Download A1989R818600002

This Week’s Citation CIassic~
[oa Heljne G. Patterns of amino acids near signal-sequence
cleavage sites.
Eur. J. Biochem. 133: 17-21, 1983.
[Research Group for Theoretical Biophysics, Depanment of Theoretical Physics, Royal Institute of
Technology,_Stockholm,_Swetieni
This paper reported a statistical study or the aminoacid sequences of secretory signal peptides. In
particular, it demonstrated that only small, uncharged
residues are allowed in positions —3 and —1 relative
to the site of cleavage between the signal peptide and
the mature protein. This observation served as a basis
for a scheme that predicts the most likely cleavage
site when only the primary sequence of the precursor protein is known. [The Sd® indicates that this
paper has been cited in over 380 publications.1
Prediction of Cleavage
in Secretory Proteins
G. von Heijne
Department of Molecular Biology
Center for Biotechnology
Huddinge University Hospital
Karoiinska Institute
S-141 86 Huddinge
Sweden
September 23, 1988
_____________________
This didn’t mate sense to me: a
peptide ought to become anchored in the
membrane, most likely with its charged
amino-terminal end remaining in the cytoplasm. I later found out that this was2 the essence of the so-called “loop model.” Fortunately, I didn’t know this at the time, or I
would never have3been drawn into the field
of protein sosting At any rate, this inspired
me to write a paper dealing with the energetics
of a polypeptide chain passing through a lipid
bilayer’
I then got interested in the primary sequences of secreted proteins, and a study of
the then-known signal peptides was a fairly obvious step. Again, I didn’t know that this had
been done
5 before on smaller collections of sequences, and it turned out that my sample
was just the right size for discerning what has
Later become known as the (-3,-1)-rule forthe
cleavage site between the signal peptide and
the mature protein: ouly small, uncharged residues are allowed in positions -3 and -1. As
it happened, an equally well-cited paper’
with essentially the same message was published by 0. Perlman and H. Halvorson within
a few weeks of my paper.
The main reason for the many citations is
that genes and cDNAs for secretory proteins
represent a Large proportion of current
DNA-sequencing efforts. The (-3,-1)-s’ule allows one to inalce a reasonehie prediction of
the site of signal p~pddecleavage in such proteiw. If a few thousand i~ protein sequences
are deduced from their DNA sequences per
year, and if, say, 20 percent of these ~.eissW
secretory proteins, and if a good number of
the papers reporting these sequences cite the
(—3,-1).nde, one is bound to end up with quite
a few cvntiineters of kience Citatfon lndexa
column-space. It is thus simple mass-market
cutrand~wul~
location across the .,.~a,th,.neof the endoplas- effects, rather than profound insight or
mic reticulum (ER). The hyde~Mcsignal theoretical sophistication, that marks the
peptide was shown as somehow squeezing success ofthis CkMion C1~. As for a moral,
tiwougli a likewise hydro,a&iubic munthrane, I guess that the story underlines the wellending up, after cleavage, as a freely soluble documented importance of Ignorance and
peptide in the lumen of the EL
French in all scientific work.
Asa graduate student in ties Blontherg’s Re
search Group for Theoretical Biophysics at the
Royal Institute of Technology in Stockholm,
I had one scientifically very fnaitful idea to
brush up my rusty high-school French. A demanding teacher made me subscribe to La
Recherche, a French popular-science inagazine. Ripping through its puges one day, I
stumbled across a short piece on protein secretion. It described the classic G. Blobel and
I. Doliberstein paper’ that ~..®.n,oted
the first
full-blown version of the signal hypothesis. A
small figure illustrated the main idea a signal
——
I. 3ie~aIG & DobIeraiem B. Transfer of pnxeins across membranes. 1. Presence of pnxcolytically processed toni
ui~rocessedanacent rmmsusoglo&tulis light chains on rneinbraiu-bound nbosonses of storme myeloma. I. CCII ilIoI.
67~g35-5l,l975~(Chad 1.895 times.) (See alser Ikèei 0. Canaan Classic. Cannes Contents/Life Scces
28(1l):1S, (8 Much 1985.]
2. DWhe..o 49*, ~
K & I..uye M. The outer nannbeane proteins of Gram-negative bacteria: biosymbasis,
assembly, and functions. Anon. Rev. Biochem. 47:481-532, (975. (Cited 340 times.)
3. v~Ikt~e G. Transcending the impenetrable: sow proteins come to terms with to nfleanen. Biohaim. Biopimys.
Acm 947:307-33, (988.
4. mu 11e1
G a ISa
C. Trsns-nwntheane translocation of proteins: the direct transfer osodel. &mr. I. Biochem.
97:175-81, (979. (Cited 170 tintes.)
5. Ana B 9* a RIdd B Ii. The signal peplide and its rotc in membrane penetration. B.~em. Soc. Syieposium
46:235-58, 1981. (Cited 15 tintes.)
6. Pne~ B & Ibbrama. IL A putative signal peptidanc recognition site and setluonce in eukaryotic and peobaryotic signal
peptides. I. Mol. 81o1. 167:391-109. (9*3. (Cited 355 linus.)
CIJRHENT CONTENTS®
~—C_ .L~L
©1989by SI®
LS. V.32. #5. Jan. 30. 1989
15