DNA FINGERPRINTING AND PHYLOGENETIC ANALYSIS OF BACTERIA. DNA fingerprinting and the bacterial 16S-23S rrna intergene region.

Transcription

1 MCB4403L SUPPLEMENTAL EXERCISE #3: DNA FINGERPRINTING AND PHYLOGENETIC ANALYSIS OF BACTERIA INTRODUCTION DNA fingerprinting and the bacterial 16S-23S rrna intergene region. Relationships among bacteria have traditionally been examined using a variety of morphological (staining), biochemical and serological procedures and grouping together those bacteria that share the greatest number of traits. The resulting taxonomy, however, does not necessarily reflect phylogeny, relationships by evolutionary descent. Most microbiologists would prefer to have taxonomic schemes based on phylogeny since the grouped bacteria should share close genetic backgrounds and thus common phenotypes. Phylogenetic analysis can be based on the amino acid sequence of proteins and on the presence of similar metabolic pathways (two early methods), but the most accurate method of determining phylogenetic relationships is the comparison of DNA composition and sequence. In the last thirty years, since the discovery of restriction endonucleases, rapid DNA sequencing and polymerase chain reaction (PCR), the analysis of DNA sequence or DNA polymorphisms has become the standard for determining relationships among the bacteria. DNA analyses can also be applied to bacterial "identifications", i.e. do two bacterial strains (species) have an identical or similar DNA pattern or "fingerprint"? These methods are quick, accurate, and do not require many cells. Often no culturing is required. The number of DNA methods for identifying or classifying bacteria is staggering. The first and perhaps easiest type of DNA analysis applied to bacterial relationships was DNA base composition, hence the high GC and low GC Gram positive bacteria. But this method is slow, labor intensive and the results are not very informative, usually placing bacterial isolates only into large groups. With the discovery of restriction endonucleases, rapid methods were devised to identify and cluster bacteria by what has been called Restriction Endonuclease Analysis (REA) and Restriction Fragment-Length Polymorphism (RFLP) analysis. Genomic DNA from most bacteria when digested with "6-basepair cutters" like EcoRI, produce about 1000 DNA fragments of varying sizes that can be analyzed (separated) by agarose gel electrophoresis. The collection of restriction endonuclease-generated DNA fragments from a given species or strain of bacteria gives a distinctive pattern when analyzed by gel electrophoresis, and this pattern can be used to differentiate bacteria or strains of bacteria. This whole-genomic-pattern method is but one example of REA. RFLP analysis is a technique used mainly to detect genetic variation in a single gene. A polymorphism (difference in fragment size) of a specific restriction endonuclease fragment has to be "linked" to a specific genetic allele. The fragment size is monitored (as above) by gel electrophoresis, and the specific fragment is usually identified by Southern blotting and probing the DNA pattern with a labeled DNA fragment that hybridizes only to the one, specific fragment. The shortcomings of both REA and RFLP analysis is that they again are slow, labor intensive and usually require a large amount of DNA. Modern DNA fingerprinting usually relies on the polymerase chain reaction to amplify a known fragment of DNA that shows variation from one species or strain to another. The resulting DNA fragment or set of fragments is analyzed by a rapid separation technique (gel electrophoresis or use of an automated DNA sequencing apparatus), which results in a species or strain-specific pattern. Only small amounts of DNA are necessary (from just a few cells) because the DNA that is analyzed is amplified nearly one million fold. The laboratory manipulations are minimal and the total analysis time is about six hours - about four hours for the PCR reaction (New thermal cyclers are cutting this time down to less than 90 minutes.) and about two hours for the gel electrophoresis and recording of the pattern. In most cases of DNA fingerprinting, the investigator is trying to "match a pattern", i.e., the pattern from an unknown species, strain or individual is being matched to a known pattern. [One of the questions often

2 asked about human DNA fingerprinting is: "How many individuals in the population would, by chance, have a similar or identical pattern?"] Because of the use of PCR and the need for accurate matching, the gene or region of the DNA that is being analyzed has to have two characteristics. The PCR fragments that are being compared or matched between individuals obviously have to vary in size (position on a electrophoretic gel), and the greater amount of size variability the more useful is the particular DNA. In addition, the DNA being amplified has to have conserved sequences that flank the variable region. The conserved sequences are necessary for the hybridization of the two primers required for PCR amplification of the fragment. These types of DNA regions, a variable region flanked by two conserved regions, are used commonly in human DNA fingerprinting. VNTRs (variable number of tandem repeats) are common in human DNA, and they fit very nicely the two characteristics required for DNA fingerprinting. One hypothetical example of a VNTR is a sequence of 12 nucleotide base-pairs that is repeated from 3 to 20 times in different individuals at one locus in the genome, and on each side of this locus are conserved sequences of at least 20 bp found in all individuals. Primers are synthesized that hybridize to each of the conserved sequences, and they are used to amplify the VNTR locus by PCR. Since humans are diploid, the usual result for this "fingerprint" is two bands on an electrophoretic gel, one paternal band and one maternal band. In this hypothetical case, 18 different bands (differing by 12 bp) are possible (3 to 20 tandem repeats), thus, nearly 200 (171) different patterns are possible for one individual. [On occasion a single band may result because both parents have donated the same VNTR allele.] In human DNA fingerprinting at least 6 of these VNTR loci are used to identify an individual, and the possibility of a random match for all six loci is less than one in a billion. [If there were 6 VNTR loci like the one above the number would be 1 in 2.5 X 10 13, but this number is dependent not only on the number of tandem repeats at each locus, but also on the frequency of each of the repeats (alleles) in the population.] VNTRs are rare in bacterial genomes and even when they are present they are species specific and would therefore only be useful for identifying strains within the species. In this experiment you will use another type of locus that has the same two characteristics, a sizevariable region surrounded by conserved regions. The DNA region between the 16S and 23S ribosomal RNA genes in bacteria (the intergene region, IGR) is variable in length, and there are highly conserved sequences on each side (the 5'-end of the 16S rrna gene and the 3'-end of the 23S rrna gene; See the figure below.). Another feature of the 16S-23S rrna gene IGR is that there are usually several copies of this region in a single bacterium (e.g., There are six in E. coli.) and there is usually size variability even within the same organism. Thus for E. coli the IGR fingerprint from one strain gives multiple bands (up to six), and between strains there is variability in the overall pattern. We have found that between species of bacteria the number of bands in the fingerprint and the general pattern of the fingerprint is reproducible enough to group bacterial isolates into phylogenetically related groups, an easy and rapid way to investigate bacterial diversity in a community.

3 What The Heck is PCR? ~~~ (from the internet Polymerase chain reaction (PCR) is a technique thought up and developed by Kary Mullis in 1983 and now is used to amplify DNA say from a single gene in order to have enough DNA to study, test, or clone. This technique can be used to identify with a very high-probability, disease-causing viruses and/or bacteria, a deceased person, or a criminal suspect. In order to use PCR, one must already know the exact sequences which flank (lie on either side of) both ends of a given region of interest in DNA (may be a gene or any sequence). One need not know the DNA sequence in-between. The building-block sequences (nucleotide sequences) of many of the genes and flanking regions of genes of many different organisms are known. We also know that the DNA of different organisms is different (while some genes may be the same, or very similar among organisms, there will always be genes whose DNA sequences differ among different organisms otherwise, would be the same organism (e.g., same virus, same bacterium, an identical twin; therefore, by identifying the genes which are different, and therefore unique, one can use this information to identify an organism). A gene's building-block sequence is the precise order of appearance, one after the other, of 4 different components (deoxyribonucleotides) within a stretch of DNA (deoxyribonucleic acid). The 4 components are: Adenine, Thymidine, Cytosine and Guanine, abbreviated as: A, T, C and G, respectively (a 4-letter alphabet). The arrangement of the letters (one after the other) of this 4-letter alphabet generates a "sentence" (a gene sequence). The number of letters in the sentence may be relatively few, or relatively many, depending on the gene. If the sentence is 1000 letters-long, the sequence would be said to be 1 kilobase (1000 bases). As an example: ATATCGGGTTAACCCCGGTATGTACGCTA would represent part of one gene. DNA is double-stranded (except in some viruses), and the two strands pair with one another in a very precise way. EACH letter in a strand will pair with only one kind of letter across from it in the opposing strand: A ALWAYS pairs with T; and, C ALWAYS pairs with G across the two strands. So: 5 - TTAACGGGGCCCTTTAAA...TTTAAACCCGGGTTT-3 Would pair with: 3 - AATTGCCCCGGGAAATTT...AAATTTGGGCCCAAA-5 Now, let's say that the above sequences "flank" (are on either end of..) the gene, which includes a long stretch of letters designated as:... These are known, absolutely identified to be, the sequence of letters which ONLY flank a particular region of a particular organism's DNA, and NO OTHER ORGANISM'S DNA. This region would be a target sequence for PCR. The first step for PCR would be to synthesize "primers" of about 20 letters-long, using each of the 4 letters, and a Machine which can link the letters together in the order desired - this step is easily done, by adding one letter-at-a-time to the Machine (DNA synthesizer). In this example, the primers we wish to make will be exactly the same as the flanking sequences Shown above. We make ONE primer exactly like the lower left-hand sequence, and ONE primer exactly like the upper right-hand sequence, to generate: 3 -TTAACGGGGCCCTTTAAA...TTTAAACCCGGGTTT-5 5 -AATTGCCCCGGGAAATTT...> and: <...TTTAAACCCGGGTTT-5 5 -AATTGCCCCGGGAAATTT...AAATTTGGGCCCAAA-3 Now. the... may be a very long set of letters in-between; doesn't matter. If you look at this arrangement, you can see that if the lower left-hand primer sequence (italics) paired to the upper strand could be extended to the right in the direction of the arrow, and the upper right-hand sequence paired to the lower strand could be extended to the left in the direction of the arrow(remembering that the... also represent letters, and opposite pairing will ALWAYS be A to T and C to G), one could successfully exactly duplicate the original gene's entire sequence. Now there would be four strands, where originally there were only two. If one leaves everything in there, and repeats the procedure, now there will be eight strands, do again - now 16, etc.. therefore, about 20 cycles will theoretically produce approximately one-million copies of the original sequences (2 raised to the 20th power). Thus, with this amplification potential, there is enough DNA in one-tenth of one-millionth of a liter (0.1 microliter) of human saliva (contains a small number of shed epithelial cells), to use the PCR system to identify a genetic sequence as having come from a human being! Consequently, only a very tiny amount of an organism's DNA need be available originally. Enough DNA is present in an insect trapped within 80 million year-old amber (fossilized pine resin) to amplify by this technique! Scientists have used primers which represent present-day insect's DNA, to do these amplifications. Here is how PCR is performed: First step: unknown DNA is heated, which causes the paired strands to separate (single strands now accessible to primers).

4 Second step: add large excess of primers relative to the amount of DNA being amplified, and cool the reaction mixture to allow double-strands to form again (because of the large excess of primers, the two strands will always bind to the primers, instead ofwith each other). Third step: to a mixture of all 4 individual letters (deoxyribonucleotides), add an enzyme which can "read" the opposing strand's "sentence" and extend the primer's "sentence" by "hooking" letters together in the order in which they pair across from one another - A:T and C:G. This particular enzyme is called a DNA polymerase (because it makes DNA polymers). One such enzyme used in PCR is called Taq polymerase (originally isolated from a bacterium that can live in hot springs - therefore, can withstand the high temperature necessary for DNA-strand separation, and can be left in the reaction). Now, we have the enzyme synthesizing new DNA in opposite directions - BUT ONLY THIS PARTICULAR REGION OF DNA. After one cycle, add more primers, add 4-letter mixture, and repeat the cycle. The primers will bind to the "old" sequences as well as to the newly-synthesized sequences. The enzyme will again extend primer sentences... Finally, there will be PLENTYof DNA - and ALL OF IT will be copies of just this particular region. Therefore, by using different primers which represent flanking regions of different genes of various organisms in SEPARATE experiments, one can determine if in fact, any DNA has been amplified. If it has not, then the primers did not bind to the DNA of the sample, and it is therefore highly unlikely that the DNA of an organism which a given set of primers represents, is present. On the other hand, appearance of DNA by PCR will allow precise identification of the source of the amplified material. In our exercise, we will be amplifying the intergene region of the ribosomal RNA clusters in bacteria. Each species of bacteria has a specific number of ribosomal RNA genes. E. coli has 7, B. subtilis has 10, S. aureus 6, etc. Ribosomal RNA has been highly conserved in bacteria and other organisms as well. One might imagine that any mutations to the ribosomes would be potentially lethal as protein synthesis depends on them. Any mutation might affect the efficiency of protein synthesis or perhaps in the extreme stop it completely. We can say that the ribosomal RNA is highly conserved and that its sequence has been used in recent years to relate species of bacteria, plants, animals, etc. to each other and that the more alike the sequence the more closely related are the organisms. Conversely, the intergene region (IGR) that is the spacer between the 16s and 23s RNAs is not highly conserved with the exception of the trnas encoded in this region. In fact the IGRs within the same bacterium usually are different. That is the 7 IGRs of E. coli may be all different. We are going to use these differences to get a banding pattern of the different bacterial species and thus enable us to identify an unknown bacterium. We can only do this because we know what species we are looking at. We could not do this identification with a true unknown say from the environment. That would require sequencing the 16s rdna. We could use this technique and have done so to determine if isolates from a hospital outbreak of methcillin resistant Staphylococcus aureus are the same and from a common source or are different and not from a common source. The sequence of E.coli 3 end of the 16s, IGR, and 5 end of the 23s is as follows: 1 CCCGGGCCTT GTACACACCG CCCGTCACAC CATGGGAGTG GGTTGCAAAA GAAGTAGGTA 61 GCTTAACCTT CGGGAGGGCG CTTACCACTT TGTGATTCAT GACTGGGGTG AAGTCGTAAC 121 AAGGTAACCG TAGGGGAACC TGCGGTTGGA TCACCTCCTT ACCTTAAAGA AGCGTACTTT 181 GCAGTGCTCA CACAGATTGT CTGATGAAAA TGAGCAGTAA AACCTCTACA GGCTTGTAGC 241 TCAGGTGGTT AGAGCGCACC CCTGATAAGG GTGAGGTCGG TGGTTCAAGT CCACTCAGGC 301 CTACCAAATT TGCACGGCAA ATTTGAAGAG GTTTTAACTA CATGTTATGG GGCTATAGCT 361 CAGCTGGGAG AGCGCCTGCT TTGCACGCAG GAGGTCTGCG GTTCGATCCC GCATAGCTCC 421 ACCATCTCTG TAGTGGTTAA ATAAAAAATA CTTCAGAGTG TACCTGCAAA GGTTCACTGC 481 GAAGTTTTGC TCTTTAAAAA TCTGGATCAA GCTGAAAATT GAAACACTGA ACAACGAAAG 541 TTGTTCGTGA GTCTCTCAAA TTTTCGCAAC TCTGAAGTGA AACATCTTCG GGTTGTGAGG 601 TTAAGCGACT AAGCGTACAC GGTGGATGCC CTGGCAGTCA GAGGCGATGA AGGACGTGCT 661 AATCTGCGAT AAGCGTCGGT AAGGTGATAT GAACCGTTAT AAC The highlighted sequences are the two primers that we will be using to amplify the IGR. The 16s primer is called C-complement because it primes in the direction of the sequence as written 5 to 3 off the complementary strand. The 23s primer is called 8 and is the complement to the sequence written here ( AGG GCA TCC ACC GTG ). It will prime off this strand into the IGR.

5 SEQUENCING THE 16S-23S INTERGENE REGION IGR 16S RNA ile (ala) 23S RNA primer CC primer TC primer 8 primer 7 PCR AMPLIFICATION primer CC primer 7 SEQUENCING PROTOCOL 150 N ~450 N 250 N CC TC ile C ala C 8 7 The cycle sequence we will use in the PCR reaction is as follows: Denaturation: 95 o C Annealing: 50 o C Extension: 72 o C These are repeated for 30 cycles and takes about 3 hours to complete. The enzyme we are using is called Taq Polymerase and has an optimum of 72 o C. It was isolated from Thermophilus aquaticus a bacterium living in hot springs thus the high optimum temperature. If the enzyme were not stable at high temperatures, we would have to add fresh enzyme at each cycle. So you could say this enzyme makes the thermocycler and PCR possible as we know it. Here is a graphic presentation of the amplification:

6 PROCEDURE First Laboratory Period Students will use the cultures from the TSA plates which they streaked from the EMB water samples. Students will work alone but share equipment with a partner. 1. Pick a colony with a sterile stick and mix it into 50ul of 10mM EDTA. Small inocula often work better than too much. Mix this well until all the bacteria are suspended evenly. Use the vortex provided. Spin briefly in the microfuge. 2. Boil this mixture of EDTA and bacteria for 5 minutes. There will be a setup for your use. 3. Allow the EDTA lysate to cool. The EDTA lysate is the source of the bacterial DNA. 4. Obtain an eppendorf tube that contains all the PCR reaction ingredients except for DNA. You will need one for each culture. 5. Add 1 ul of the DNA lysate to this reaction mix. Make sure you have labelled the tube on the top of the lid Label the side of the tube as well. Label it with your section, sample # and your initials. 6. Mix well. 7. Spin in a microfuge briefly to get the contents of the tube to the bottom. With such a small volume 20ul this is necessary. 8. Place your labelled tubes on ice with the rest of the class. Your TA will put these tubes into the thermocycler where the PCR reaction will run through 30 cycles of denaturation, annealing, and extension. Second Laboratory Period 1. Set up the gel apparatus the way your TA instructs you. Be sure to follow directions carefully. Pour your gel. 2. Allow the gel to solidify. Remove the scotch tape. Pour enough buffer to cover the gel (about 300ml). Carefully remove the comb by moving it side to side gently and pulling up. You should have 12 wells. 3. To each of your PCR reaction tubes add 4 ul of gel loading buffer. This buffer contains 2 dyes, bromophenol blue and xylene cyanol to track the running of the bands on the gel as reference points and ficoll to make the samples heavy so they will sink into the wells (formed by the comb) and not diffuse into the buffer. 4. Mix well and spin in the microfuge. 5. Pipet out 20ul of your sample and load it into a well in the gel. Make a note of which sample goes into which well. This is very important or else you will not know which well has which sample. 6. Load additionally the 100 bp DNA ladder so that when you view your bands you will know what size they are approximately. You can use this standard as a reference point also. Put it in a convenient lane. 7. Run the gel for hours at 100volts or until the bromphenol blue dye is a centimeter from the end. Make sure the black electrode (-) is at the top of the gel. The electrons and negatively charged ions run from the negative pole to the positive and so will DNA ( it has an overall negative charge due to the PO 4-2 on the ends). Check to see if bubbles are forming at the (-) pole. This means all the connections are okay. One person from each group should stay and complete the rest of the exercise together with the TA. 8. Turn off the power supply and unplug the electrodes. Remove the top of the apparatus and remove the gel by sliding it off the holder into a container for staining. Cut off a corner of the gel so you can identify it later among the other gels. Add the ethidium bromide (wear gloves). Ethidium bromide is a dye that intercalates with the DNA causing it to fluoresce under ultraviolet light. It can also intercalate with your DNA causing damage if exposed to UV light too. It is a mutagen so be sure to wear gloves and don t splash it around. 9. Let the gel shake gently in the EtBr for 30 minutes. Pour off the EtBr into the special EtBr container. Rinse your gel 3 times with tap water. 10. View the gel on the Gel Doc. This is a special apparatus that has an UV transilluminator and camera link to a computer. Take a picture with it and save it on the computer. 11. Your TA will post a copy of the picture of the gel on the web site. 12. Can you identify your unknown bacterial culture? Third Laboratory Period: Discussion of Results PCR REACTION MIX:.2ng/ml Primers, 4mM MgCl 2, 50uM dntps, Taq 25u/ml, 10mM TrisHCl ph 8.3, 50mM KCl. Electrophoresis buffer: 90mM Tris base, 90mM boric acid, and 2mM EDTA; ph