Bacteria have been the favourite tool of geneticists and molecular biologists for a number of years because of their simplicity.
This simplicity is both in structure and in their genetics, and also in their life cycle.
They’re not only the simplest cellular organism, but extremely small in size and with other properties of rapid division and so forth
that make them ideal objects for study in the laboratory.
If I could have the first slide.
I’ll show a scanning electron micrograph of a typical bacterial cell, one that we study in our own laboratory, haemophilus influenza.
Now, for the purposes of my lecture, let me describe, very briefly, some of the characteristics of these bacteria.
First of all, bacteria can be divided into two major categories based on the structure of their cell envelope.
We have the gram-positive bacteria that have an outer boundary consisting of a cytoplasmic membrane and a rather thick cell wall.
On the other hand the gram-negative bacteria have, in addition to these two layers, an outer membrane.
So that when we talk about gene transfer, we have to think of the problem of getting genes, that is DNA,
out of one cell, through these layers, and into a recipient cell.
Now the bacteria also is rather simple inside.
There is no true nucleus, the chromosome is in direct contact with the cytoplasm and genetic expression occurs in a coupled fashion,
with the genes being transcribed and then directly and immediately translated into protein.
The chromosome is a single molecule of some one hundred, some million or so base pairs,
and the nucleotide sequence, of course, carries the complete genetic programme
which specifies not only the structure of the bacterial cell, but also the complete life cycle.
Perhaps we could show the first slide.
The nucleotide sequence itself which carries this genetic information is organized into a series of contiguous units, which we call genes.
And in a typical bacterial cell there might be some three thousand such genes
which play out the genetic programme of this organism every twenty or thirty minutes as the cell divides.
Here is the haemophilus influenza cell that we study and you should keep this in mind for future reference.
These cells are only about one or two microns in size.
Now, bacteria in general, if one looks in nature, grow in extremely large populations.
And consequently, and they also divide very rapidly, a matter of twenty or thirty minutes for a cell, per doubling time.
In addition, these large populations collectively carry an enormous variety of mutations.
So that in adapting to their environment they can select for favourable mutations
and in addition they have developed a variety of means of exchanging genes between cells
so that they can play with various combinations of mutations in order to develop the most adaptable organism.
So that many modern biologists, because of these reasons, have considered that bacteria may be the most highly evolved organism.
This is not to mean that they are the most complex in structure, but genetically the most evolved.
They have a genome with the highest density of genetic information of any that we know, other than the viruses.
If I could go to the next slide then.
Let me go over the known mechanisms for gene transfer in bacteria, as this is the subject of the lecture.
We have transformation, transduction and conjugation.
In transformation, a donor cell in the population releases its DNA into the medium by lysis or perhaps in some cases by a secretion mechanism,
and other cells in the population act as competent recipients.
That is they have become able to take up DNA from the medium into the cell.
In transduction, you have a similar transfer from donor to recipient, except that the vector for transfer is a virus,
which in a small frequency of the cases will package a piece of the bacterial DNA rather than phage DNA.
In conjugation, we have a highly specialized mechanism in which there is an actual bridge between the donor and the recipient cells
and there is a plasmid-mediated linear transport of the donor chromosome, or a copy of the donor chromosome into the recipient.
Now in all these cases, once the DNA gets into the cell, into the recipient cell, if it contains homologous sequences,
it can be recombined into the recipient chromosome to form recombinants.
And this is ordinarily by several different biochemical pathways that all cells carry that enable them to recombine homologous sequences.
My lecture will deal only with the transfer mechanism, and not with what happens in the cell after the DNA gets in, as this is fairly uniform.
Alright let me then talk a little bit in more detail about each of these mechanisms.
And I’ll start with transduction which in many ways is the most universal transfer mechanism
because, as far as we know, all bacterial cells can be infected by at least some viruses, which are capable of the transduction mechanism.
Let’s go to the next slide.
And I want to concentrate really on the most salient features for this particular comparison.
Here we have the essential mechanism for generalized transduction.
In this case the virus itself replicates its DNA in the form of a tandem polymer
in which individual viral chromosomes have been joined end to end,
either by recombination or in the process of a rolling circle type of replication.
And the mature virus is formed by packaging genome units sequentially along this polymer.
Much as you would take an egg shell and stuff a long string into it.
But the important feature is that the headful packaging starts from a particular site,
identified by a nucleotide sequence which is called the “packed sequence” which occurs in each genome.
But more or less randomly the packaging will start at one of these sites and then continue along,
packaging slightly more than a complete genome unit without the requirement for additional recognition of the pack side as you go down.
It’s only the initial one which seeds the packaging mechanism.
Now this works very accurately in the cell so that the vast majority of packaged pieces of DNA are viral.
However there are mistakes built into the mechanism.
Bacterial DNA itself contains a few sites which are very similar but perhaps not identical to these sites, and this fools the packaging mechanism
so that occasionally viral packaging will start on the bacterial chromosome and proceed sequentially,
thus forming particles which contain bacterial DNA, and these are transducing particles.
The other way that mistakes are made are by actual mutations within the viral protein, present on the phage head, which recognizes a site.
And this, we know that these mutations play a role in the formation of transducing particles
because in certain single bursts one sees a variety of transducing particles formed, originating from a number of sites in the bacterial cell.
Now, if I could go to the next slide.
I show the mechanism of specialized transduction.
In this case, we have a virus which is integrated as prophage into the bacterial chromosome with neighbouring genes on either side,
and when the virus induces at some subsequent time and begins to replicate, it must excise itself from the bacterial chromosome.
Normally it would do this with great accuracy so that the excision occurs precisely at the two ends of the virus.
But about one percent of the time, an error is made and the virus loops out and excises so as to lose some of the viral genes
and gain some of the neighbouring bacteria genes.
Thus you form a recombinant molecule containing virus and bacterial DNA.
This can be packaged into a viral coat and then can subsequently infect another cell in the population
and transfer these donor genes into the recipient cell.
That’s nature’s way of doing recombinant DNA and it’s been known for some twenty to thirty years.
Now let me emphasise in both cases that the transducing particles arise
as a by-product of the normal replication in life cycle of these bacterial viruses.
And the question arises as to whether nature could have evolved a more accurate process so that you would not make so many mistakes.
But on the other hand you have to ask the question as to whether nature has found it more desirable
to design a certain number of errors into the system, for the benefit of the host cell
because ultimately the host cell must survive in the environment if the virus itself is to live.
Let’s go to the next slide which will give us a brief picture of conjugation.
Now this is actually quite a complicated mechanism.
It’s one that is not yet clearly understood although we have a general picture of it.
Here is the classic F factor mating cycle, and it starts with a bacterial cell, typically E-Coli containing the F factor,
which is a plasmid, in supercoiled form, as shown by this figure of 8.
Now the plasmid itself can replicate inside the cell and maintain itself in this cell and its daughter cells.
But in addition, it has a mechanism to spread horizontally in the population, and this is the conjugation mechanism.
The plasmid specifies, genetically specifies the synthesis of a tube-like pilus on the cell surface
which contains specific recognition proteins at its tip that can interact with a so called female cell, which does not carry the plasmid.
So you have here a donor or a male cell and a recipient, or female cell in contact.
And when contact is made, a single break occurs at a specific origin in the plasmid, and thus relaxes the supercoil.
We also have to imagine that the plasmid molecule is attached, probably in the vicinity of that nick at the base of the pilus.
Then over a period of a few minutes, the pilus retracts by a mechanism that is not understood,
and the cells are drawn together, so that an actual contact or bridge can be made.
Replication of the plasmid then ensues from the original break in this one strand, and you have a linear transfer of the plasmid into the female cell
and by DNA replication in the recipient, one forms a double helix and the molecule is rejoined to generate again the supercoiled molecule.
And the cycle is complete.
So that one has now, the plasmid has now successfully transferred itself into a previously non plasmid carrying cell in the population.
In this way, one can have a very rapid infective process in a cell population.
If one adds only a few fertile donor cells to a culture of female recipients,
within a few hours the entire population can be converted to plasmid carrying cells.
On this slide we see the genetic structure of the F plasmid.
The complexity of the conjugation process is mirrored in the number of genes that are required to specify the process.
There are some nineteen genes that have been located in this segment of DNA called TRA for transfer genes.
And these genes not only specify the pilus structure but also specify a new and independent replication from this transfer origin,
this replication mechanism being quite separate from the replication mechanism which maintains the plasmid in a particular cell.
Now what I have described so far is simply the normal reproductive process for this plasmid,
but the plasmid can also transfer bacterial genes under rare circumstances.
And it does this by literally incorporating the entire bacterial chromosome into a particular site on the plasmid.
And the incorporation occurs at particular sequences which occur also in the bacterial chromosome,
so that one has homology between the plasmid and the bacterial chromosome at specific points,
allowing a genetic recombination and co-integration of the two structures together.
Then when the transfer occurs by the mechanism I described, the bacterial chromosome which now is part of this plasmid,
is simultaneously transferred in a linear fashion.
And so we have then the very useful transfer of bacterial genes which is of great benefit to the bacterial cells themselves.
And one can, again, use the same argument, that although the mechanism is primarily for the benefit of the plasmid,
it also has designed into it features which enable the bacteria to gain benefit and survival
which is useful to the plasmid in a secondary way because the bacterial cell is a host for that plasmid.
Now let me move onto transformation on the next slide.
And here I want to describe the mechanism separately for the gram-positive cells as opposed to the gram-negative cells.
Because we have in these two types of bacteria, a difference in the cell envelope
which apparently has made it necessary for nature to evolve two separate mechanisms
that are designed to allow naked DNA to penetrate the cell envelope and undergo recombination.
The gram-positive transformation mechanism has been studied for some fifty years and only in the past few years have we begun to understand,
through the work of a number of laboratories, the details of the mechanism.
But I can make it very clear that it is not understood at a biochemical level.
What we have really is a molecular description of some of the events.
The slide here illustrates transformation for pneumococcus, which is a most widely studied organism.
But the features hold also for the other gram-positive organisms, for example bacillus subtilis and a number of the other streptococcal strains.
There are two stages to the transformation process.
First we have confidence development which involves the induction of certain changes in the cell
which allow it to become permeable or to transport DNA.
As the pneumococcal cells grow in a broth medium, they elaborate an activator molecule which is secreted into the medium.
As the population of cells increases in density, the amount of activator molecule builds up in concentration
until it reaches a critical level at which the entire population of cells is induced to competence.
The induction mechanism itself involves the binding of the activator to a membrane receptor,
much as a hormone would act in a eukaryotic organism.
This process, by unknown means, causes the induction of a series of a dozen or so genes, which specify proteins necessary for changes,
which specify a protein that exposes certain binding proteins on the membrane surface,
and also a number of proteins which act internally to facilitate transfer of the DNA.
Now here we have a confident cell which is bound to a large DNA molecule and I show sequentially some of the steps in the uptake mechanism.
The first thing that occurs is that this binding protein interacts with the DNA and produces a nick in one strand.
This black protein then completes the break and in some cases DNA comes off, but some of the strands are then taken in to a space
which is still outside the cytoplasmic membrane, here is the cell wall, and the DNA is converted to a single strand.
And here we see the process further along.
In addition there is a protein which has been induced by this process
which binds to the DNA, to the single stranded DNA, in much the same way that a viral molecule would be packaged.
This has a structure which is highly protected against nucleases
and is stable enough to be isolated as a DNA protein complex in caesium chloride gradients and by several types of chromatography.
Then by mechanisms which are not understood, we have an incorporation of the single strand into the chromosome to form a transformant.
Now if I could move quickly to the gram-negative cells shown on the next slide.
These are bacteria that my own laboratory is studying and we became interested in the whole process of how haemophilus bacteria take up DNA
because of a discovery by a colleague John Scocca that haemophilus has the ability to recognize its own DNA during transformation.
If you, for example, take a mixture of several types on DNA, haemophilus DNA, E-Coli DNA, calf thymus DNA and so on,
the cells will selectively take up only the haemophilus molecules and incorporate them.
And this is shown by radioactive label experiments here.
Haemophilus DNA as opposed to the absence of uptake of a variety of foreign DNAs.
It was clear from these experiments that the cells, somehow, at the surface, could identify which molecule was which,
and we decided to determine what the recognition mechanism involved.
If I could have the next slide.
To do this we took a pure piece of haemophilus DNA that we obtained by molecular cloning, using the recombinant DNA techniques.
That piece of DNA was then cleaved into a dozen or so fragments using a restriction enzyme ALU1.
And the fragments were all labelled with radioactive phosphorus.
And here is the mixture.
That mixture then was incubated with competent cells and they were allowed to take up DNA.
We found that they took up only two fragments out of this entire mixture.
They recognized those two fragments but not the others.
So the natural question is, what is there about the sequence in those two fragments that is different from the others?
To obtain that answer, we continued to use restriction enzymes to break these fragments into smaller and smaller pieces,
again asking which pieces are taken up, so that finally we obtained as shown on the next slide, four different fragments,
which were small enough that we could easily determine their nucleotide sequence.
And when we then allowed a computer to scan these sequences, it told us that they all contained an eleven base pair sequence in common.
And they also told us that the probability of that event randomly occurring was about 10^-11.
Well, if I could have the next slide.
We continued then to study this presumed DNA uptake sequence which is shown here.
We found that the sequence is present on all fragments which can be bound and taken up by the bacteria,
whereas a number of sequenced molecules that are available that do not contain that eleven base pair site are not taken up.
We also found that if we modified certain bases in this sequence, that would affect the uptake,
implying that the cell was actually interacting with this site in a direct way.
Finally we were able to obtain, by collaboration with Saran Narang and Ottawa, a chemically synthesized eleven base pair sequence
which could be inserted into any foreign DNA molecule and then conferred on it the ability to be taken up,
thus completing the proof that this sequence must be present and is the necessary and sufficient condition for uptake
or at least for catalyzing uptake.
Could I have the next slide please?
Well, at this point we were very curious as to how the cell recognized the sequence in order to initiate uptake.
So we began to look at the proteins on the membrane.
And first let me point out that confidence induction itself involves the new synthesis of a number of proteins
which are implanted in the membrane, and this occurs in a synthetic media over a period of about ninety minutes.
If the cells are then returned to a rich medium, they lose the competence very quickly.
So it’s a true induction and then deinduction mechanism
which is called into play presumably only when the bacteria need to genetically recombine.
The next slide shows our attempt to purify the membrane receptor.
Here we have introduced S35 label during the induction step and then extracted the membranes
and stabilised the membrane proteins with detergent, put then over a DNA affinity column.
And we find that there is a small fraction of the proteins, about 3 or 4%, which are specifically bound to haemophilus DNA and can be eluded.
This fraction contains about six different polypeptide chains which we believe to be a part of the transport mechanism for DNA,
and one or two of those polypeptide chains probably are involved in the specific recognition of the eleven base pair sequence.
If I could have the next slide.
I show here evidence that the receptor is, in competent cells, is in the outer membrane,
because if you separate outer membrane fragments from inner membrane fragments by a suitable density gradient,
the specific binding activity is predominantly in the outer membrane fraction.
The next slide shows a summary of what we believe to be the initial steps in the transformation process.
One has then the outer membrane receptor present probably in only a few copies on the cell membrane
and donor DNA containing eleven base pair sites.
These interact in a reversible fashion which we have been able to demonstrate to form a complex
which then at some point becomes irreversible and transport proceeds into the cell.
Now here is the really the unknown and more difficult part of the problem.
How does this highly negatively charged molecule actually penetrate the membrane?
The receptor DNA interaction is only the trigger for this process and we are only now beginning to get some notion as to how this occurs.
I see that my time has run out.
The next slide shows a scanning electron micrograph of confident cells.
And if you remember the first slide I showed, the surface was smooth.
As the cells developed confidence, they formed little vesicular blebs on the surface, about a hundred nanometres in diameter,
which are sufficient in size to contain fairly large molecules of DNA.
If the cells are exposed to DNA, the vesicles disappear and in some cases can be visualized inside the cell.
If the confident cells are returned to a rich medium to promote deinduction, the vesicles are released into the medium
and we can harvest these vesicles.
This is work done by a former student, Dr. Robert Dyke.
The vesicles can be harvested and they, by themselves, will take up DNA in a specific fashion,
so that it becomes tightly bound and is resistant to nucleases.
In addition, if one looks at the protein content of the membranes of these vesicles,
it’s highly enriched for the six proteins which we had previously purified from whole membranes.
So we are building a circumstantial case for the involvement of these vesicles in the transport process.
On the last slide, I show a schematic version of how the transport might take place.
Here we have vesicular transport in which DNA could perhaps bind to the vesicle,
which would then evaginate and package the DNA much as a virus would package DNA.
And then transport it to the inner membrane, where it could be, by fusion, the DNA could be injected into the cell.
I would like to discuss this with Dr. Luria perhaps to see if it is a feasible mechanism.
On the other hand - so we believe that this is a likely mechanism for the haemophilus, or other gram-negative bacteria,
whereas for the gram-positive bacteria we think it’s more likely that there is...