Download Molecular Koch`s Postulates Applied to Microbial Pathogenicity

Molecular Koch's Postulates Applied to Microbial Pathogenicity
Author(s): Stanley Falkow
Source: Reviews of Infectious Diseases, Vol. 10, Supplement 2. Microbial Surfaces:
Determinants of Virulence and Host Responsiveness (Jul. - Aug., 1988), pp. S274-S276
Published by: Oxford University Press
Stable URL: http://www.jstor.org/stable/4454582
Accessed: 24-08-2016 23:57 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to Reviews
of Infectious Diseases
This content downloaded from 128.103.149.52 on Wed, 24 Aug 2016 23:57:47 UTC
All use subject to http://about.jstor.org/terms
REVIEWS OF INFECTIOUS DISEASES * VOL. 10, SUPPLEMENT 2 * JULY-AUGUST 1988
? 1988 by The University of Chicago. All rights reserved. 0162-0886/88/1004-0051$02.00
Molecular Koch's Postulates Applied to Microbial Pathogenicity
Stanley Falkow
From the Department of Microbiology, and Immunolog
Stanford University, Stanford, California
Microbial genetics and molecular cloning now permit us to routinely isolate specific genes
from a variety of microbial pathogens. Obviously not all genes from pathogenic microorganisms play a role in pathogenicity or virulence. Just as Koch's postulates were formu-
lated to identify the causal relationship between an organism and a specific disease, the
notion is presented here that a form of molecular Koch's postulates is needed when examining the potential role of genes and their products in the pathogenesis of infection
and disease.
Bacterial pathogens possess distinct genetic proper-In our laboratory, we have defined these postu-
ties that provide them with a significantly greaterlates
ca- as follows: (1) The phenotype or property unpacity to compete with other bacteria, to gain a footder investigation should be associated with patho-
hold within a specific host, to avoid normal host
genic members of a genus or pathogenic strains of
defense mechanisms, and, once established, to mula species. (2) Specific inactivation of the gene(s) astiply. To the microbial geneticist, the challenge issociated
to
with the suspected virulence trait should
dissect these genes from the more common genetic
lead to a measurable loss in pathogenicity or virutraits found both in pathogens and in nonpathogens.
lence. (3) Reversion or allelic replacement of the mu-
This challenge is being met more and more fretated gene should lead to restoration of pathogequently. Consequently, the classical approachnicity.
of
comparing pathogenic and nonpathogenic isolates
Alternatively: (2A) The gene(s) associated with the
of the same species has given way to precise mutasupposed virulence trait should be isolated by motional analysis utilizing insertional inactivation
or methods. Specific inactivation or deletion of
lecular
site-directed mutagenesis. Molecular cloning permits
the gene(s) should lead to loss of function in the
us to isolate specific virulence genes, to focus on their
clone. (3A) The replacement of the modified gene(s)
function and structure, and to modify them in a for
pre-its allelic counterpart in the strain of origin
cise way. Hence, one can now hope to routinely comshould lead to loss of function and loss of pathogepare bacterial cell lines identical except for a single
nicity or virulence. Restoration of pathogenicity
should accompany the reintroduction of the wildIn some ways, our capacity to isolate genes type
has gene(s).
outstripped the experimental means to document the
These postulates place a heavy burden on an indefined pathogenic property.
essentiality of a given genetic property to pathogevestigator. They insist that genetic manipulation of
nicity. Thus, the simple cloning of a gene considthe microorganism is a prerequisite for success, and,
ered important in pathogenicity, submitting it
of to
course, for some pathogens, such study is not yet
functional analysis, and even sequencing it are not
possible. Moreover, for either alternative, it is essensufficient unless one can rigorously prove that the
tial that the test of pathogenicity be performed with
loss (or gain) of the gene in the species of origin the
has species of origin using a relevant model of
a well-defined effect. In my view it is imperative when
pathogenicity. One must also take into account that
pursuing the genetic analysis of bacterial pathogenit is possible to affect genes associated with pathogeesis to apply some molecular form of Koch's postunicity indirectly; thus, precise characterization of
lates.
structural genes and the information about genes af-
fecting biosynthesis and regulation are mandatory
to an adequate fulfillment of these postulates.
I wish to thank the undergraduate, graduate, and postdoctoral
Because
genetic manipulation is not yet available
for all microorganisms and because there are some
students who have worked in my laboratory for the past two de-
cades for their interminable questions that led in part to this paper.
pathogens, e.g., the gonococcus, for which a relePlease address requests for reprints to Dr. Stanley Falkow,
vant model of pathogenicity is not available, one
Department of Medical Microbiology, Sherman Fairchild Building D307, Stanford University, Stanford, California 94305.could reasonably argue that the induction of spe-
S274
This content downloaded from 128.103.149.52 on Wed, 24 Aug 2016 23:57:47 UTC
All use subject to http://about.jstor.org/terms
Molecular
Koch's
Postulates
S275
subsequently
identified
a second genetic loc
antibody
to
a
defined
g
cific
tralizes
ceptable
Such
in all pathogenic Yersiniaor
that mediates
a
pathogenicity
vir
to and entry into cultured
cells. I
alternative
to epithelial
a
mol
an
sonable
tion Kathleen McDonough
a plasm
approach
has cloned
been
diated adhesin from
the pathogenic
alternative.
To
be membe
sur
It is not clear that
inactivation of
cines
against Yersinia.
several
comm
of these determinants leads to a total loss
of
on
such
an
immunologic
ap
nicity. An antiserum raised
to the
inv gene
and
this
approach
is
still
does appear to be protective.
Yet, how
these
Of
course,
knowing
how
to
p
determinants contribute
individually to pat
and
understanding
microbial
ity and act in concert can be
moreis
readily
necessarily
equivalent.
It
m
ity,
by genetic manipulation than
by antibody n
understanding
micro
ization. Moreover,
it is important
to see that
approach
is
often
not claim to have
molecular Koch's
preferred
method.
Ifulfilled
offer
tw
text
of
genetic
lates for the inv gene since it is arguable th
case.
culture-invasion
First, it is a widely held view that bacterial
pili are model is not a relevant substitute
for
invasion
of whole animal tissue.
the means by which many microorganisms achieve
The Henle-Koch
postulates were intended as a
their specific adherence to host cells in order
to ocmeans of identifying the causal relationship between
cupy a unique niche. Indeed, there are considerable
an organism of
and a specific disease. Koch recognized
data showing that in some cases immunization
that the dispostulates were not rigid. The limiting facanimals with purified pili is protective against
tor even
in Koch's time was the lack of suitable exease. The molecular analysis of bacterial
adhesins
animal models in which human disease
has provided a more specific view. For perimental
the enteric
could the
be reliably
bacteria, which cause urinary tract infection,
spe- reproduced. Moreover, the inability to grow
presumed pathogens, e.g., the leprosy bacific adhesin employed for binding to a specific
host
cillus, was
as relevant in Koch's time as it is now. The
cell receptor is actually a gene product distinct
from
evolution
thought regarding Koch's postulates is
that specifying the pilus subunit. It is likely
thatofthe
documented by Evans [1]. Thus, the discovpilus and the specific adhesin are closely nicely
associated,
ery
of viruses and their role in disease underscored
and then often are co-purified. As the
molecular
the limitations
of the original postulates. It became
analysis of the bacterial adhesins employed
by enclear, forthat
example, that the establishment of a given
teric bacteria progresses, one begins to believe
virusdirectly
with a disease syndrome was possible only unplasmid-mediated pilin molecules may bind
der certain picircumstances which included the imto their receptors while chromosomally mediated
munologic
status
of the host. Similarly, Evans traces
lin molecules form a scaffolding for an adhesin en-
establishment
of "Koch-like" postulates to the
coded by a different structural gene. No the
matter,
the
immunologic
available genetic and molecular data now
paint a proof of disease causation; to the
search for and
the association of slow viral infection and
more precise picture of the pathogenic process
neurologic illness; and, more recently, the
provide us with more specific targets for chronic
the achieve-
ment of protection against disease.
causal relationships between viruses and cancer as
wellfindings
as the causative factors in certain chronic disAs another example, I offer some recent
eases brought
from my own laboratory regarding bacterial
inva- about by exposure to toxins. Clearly,
sion. We recently reported the isolation the
of postulates
a single in their original form or as modified
for a given situation will only be as good as the availgenetic locus, inv, from Yersinia pseudotuberculoable coli
technology.
sis, which when transferred to Escherichia
K12 By the same token, it must be posto modify the postulates as new technology perpermits this innocuous microorganism tosible
effectively
mits
us toby
examine new aspects of the pathogen and
enter eukaryotic cells. We inactivated this
gene
theit
host.
a well-defined mutation and exchanged
for the
I was asked
to present an overview of the genetic
wild-type allele, to yield a Y pseudotuberculosis
invand molecular
means by which one may investigate
strain. However, this derivative still exhibits
low-level
pathogenicity that could serve as a founentry and strong adhesion to the surface microbial
of cultured
dationMiller
for the discussion of the exciting developmammalian cells. In my laboratory, Virginia
This content downloaded from 128.103.149.52 on Wed, 24 Aug 2016 23:57:47 UTC
All use subject to http://about.jstor.org/terms
S276
Falkow
ments
tak
the logic or wish to define their
own criteria. For me
these
postulates
underscore
the
need
to define and
pathogens
the
metho
refine genetic exchange mechanisms among microorganisms and for new and better animal models or,
bacterial
preferably, alternative models
of microbial infection;
model
of
I believe these are as much the keys
to success in unlecular
bio
study
derstanding microbial pathogenicity as is molecular cloning. The only point I wish to assert unequivoimal disease. In fact, the study of pathogenicity at cally is that it is no longer necessary to speak of the
the genetic and molecular level has become increas- promise of molecular biology and genetics for the
ingly refined over the past several years. To meet this study of pathogenic microorganisms. That promise
growing sophistication, it is necessary to redefine a is already being fulfilled.
the
of those microbes that contribute to human and an-
set-of acceptable criteria that can be applied to the
analysis of microbial pathogenesis. The molecular
Koch's postulates described here were not intended
to be anything more than an attempt to provide the References
basis of a dialogue among interested investigators. 1. Evans AS. Causation and disease: the Henle-Koch postulates
There is little doubt that some will find fault with
revisited. Yale J Biol Med 1976;49:175-95
This content downloaded from 128.103.149.52 on Wed, 24 Aug 2016 23:57:47 UTC
All use subject to http://about.jstor.org/terms