragments that were used for se-
quencing. Altemat tively. the discrepancy
between the DNA and prote tamay
reflect a processing step in l"which the
newly synthesized thy-l molecules are
cleaved to yield mature molecules of a
difl’erent size. To funher explore this
possibility, “pulse-chase" experiments
were performed as follows. Cells from a
murine thytnorm. BWSIA7. which ex-
presses thy-l on the cell surface were
labeled with [”Slmethionine for 5 min-
utes. followed by incubation with unla-
beled methionine for various periods of
time. Included in these experiments was
the compound deoxymannojirimycin
which inhibits the cleavage ofman mannose
residues from N- linked glycans of thy- l
and consequeme simplifies the patterns
observed (U). lmmunoprecipitation of
the thy-l molecule. with a rabbit antise-
rum to rat thy-l antibody ([4) that cross-
reacts with murine thy-l. revealed that
thy-l was initially present in two forms
of dilferent molecular weights and that
within )0 minutes the larger fortn was
convened to a smaller one (Fig. 3). To
determine whether this conversion was
due to cleavqe ofthe carboxyl terminal
31 amino acids don: the thy-l molecule.

terrains] 3| amino acids (at position l24)
and absent from the rest ofthe molecule,
cleavage of this extra 3| amino acid
stretch would result in the production of
an unlabeled thy-l molecule. Although
incorporation of [’Hltryptophan is low.
both molecuhr wcuht forms were visi-
I mauuv ms

79

[u 3] film [’ "lhltopban
‘ ‘g z : \ \-

 

10 d6
omnutas
F1. 3. Biosynthesisof thy-l it the BWSM?
yours-cell line. SW“ 47 cells were
transferredm to methionine- or Iryptophan—free
medium (5 X rot cells per nucroliter). heu-
60 minutes. and exposed to

ophanat shoal
tratsond IN and 250 uCi/ml. respec-
tively After 5 mm es. ctive unino
acidwasaddedtoaconccnttationoflmhl
(zero time point) The oligosaccharide cleav
fie inhibitor deox cin was in-
cludedd uring '
contmuou
(5 X IO‘ cells) were withdrawn at the time
points indicated and processed for immuno-
precipitation lrnrn unoprecrpr'tates were ana
lyzed by sodiu mdodecyl sulfate—polyacrila-
snide gel electrophores is s.(l2 5 percent gel).

ble when thy-l was labeled with
[’HltryptophaMFig. 3): furthermore. the
thy- 1 molecule was visible even after
labeling was followed by an extensive
chase (‘5 minutes in the presence of
unlabeled tryptophan) indicating that the
mature thy-l molecule extends beyond
the IIZ amino acids proposed by Camp~
bell er al. (9) and at least to amino acid
[24. Our inability to detect any other
conversion step even after a 90-minute
chase (data not shown) suggests that the
mature thy-l molecule has a size consist-
ent with that predicted from the cDNA
and genomic data and that its mode of

intention in the membrane is Via the
stretch of 20 amino acrds
present at the carboxyl terminus.
Tersunoat Sen
Cellular and Molecular Biology Unit.
Department of Rheumatic Diseases.
Hospital for Joint Diseases.
New York IMJ
Hsru-Citmc CHANG
Tnsura Moxrucm
thocz IDENOME
Department of Microbiology and Public
Health Michigan State University.
East Lansing 4382‘
Hum: Ptoecu
Institute for Genetics.
University of Co log
Cologne. Federal Republic of Germany
ILVEI
Cellular and Molecular Biology UniI.
Department of Rheumatic Diseases.
Hospital for Joint Diseases
Mm .d N‘-
E. Ieif and J M Allen. Nalure (wisdom
.Trnnsplant. kei. I 32 2M
£1

. , . 4.1.5 l054tll972)
.mne-Mturhearlj R T Act F. Wil-

!>
:1

cnzselandl. W. Fabre. Transplanta-
“liar w Fibre. J. Exp Med. m
)0
rIIIlOrdIndl. Goldschneider I. learnt:

Williams Gapon.$rrenre1l6.996

Mon H C Chang. ll. Denome. J.

ver Nature (Landon: 310nm”)

h. H._( Chang T. uc.hi ll Deno no.mc

Fuhrmann E I'llsg'a‘use.G‘.:m lagler. H. Ploegh.
ibul.’1.755(l%4)

. antiserum to rat thy] antibody was
received from N Ducts and A Williams.
Oxford University.

IS. A ll. Hanan and W Gilben. Nelhals Eno-

nolU IWO)

l6. P. F. Balesandll A SI'ill Genefl. ”70”,).

I7. Supported by NIH grant CAM

9 July ISA; accepled ll October I984

D. Campbell JG KMB Reid A
F. Williams. Durhem I l”. l5(l98l).

A F arth

(

I

80

 

 

 

GAME m 3” 7 RIIL’AIY I.“ WTOMM
Structural organization of A.....z..-..~...-.....‘............-...
I :msmacanmcmaesmermcmcrsmtcasmcm
the rat thy-l gene ,, _
“bumtnmmadluh-l-fba-toahlumlu
mammnmt‘tassutaacassmcmmcstetasur
Tess-noel Sebl'. Tetsuya Morlucblt. ..
Hutu-CblungsanglogerDesomcli-lkifll'fl‘ 2::Rfig‘fizzflmfigfifilz
'CellularandMolecularliologyUniLDepartmcntof mending-msrwmnawmmxmmw
”climaticDiseases.flospitalforloinr0iseases,NewYorb. m“'"“"“‘““"“""°‘°‘m“‘mmm
NnYorblMS.USA to
'DepurtmentofMicrob' _ iologyandPublicHealth.MichiganState 32:}:r,Ԥ3333:;3333
University. East Lansing. Michigan 48124. USA on n
e. w
Niflymmmmaubflnklhflybhuh~
uumrmusmassmmmcxmrcceaa
Tbon a a diluentlatlou marker ex predominantly on .. "”
Meg'rcelk and bull tlssue.ltspreseuceuumurlue hrhramtnhruea-mlllohhmbumlgcyu
pedpberalTeellsbutuothellsbaslougbeesusedtodlstluguisb ficmmmmmwmwmmmm

between these two populations of lymphocytes’. Altbougb
analogues of T'by-l have been described la several mammalian
species". to tissue distribution is dilereut species varies widely”.
precluding its use as T-eell-specific marker. The Tby-l molecule
b a cell—surface glycoprotelu of relative molecular mass lm.
une-tblrd of wblcb represents carbohydrate’; the protein moieties
u! the rat and marine Thy-I molecules” have been sequenced and
found to consist of Ill and "2 amino acids. respectively. Au
unusual aspect of Tby-l is the apparent absence of a bydropboblc
segment comparable to tbat observed in other membrane gly-
eoproteius which would allow integration of Tby-l wltbiu the
membrane lipid bilayer. This has prompted speculation tbat Tby.l
B anchored to the cell surface by some otber hydrophobic com-
ponent such as glycollpld. Here we report the structure of thy-I
complementary DNA and genomic clones and describe the exon-
btrou organization of the gene. More importantly, our data ludi-
cate that Tby-l is initially synthesized as a molecule of I42 amino
ucids. 3| amino acids longer at the carboxyl eud than the Tbyol
molecule isolated and characterized by Campbell a ul". An
extremely hydrophobic region of 20 amino acids lies witbiu this
JI-amlno acid stretch and may represent the transmembrane seg-
ment responsible for anchoring T'by-l to the cell membrane.

We have previously isolated a rat (It '-I cDNA clone. pT64,
which terminates prematurely at amino acid l0}. preventing us
from defining the carboxyl end of the molecule". Screening of
a second rat thymocyte cDNA library using pT64 as a probe
yielded an additional thy-l cDNA clone. p136. The DNA
sequence of the insert (Fig. l) encodes the entire Ill amino
acids of the Thy- l molecule previously obtained by conventional
protein sequencing methods. Surprisingly, however. the reading
frame encoding this sequence continues for an additional 3|
amino acids before reaching a termination codon (TCA). These
31 amino acids contain an extremely hydrophobic stretch of 20
amino acids (note the six consecutive leucine residues). which
strongly resembles segments found in other membrane proteins.
This unexpected observation prompted us to determine whether
these extra 3| amino acids are also present in the normal thy-l
gene.

A rat genomic library prepared in A Charon 30 was screened
with the thy-I cDNA clone. and of several positive clones
obtained. one was selected for further structural analysis. Its
DNA sequence was determined starting ~ l00 nucleotides 5' to
the initiation codon of the gene. By comparing the genomic
sequence with the sequences of the two cDNA clones and
performing 3‘ S. nuclease mapping (data not shown), we
deduced the intron-exon organization of the rat thy-l gene.

The coding sequence is distributed among three exons (Fig.
2) the first one shown (actually the second exon of the gene,
because an intron is present in the S'-untranslated part of the
gene) encodes part of the S‘-untranslated region and the first (2
amino acids of the signal peptide. This is followed by an intron
of 667 nucleotides, then another exon encoding the remainder
of the signal peptide and amino acids l- IOS of the mature Thy-I
protein. Beyond this. there is then an additional intron of 402
nucleotides and an additional exon encoding amino acids 106-

no
Ilrblynulurtmaauulalltahhmrmuutnumluu
mussiummmmmmmm

i). too

Unuuhrhubrhsuuflauufmmhullulurlmu ‘

W

mmmmmmm

. “y

reccmreaccrflqfcicceemcmcauumemcuu
“If

F.. I Nucleotide sequence of the cDNA insert of pm and the
predicted amino and sequence of the complete rat Thyol antigen.
The cDNA library was constructed using mRNA from “NH rat
thymocytes as described previously”. One hybridization-positive
clone (p136) was isolated from ~l0.000 colonies using the entire
Bil insert of the first thy-l clone. pT64. as a probe. A restriction
map was constructed based on the size of DNA fragments obtained
after restriction endonuclease digestion. The entire DNA of p136
was cleaved at seleaed sites with restriction endonucleases (URL)
and fragments corresponding to the insert were purified by acry-
larnide gel elecrrophoresis and electroelution. Purified fragments
were treated with calf intestinal alkaline phosphatase ( Boehn'nger).
”P-labelled at both 5' ends with T4 polynucleotide kinase plus
[tr-”FIAT? as described previously” and cleaved secondarily to
generate subfragments with only one labelled end. The restriction
enzyme sites shown and both Pstl sites are those that were labelled
at the 5' end. The subfragmenu were separated by acrylamide gel
electrophoresis. electroeluted and subjected to partial chemical
degradation sequence analysis as described by Mann and Gil-
ben". Both strands of the insert cDNA were sequenced. The
hydrophobic 20-amino acid segment is underlined and the termina-
tion codon is indicated by an asterisk.

I42 plus the termination codon. TGA. Two polyadenylation
signals. AAUAAA. are located 569 and 1.055 nucleotides.
respectively, downstream from the termination codon. although
only the latter one is actually used for polyadenylation. Thus.
the sequence of the rat thy-l gene is perfectly consistent with
the cDNA sequence and strongly suggests that rat Thyol is
synthesized initially as a polypeptide of I42 amino acids rather
than Ill amino acids We have recently isolated and sequenced
\the mouse and human thy-l genes and find that both also contain
an additional 31 amino acids at the carboxyl end (TS. er al..
manuscripts in preparation).
To eliminate the possibility of a second thy-l gene encoding
a molecule of III amino acids or difierential processing of a
thy-l nuclear transcript which would give rise to a second
messenger RNA encoding a Thy-I molecule of III acids.
Southern and Northern blots were performed. The Southern
blots (Fig. 3) indicate the presence of a single thy-l gene in
both the rat and mouse genome. Similarly. Northern blots (Fig.
4) demonstrate the existence of a single mRNA species of ~ 1.85
kilobases (kb) in brain and thymus tissue. This mRNA hybrid-
izes to a nick-translated DNA fragment corresponding to amino
acids 106- l42 of the Thy-l molecule (data not shown) and thus
includes the extra 3| amino acids observed in the cDNA and
genomic clones. There is no evidence for a second smaller

81

 

 

 

 

 

 

 

 

 

- MTOMM VATUIF V“, ‘l‘ 7 ".3! AIY lOI(
H. 0.
Hh'OO‘IHOhXIOfiI‘IUUWUV
O I ~
I I ".
"Illw'Ulwl “ l n.
w. - ‘ III-t ‘ ‘ ‘ w
rat """" ”I
‘—‘ ...— nun u.
.7 0‘ I
at Ito-Clout brargltyflavgtat Ital-rue
, W m
-“Imm‘l-mhhhlfllumm.l.lihh‘rhmMXIOQIIOOINMM ”
W 9
‘1‘ ‘1 In... hill-haul.“ MIM‘X-fi I ‘01 bah-.7“ '
Wmfi cu 7% I ”flew“ “PM“ °r a”
1 m w rat thy-I gene. Agenomic library was
a”. IOUINIIOU hn.&~ ”'HOHCIIIIIDIMHQWQIMh'U”9‘11.”b. '
W rm ””"d "‘ 'h‘ ‘ "a" 0"” 3°

using a partial Moo] digest of rat

thymus DNA. The isolated gene was

------ __ ---- n)- sequenced according to Matam and
__ .3. Gilbert". The underlined portions
___ ‘ .... represent the exons deduced from
““‘m' "9 the cDNA sequences and the 3'5.

ias nuclease mapping. The protein

“ z u" sequence is numbered from -l9 to

-l (signal peptide) and I-l42. The

two polyadenylation signals at

80' nucleotide positions 2.54 and 2.6“
are indicated by double lines.

 

res
MNMUHM‘DuaA-Nlmml
W

 

‘11
hulk!Dawsflyflyuelavhula-filflnlnohrhrhpmunmhnlaouahrbohrhaloofla
W

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

in
at. is: a» a. no a» Lou 00° Asterisks indicate the termination
' ' ' ' ' ' M, codon.

Manure-rt- - ....Luu (lhmmnmI-nIIKIIKIIKK1 ...nmn um: I"!
mu....mmnmu.um._..u._.“....-.” ... m... m;
m .mcn-rnunenmmn - r- mwtmu ,- nu ---- no.
u m“— ------ --mmmm-nmm m3
mam-‘- lmm'rr-HW“ was: an
Irranmrnmmu ...— .ufintfirrumaetracnmmm mt
MalfunflllmmaHAm I“; 1.xxanrmu.mLu.Lnumx-nun n x M 865
meaacnccut ‘ “ ‘fnrr'amaum - u u. “unsanramrbfamaamatfit a"
warmer .ccs-mF-amrsrma. -‘ - - "wane-"nun- n1 um. “flung.“ “1 I.”
mamccccautmmmm l " '_”‘ _‘ fi‘ unusatlumknmm‘n m
mummummu ”0)

80m Hl

Eco RI
MouseHmd III
MouseHmd Hl
Spleen
lhymus

.E
o
h

a

Rat
Rat

Ft. 4 Northern blotting of

RNA from rat brain. spleen and

thymus. Total cellular RNA was
" '9. kt isolated from rat brain. spleen 2
. ¢ 4 9 kt and thymus using the guani-

diniutn isothiocyanate/caesium
chloride method". RNA no ug)
were electrophoresed through
IS agarose gels containing for-
maldehyde. RNA was then trans-
ferred to a nitrocellulose filter
and subsequently hybridized with
a nick-translated Psrl fragment
of NW». The position of rRNA
was visualized by staining the gel Q
Fu. 3 Southern blotting of DNA from rat and mouse thymus. “‘5 “Vial" 0'1"“ "3d Hindlll
DNA of high relative molecular mass, isolated from rat and mouse fnsmems 0' * DNA "74'5“!“
thymus. was digested with restriction enzymes BornHl. Ecolll and n the 5' "l" “7' “’9‘ as “39
Hutdlll. Each of the digests (10 ug) was electrophoresed through markers.
0 0.7% agarose gel and transferred to a nitrocellulose filter. The
All fragment of prl was nick-translated and used as a probe.
Hybridization was carried out in 5x Denhardt's and 6 XSSC. at
63‘C.

33 i :2;
ll l lll

 

82

 

 

rum: VOL "HM" "'-‘ WTONAM

mRNA encoding a shorter Thy-I molecule. S. nuclease mapping
(data not shown) indicates that the large size of the mRNA is
the result of a large 3' untranslated region “.062 nucleotides)
extending from the end of the coding region to the second
polyadenylation signal.

There are two possible reasons for the discrepancy between
the cDNA and genomic data and the protein sequence data of
Williams and Gagnon'°. First although the purified Thy-l
molecule is indeed 142 amino acids long. the hydrophobic
peptide containing the extra 3) amino acids may have been lost
during the preparation and purification of the peptide fragments
used for sequencing. Alternatively, there may be a processing
step in which the newly synthesized Thy-l molecules are cleaved
to yield mature molecules of a diflerent size. Pulse-chase experi-
ments to explore the latter possibility further failed to reveal
any post-translational proteolytic processing. Furthermore.
when ’H-tryptophan is used for incorporation. the mature Thy-l
molecule is visible even after an extensive (30 min) chase (TS.
et al., manuscript in preparation). Thus. the mature Thy-I
molecule extends beyond the Ill amino acids proposed by
Campbell et at” and up to at least amino acid l23. the position
of the sole tryptophart residue. These data suggest strongly that
Thy-l is anchored to the cell surface by a conventional hydro-
phobic transmembrane segment. It is intriguing. however, that
the intracytoplasmic segment described for other membrane
glyc0proteins is absent and the functional significance of this
remains unknown.

This work was supported by NIH grant CA 38404.

W 20 August. accepted 2 November I”!

. QAEAAllenJMNoMUJZI-SIJHH),

. H C YWr-M 6.5240077”

las.T C I up Add In IOSA-IOIIITTII

rte-Ilium. M.. Anon. I T A Williams. A f leash I “J. Sl-Ol ”974)

.E ".thl F.Mon.l T lhkh.C N I m Md ISLMHMI
clean.

52::

l L d FaMJ ‘ Transplantation 3!. 275-32 (I’ll)

. I. I Fibre.) W l w Md I0. ”b-fil ”'79)

(ll-lo". M I Wt.‘ J I... ”A ”9470 (DID)

W. H. Chaney, ‘ 0 l heel-an. W.) Maire Iain-a 3|. COS-4|) HO“)
WM .1 Minus—tout”):

F Gags-on
ll C-pbell. G-Ga¢lon.1 lent! I H Ammo-LA F WI HIS-”(I'll
t2 Mandi. 1.0.... H Cw D.M.aasha.) “SLOWDHVISI
I). Mann. A It A Gilbert," Hedi Earns sensuous».
A! htaeDonaldJ.’ AIM.N.J.W&SMSM

ppwouevw

J
H
A
D

 

 

 

M'Tlllllllllfllllfl@tfllljillfllyljfillll“