Undergraduate Research at UW-Superior

Dr. Ralph Seelke:  Exploring the Evolutionary Potential of Escherichia coli

 My research is focused on a simple and very basic question:  What can evolution really do?

While historical, anatomic, and DNA evidence has made a strong case for evolution, the experimental support for evolution – those aspects that can be demonstrated in a laboratory setting - has been weak.  Moreover, evolution has often been compared to economic theory- excellent at explaining past events, but offering little in the way of prediction.  The purpose of my research is to put evolutionary theory on a firmer experimental footing.  To do this, I have taken advantage of the most well-studied organism on earth - the colon bacterium Escherichia coli .  My objective is to follow the evolution of E. coli for thousands of generations, asking very specific questions about its ability to evolve new functions.  Our model strain for this has been E. coli AB1157, a strain with over ten defects in sugar utilization or amino acid synthesis genes.  My students and I have put AB1157 under a variety of selective conditions, asking if evolution is able to correct these defects.  Our experimental approach is simple: we allow AB1157 to grow under conditions in which mutants that have gained a selective advantage  rapidly dominate a population. We grow our bacteria each day, and then transfer 1% of the population to new growth medium the following day.  In this manner, we can observe 6.64 generations of evolution each day, over 46 generations per week, almost 200 generations per month, and over 2,400 generations in a year. In 10 years, using this method, we could follow the evolution of a culture for over 20,000 generations, the human equivalent of over 600,000 years of evolution. 

We have been using this approach since June of 2001.  However, our early results were so interesting that we have not followed a microbe’s evolution for more than a few hundred generations yet.  We find that some genes evolve rapidly, almost overnight; others take weeks to evolve; and one has yet to show any indication of evolution.  Our objective is to follow the evolution of several genes for at least a thousand generations, and then use DNA sequencing to determine what is an easy task for evolution, what is a harder task, and what is a task that cannot be done in the time allotted.

Our research has been supported by a $60,000 grant from the Merck Foundation, through their Undergraduate Science Research Program.  Eight students have participated in this research, producing four student presentations and one accepted submission to UW-Superior’s pending Undergraduate Scholarly Activity Journal. Currently, my research team consists of four students.  In addition to the valuable experience they have gained, three of them will also have the opportunity to conduct research at one of the Merck labs this summer.  The grant has also produced the preliminary data for the grant proposal that Dr. Jim Lane, Dr. Sanjay Shukla of the Marshfield Medical Clinic, and I submitted to the National Science Foundation earlier this year.  If funded, this grant proposal will allow even more students to be involved in  undergraduate research.


Summer Undergraduate Research at UW-Superior


In February 2001, the departments of Biology/Earth Sciences and Chemistry at UW-Superior received a three-year, $60,000 grant from the Merck Foundation to support undergraduate research at UW-Superior.  This grant provides money to support at least five student researchers in the laboratories of either Dr. Ralph Seelke, Department of Biology/Earth Science, or Dr. Jim Lane, Department of Chemistry.  Currently, Dr. Lane has filled his positions.  Dr. Seelke is seeking at least three students interested in pursuing undergraduate research at UW-Superior on a full-time basis for the summer of 2002.

Dr. Seelke’s lab is asking basic questions about the evolutionary potential of the bacterium Escherichia coli .  The research will last for a minimum of eight weeks and a maximum of ten weeks, from late May until mid-August.  A stipend of $300 per week will be provided.  Participants will be expected to register for one credit of undergraduate research as well.  Because the research will be full-time, it is expected that participants will not have other employment during this time.  The participants will have some flexibility as to the period of time that they work, but should be prepared to begin no later than June 10.  As part of the research experience, the participants will also give a presentation at the regional American Society for Microbiology meeting in October, 2002.  In addition to their research, the participants will have the opportunity to be involved in a variety of scientific and recreational activities in the Twin Ports area.



UW-Superior Summer Undergraduate Science Research Program

Application for Summer Research Assistant Position

The Departments of Biology and Earth Sciences, and the Department of Chemistry at UW-Superior are the recipients of a grant from the Merck Foundation to support undergraduate research.  We invite motivated and qualified students to apply for a paid Undergraduate Research Assistant position at UW-Superior for the summer.  At this point, the Merck Foundation can only support research in Dr. Ralph Seelke’s lab.  For a description of his research, see http://www2.uwsuper.edu/rseelke/research_instructions.htm


Name______________________________ Address_____________________________

Phone number/email address:_________________________________

Major :_________________________________

Number of credits you will have completed by May 2002 and GPA: _______________________

Number of Science/Math credits you will have completed by May 2002 and  GPA (Biology, Chemistry, Geology, Mathematics): _______________________

Please list activities or on-campus work that you have been involved in at UW-Superior, including the dates of your involvement, and type of involvement (e.g., vice-president of a club, etc.)






Please list up to five honors or awards that you received, either in high school, or while at UW-Superior.






Please list the names, addresses, phone numbers, and email addresses (if available) of three references who can comment on your academic ability, work habits, and interest in science:




On the space below (or on a separate page), please describe any experience you have had doing research, and explain why you would like to do research this summer, including how this research might fit into your long-term career goals.   

For consideration, please return this form by April 15, 2002 to:  Dr. Ralph Seelke, Department of Biology and Earth Science, UW-Superior, Superior, WI   54880.  Please include a copy of your transcript as well (unofficial copies are acceptable).

I hereby certify that the information contained in this application is accurate to the best of my knowledge.


Signed_________________________________  Date__________________________




Note- this section deals with some specifics for my research.


Instruction for the Experimental Evolution Researchers: I'm just trying this out, so we'll see how it works!

We now have five derivatives of AB1157.  Four have single sugar changes:  a Gal+, Lac+, Mannitol+, and Xylose +.  One is Ara+, but also Xylose+ .  The Gal+,  Manitol+, and Xylose+  were all spontaneous revertants of AB1157.  The Ara+ came from the mating I did, as did the Lac+.  These have been named AB1157 G, AB1157L, AB1157AX, AB1157M, AB1157X

We can now use these to study evolution. The first thing question we want to ask:  What is the advantage of these traits in media that allows both a sugar+ and sugar- to grow?

We'll answer this by doing a competition experiment- Here's an example using the Lac+ competition.  The idea is to determine the # of each type before and after growth, using the fermentation differences as markers.

NOTE:  When it says "plate at 10-6 dilution", this means a FINAL dilution of 10-6.  You dilute to 10-5, and plate 0.1 ml, which makes a final dilution of 10-6.

1.  Grow up an ON (overnight) of AB1157 and a sugar+ strain in nutrient broth.  There will be about 109 bacteria/ml in each tube.

2.  Add 10 ul each to a 1.0 ml of diluent- nutrient broth, or 0.9% saline, or buffered DI H2O.  This is approximately a 10-2 dilution of each.

3.  Add 20 ul of the mix to 2 mls of nutrient broth + 1% lactose (1.9 mls NB + 0.1 ml 20% lactose).     

This is a total of 10-4 dilution. 

Plate 0.1 ml of a 10-1 dilution (= a 10-2 dilution) and 10-2dilution (= a 10-3dilution) of this mix onto Mac-lac or TTC lac.  This should give 100- 1000 colonies per plate.  Prepare two plates of the 10-2 and three of the 10-3dilution.

4.  Grow the remaining mix ON @ 37 C, with shaking.

5.  When grown, plate two plates @  10-6 dilution (final) and three @ 10-7dilution (final).Plate onto Mac-lac or TTC-lac + strep plates.  Count Lac+ and Lac- colonies

Ideally, the Lac+ will be more fit (produce more offspring), than the original AB1157.  We can then calculate a fitness index:

From Lenski's paper:  Let the initial densities of theAra + and  Ara- strains be N1(0) and N2(0), respectively, and let N1(1) and N2(2) be their corresponding densities after one day.  The average rate of increase (or realized Malthusian parameter), mi, for either competitor is then calculated as :

mi= ln[Ni(1)/Ni(0)/(1 day). NOTE: this is almost the same as the # of generations per day.

The fitness of one strain relative to another, Wij, is estimated as the ratio of their Malthusian parameters (Lenski et al., 1991):

Wij= mi/mj.

7/26/02 What we have done, and what we still need to do for the evolution of AB1157 story.

What we've shown: using serial transfer, under conditions of selection, we can observe evolution of AB1157 as it evolves towards prototrophy. It is defective in genes for five amino acids: Threonine, Leucine, Proline, Histidine, and Arginine.

Three of these defects were curable by evolution with one week of transfers:  Arginine, leucine, and threonine.- that is, within 42 generations of evolution.  2 ug/ml was selective for these traits.

Histidine and Proline did not evolve initially, under these conditions.  For both of these AA's, 2 ug/ml was not sufficiently selective- cultures gave similar cfu/ml as with 20 ug/ml.  For histidine, the amt needed to be reduced to 0.3 ug/ml, and for proline, to 0.2 ug/ml to achieve selective conditions.

Under these conditions, evolution to histidine prototrophy was observed within 100 generations.  With proline, only one tube showed evolution after--- days of transfers.

This system also shows the sequential evolution of traits.  Evolution of arginine and threonine prototrophy was observed after 9-15 days, when conditions were selective for both.  It appeared that evolution was sequential; the population evolved the ability to produce thr, then arg. For arginine and leucine, the population evolved the ability to produce arginine, and then evolved the ability to produce leucine.  The number of generations required for this evolution to occur varied from 50 to 250.  In one set of experiments, a slow-growing leu+ revertant appeared.

Something we saw, but may not be repeatable:  Pravien found that, when his Arg+ evolvants were tested, many were also thr+ as well, and vice versa.

These are things we want to be able to say, but need further experimental evidence.

The system also shows the increased fitness of a population in a selective environment, even when full evolution does not occur.  Selection under low proline concentrations resulted in a -------% increase in fitness, compared with the ancestral strain.

The time required for evolution roughly corresponded to the frequency of reversion seen in platings.  For this, we need to be able to show the reversion of AB1157 on a plate-

I just looked through my old notes- reversion on a plate, w/o added AA's at low levels, w/ cells grown in RB, gave few or no colonies.  A tube full of AB1157 grown in MD+ 5 AA's may be quite different in # of revertants than one grown in RB. In looking at my old notes, I had 780/ml Arg+ in an overnight of AB1157 grown in MD+5AA's + glu- 78 on a 10^-1 plate!

With platings from AB1157 grown in RB, the results were MUCH worse- usually 0.


Evolution of Xylitol utilization.

What we've done: we isolated a xyl+ evolvant of LAS 100; we've evolved it in 0.2% xylitol for over 130 transfers, over 700 generations.

What this study should look like:  Our goal is to follow the evolution in terms of

-fitness of the evolved strains- here we'll use the lac--tetR, mannitol-/tetR derivatives to show the fitness of these strains after evolution.

We'll compare fitness in RB, in MD +glu, and MD + ribitol.

increased fitness in xylitol, from start to finish.


Enzyme studies:  here we plot the amount of enzyme (RDH) produced, as measured by specific activity of crude extracts, and also the activity of the enzyme, as measured by Km values.

We also need to compare the activity of these enzymes in xylitol, vs their activity in ribitol.  Here, we may need to get concentrated amounts of the enzyme, to compare the RDH from las 100 w/ the RDH from X-3 after hundreds of generations of evolution.

Also, we need to measure the inducibility of the enzyme as well- compare activity in MD+glu, vs RB, vs MD+ Ribitol.


Things to work on for the next study

Right now, I'm looking at the evolution towards prototrophy of tryptophan biosynthesis.  The tryptophan synthase gene is well studied- it has an alpha and beta subunits.  There are LOTS of mutations in the A gene, for the alpha subunit.

I think we could do a very sophisticated evolution experiment with the trpA gene.  There are numerous mutant alleles of this gene, most with single AA substitutions.  several key AA residues have been identified, such that substitutions at those  locations cause the trpA gene to become non-functional. 

Right now, what I'm looking for is a list of the mutations, and the amino acids affected.  I think I have the literature needed for this, although it goes back to the 60's.

My plan now is to use in vitro mutagenesis to create a trpA gene with multiple base substitutions, any of which would render the gene non-functional. We would then ask if we can obtain revertants with one, two, three, or four defects.

What I'm missing is the information on the AA's and their substitutions that can result in a mutant trpA gene.

Research Plans- October 21, 2002

Where we've been and where we need to go.

We're close to finishing our work with AB1157; we have documented its ability to evolve.  What we need, to put the final touches on a paper, would be data on:

1) mutation rates of these genes;

2) studies using larger cultures; both of these will build on what we already know.


Mutation rates:   I don't want to spend a lot of time on these studies, but they would be useful.

Mutation rates are calculated from the number of mutations that occur, as a microbe grows from some number no to n.

the mutation rate, a, is equal to the average number of mutations occurring (m), divided by the increase in cell numbers while the culture is growing; there's a ln2 factor that is included: 

a= 0.69(m)/n-no

The mutation rate can also be calculated by the number of populations in which there are NO mutations. In this case,

m= -ln(Po), where P is the fraction of cultures with zero mutants (and hence 0 mutations).  m is the average # of mutations per culture.

To find the average mutation rate:  Take AB1157 grown in RB + strep;

Materials:  16 Rich agar plates, supplemented with strep

overnight culture of AB1157

30 plastic culture/centrifuge tubes.

30 minimal Davis agar plates, with 4 of five amino acids + 0.2% glucose, + strep.

Buffered DI H20, pipettes, dilution tubes, etc.

All material must be sterile before starting.

Dilute the culture to 10^-6.  Plate 100ul of this dilution onto three plates, and 100 ul of the 10^-5 dilution onto one plate.  These are 10^-7 and 10^-6 dilutions. (see worksheet on dilutions).  Use Plate count agar, or other rich medium, supplemented with 50ug/ml streptomycin.

Take 0.6 ml of the 10^-5 dilution, and add to 60 ml of RB.  This will add 1-300 cells/ml to the broth.

Dispense into 30 plastic culture tubes.

Grow overnight, with shaking, @ 37C.

For 3 tubes, dilute and plate (in triplicate) to obtain the # of cells/ml, as described above. Only remove 10ul for this plating. Plate on PCA or other rich plate + strep. Count these colonies the next day, and use this value to determine the average number of cells/ml in each tube.

Centrifuge the tubes, 5,000 RPM, for 3 minutes; remove the supernatant.

Resuspend in 100ul of buffered DI H20.

Plate the entire contents onto a minimal Davis plate with four of the five amino acids, but missing the one for which you are determining the mutation rate. As a control, also plate 100 ul of your diluent.

Incubate for 4 days @ 37C.

Count the number of colonies per plate.

Determine the number of plates lacking ANY colonies.

Calculations:  let's say that, out of 30 tubes, 10 have no mutants.  Then Po is 0.33, and m= -ln (.33)= average of 1.1 mutations per culture.

Let's say you have an 200 bacteria on your 10^-7 dilution plates, for the number of bacteria/ml after growth.  You thus have 2X 10^9/ml, or 4X 10^9/tube (2 ml /tube). The starting # can effectively be ignored, since it's only  1-300.

thus, a= (0.69) (1.1)/4X10^9; = 1.9X 10^-10 mutations/cell generation.

Find up to five plates on which there are no colonies.

Remove the filter from those five plates, and place the filter in a large test tube with 10 mls of buffered DI water.

Dilute and plate to determine the number of bacteria/ml.