This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Patrick Yuh Spring 2007 Bio 105 Final Study Guide Terms: linkage map (genetic map) physical map polytene chromosome mitotic recombination chromosome rearrangement prototroph auxotroph complementation intergenic recombination intragenic recombination fine structure analysis transcription translation sense strand antisense strand homology codon intron exon untranslated regions (UTRs) transfer RNA (tRNA) restriction enzyme sticky ends restriction map gel electrophoresis plasmid cloning transformation genomic library cDNA library reverse transcriptase RFLP SNP telocentric acrocentric metacentric aneuploid monosomy trisomy Cvalue paradox nucleosome heterochromatin (2 kinds) operon promoter operator repressor induction negative gene control allostery merozygote / merodiploid Concepts: deletions (deficiencies) duplications tandem duplications unequal crossing over (mispairing) inversions paracentric pericentric inversion loop translocations alternate segregation adjacent segregation one geneone enzyme hypothesis the molecular basis of dominant/recessive genetic screen biochemical pathway complementation intragenic recombination deletion mapping complementation groups the central dogma transcription (how RNA polymerase works) initiation, elongation, termination RNA processing (5' cap, splicing, polyA tail) translation (how the ribosome works) properties of the genetic code wobble pairing missense mutation nonsense mutation nonsense suppressor mutation the RNA world hypothesis restriction digests DNA fingerprinting eukaryotic chromosome packaging regulation of gene expression cis and trans acting elements 1 Patrick Yuh Spring 2007 Examples to understand: retinoblastoma lactose intolerance Duchenne muscular dystrophy Philadelphia chromosome Beadle and Tatum's Neurospora experiments lac operon hemoglobin characteristics [Note after I finished all 8 pages: this ended up being more about concepts than problem solving. There are simply too many kinds of problems to talk about here, but I did try to include both some advice on how to approach them and some questions to think about. Practice is the only way to get good at this!] In my mind, there were about 7 major topics covered since the midterm. My advice: spend at least some time on each topic, since most of them will appear on the final in one form or another. My opinion on most to least important topics: V, IV, III, VII, II, I, VI. Prioritize your study time accordingly. I. Mitotic Recombination If you understand meiotic recombination from the first part of the course this shouldn't be too difficult. Remember, mitosis resembles meiosis II in terms of how the chromosomes line up on the metaphase plate. The difference in mitosis is that the homologues aren't supposed to pair up like they do in meiosis I...because they aren't supposed to recombine...but sometimes this does happen. Draw out the mitotic homologues, draw in a crossover event that causes two alleles to switch places, and put them on the metaphase plate in an orientation which would lead to genotypically different daughter cells. II. Chromosome Rearrangements Many of you have trouble with these. Take another crack at it, but study pragmatically don't spend all your time on this at the expense of understanding the other topics. In order to get chromosome rearrangements, it may help to think of DNA as a physical object (since it is a physical object) like a piece of string. In all rearrangements except duplications, you have to make two cuts in a string. One other thing you can be homozygous or heterozygous for a rearrangement. Deletions (aka Deficiencies) are just that a part of the chromosome (or string) that got deleted. You had to make two cuts to remove this part. Think about what the effects would be if there was a homozygous deletion which contained important genes. Duplications are just that, a region of the chromosome that got duplicated. How do they happen? One way is an error in homologous recombination (unequal crossing over). Imagine the homologues pairing up in mI and they're just a tiny bit misaligned. One sister chromatid ends up with both copies of a gene, and the other sister has none. If the duplication is right next to the original, this is called a tandem duplication. There can be a `chain reaction' of tandem duplications, where each time you go through meiosis you have unequal crossing over. This leads to massive gene duplications, which is actually believed to play a major role in evolution. The hemoglobin genes are one example of this. 2 Patrick Yuh Spring 2007 Inversions involve making two cuts in one chromosome, flipping this piece 180, and putting it back. If you are heterozygous for an inversion, this makes meiotic recombination an interesting affair. One of the two pairs of homologues must make an inversion loop so the inverted genes line up with the non inverted ones. Recombination frequency is greatly reduced in inversion heterozygotes, because most of the gametes you get from this recombination event are not viable. I didn't explain this correctly to some people. Draw out the inversion loop and see that the gametes are missing some genes and have two copies of others. In other words, they're inviable. There are two kinds of inversions paracentric and pericentric. Paracentric inversions do not include the centromere (para means next to or alongside, so you could think of the inversion as sitting next to the centromere much like a kid sitting next to a sandbox). Pericentric inversions do include the centromere (peri means around, so the inversion `wraps' around the centromere). If you have recombination in a paracentric inversion heterozygote, you end up with one pair of homologues with two centromeres each (dicentric) and the other pair with no centromeres at all (acentric). This is really bad. The dicentric will get microtubules attaching to both centromeres, maybe pulling in opposite directions, tearing the chromosome apart. The acentric will simply be lost. Translocations are recombination events between nonhomologues, say chr 1 and chr 12. If this happens once in a diploid organism, you now have a chr 1 and 12 that recombined, and the other chr 1 and 12 which are still okay. All four of these pair up in meiotic recombination by way of a cruciform structure (it looks like a 4way intersection). Draw the okay chromosomes diagonal to each other, and the translocated ones should fit in so you get homology everywhere. The TAs (myself included) made a slight mistake explaining the segregation patterns, so allow me to correct it here. There are 3 possible ways to segregate the chromosomes in the cruciform: leftright, topbottom, and diagonally. Diagonal is called alternate segregation, and produces viable gametes. The other two are called adjacent segregation, and produces inviable gametes. The mistake was in saying that these 3 ways are all equally likely, which isn't quite right. Alternate happens 50% of the time (so 50% viable gametes; adjacent (both of them put together) happen the other 50% (so 50% inviable gametes). This is why fertility is reduced by 50% in the next generation. III. Biochemical Pathways (one geneone enzyme) This topic encompasses 3 other topics: complementation, intragenic recombination, and deletion mapping. Complementation is the idea that two strains with mutations in different genes can `complement' each other when crossed, producing phenotypically WT progeny. The mutations have to be in different genes for this to work. For example, aaBB x AAbb AaBb. The progeny are heterozygous for both mutations, which allows them to survive. This can be useful in determining whether two unknown mutations are in the same gene. If they fail to complement, that means the mutations are in the same gene. 3 Patrick Yuh Spring 2007 Intragenic recombination is simply recombination within a gene. This happens very rarely, but when it does a wonderful thing can happen: you can start with two parents with mutations in the same gene, and end up with a few progeny with no mutations in that gene! If the recombination occurs between the actual mutations (let's assume they're point mutations), one gamete will end up with both mutations, but the other one will avoid both mutations and end up WT. Pretty nifty Deletion mapping is a useful tool for finding out where a mutation is located. You have one strain that has a mapped deletion (i.e. you know where the deletion is), and a mutation whose location you want to find out. If you cross these two and find that they complement, you know the mutation lies outside the deletion. If they do not complement, you know the mutation lies somewhere in the deletion. These concepts can help you construct a biochemical pathway. You can put it together going forwards or backwards, it doesn't matter. Just remember that the fewer compounds that can rescue a mutant, the later that mutation occurs in the pathway. Conversely, the more compounds that rescue, the earlier that mutation occurs. One more thing pathways can be linear or branched. Linear pathways tend to have more straightforward data to interpret. IV. The Central Dogma (DNA RNA protein) The central dogma of molecular biology explains the flow of information in a cell. DNA is read by RNA polymerase, and RNA is read by the ribosome. These processes are called transcription and translation, respectively. You should be familiar with both of them. 5' GCTACCATGGGCTTACCG...TTCACGATTACATGACGCATCTT 3' 3' CGATGGTACCCGAATGGC...AAGTGCTAATGTACTGCGTAGAA 5' Here we have a piece of DNA. RNA polymerase binds to an upstream sequence called the promoter. Once bound it starts reading down the DNA, and starts transcribing when it sees an ATG. RNA polymerase reads the 3' 5' strand (the template), and makes the complementary RNA in a 5' 3' direction. So it is essentially making a molecule that looks very similar to the 5' 3' strand, with U's instead of T's. This means either DNA strand can serve as the template. In yellow are two ATG start sites, one on each strand. RNA pol can start transcribing from either one. Make sure you understand the direction RNA pol travels as it transcribes. Transcription terminates when the RNA pol gets to one of three stop sites TAA, TAG, or TGA. In blue above are the stop sites for each strand. And there you have it. This initial RNA strand is called the primary transcript, or premRNA. In eukaryotes it gets processed before being read by the ribosome. Specifcally, three things happen: the 5' end gets capped (by 7 methylguanosine); the 3' end is polyadenylated (polyA tail); and introns are spliced out. So introns and exons exons are the regions that actually get read by the ribosome; introns are removed during RNA processing. You don't need to worry about what determines the boundary between an intron and an 4 Patrick Yuh Spring 2007 exon, but suffice it to say that the splicesome (a big complex like the ribosome that splices out the introns) recognizes certain RNA sequences and knows where to cut. Once the mature mRNA is ready, it gets exported out the nucleus into the cytoplasm where the ribosomes live. They read the RNA and translate the nucleotides into amino acids. Every 3 nts make up a codon, and every codon specifies for a certain amino acid. So the AUG in the RNA is the start codon, which codes for methionine. As the ribosome reads down the RNA, new amino acids come in and link up to the previous one. This growing polypeptide chain gets cut off from the ribosome when it reads a stop codon (UAA, UAG, or UGA), folds up, and becomes a protein. Properties of the genetic code. I won't rehash them here, but they're important to understand. Molecular nature of alleles. One thing I will mention is that we're now able to clearly define the dominant and recessive alleles we worked with before. That is, the nature of many recessive alleles is that they are mutated with respect to the WT allele. Since every 3 nt = 1 codon = 1 amino acid, if you have a deletion or insertion of 1 bp in your DNA that will totally mess up what the codons are, which will then completely change the amino acid sequence. These are also called frameshift mutations because they shift the reading frame of the RNA. Another kind of mutation is called a point mutation, which changes one base pair to another (e.g. AT GC). These mutations can vary in severity depending on the net effect on the protein. A nonsense mutation introduces a new stop codon, which results in premature termination of protein synthesis (very bad). A missense mutation changes one amino acid in the protein (potentially bad, but usually okay). A silent mutation does not change the amino acid. This is due to degeneracy in the genetic code. Degeneracy means that more than one codon can specify for the same amino acid. The reason for this is wobble pairing in the ribosome. The third nucleotide in the codon doesn't have to match perfectly; it can `wobble' in the ribosome. V. Techniques in Molecular Biology This is probably the biggest topic of the second part of the course, and certainly the most practical. Gigantic manuals have been published solely describing these techniques, so I will just go over the most important points here. I'll touch on genomic and cDNA libraries; gel electrophoresis; restriction enzymes; cloning; PCR; Southern and Northern blots; RFLP analysis. Genomic and cDNA libraries. These are libraries of plasmids (circular dsDNA) that each carry a small piece of DNA from a given organism. In a genomic library, the DNA comes from actual DNA in a cell. In a cDNA library, the DNA comes from mature mRNA. The mRNA is reverse transcribed (with reverse transcriptase) to make what's known as cDNA (complementary DNA). So the difference between these two libraries lies in the differences between genomic DNA and mature mRNA. I'll let you ponder that. 5 Patrick Yuh Spring 2007 Gel electrophoresis. Lightning fast review for those who still don't know what this is. You can separate a mixed sample of DNA or RNA based on their size, by running it through a porous gel and applying a current. Negativelycharged nucleic acid will migrate through the gel towards the positive end. You can visualize some of your bands with a radiolabel, or all of them by staining with EtBr. Restriction enzymes. Bacteria contain enzymes that cut dsDNA at specific restriction sites in order to defend themselves from foreign DNA. Different enzymes recognize different restriction sites. A restriction site is just a DNA sequence, like GAATTC, which is the EcoRI site. If you don't know where the restriction sites are located, that's when you make a restriction map. You digest a linear or circular DNA and run it on a gel. Typically you will digest with at least 2 enzymes, separately and together. Based on the fragment sizes, you can put the DNA back together like a puzzle. The only way to get good at this is to practice, but here's how I go about it. Start by drawing in the sites for just one enzyme. Then draw in the other enzyme, remembering that this second enzyme will likely cut in between sites for the first enzyme. Match the bands in the single digest that got smaller in the double digest to figure out where the second enzyme had to have cut. To check your answer, simply digest your map and see if you get the same size bands. Cloning. If you want to study a gene that you found in a genomic or cDNA library, you'll need lots of copies of the gene. We use E. coli to make copies for us. Take your gene which has been put into a plasmid and push it into some bacteria (this is called transformation). If it worked, your bacteria will carry a copy of your plasmid and will make a copy every time they divide. You can select for only cells that have taken up your plasmid by including an antibiotic resistance gene on the plasmid. By growing your cells up in media containing the antibiotic, only the ones carrying the plasmid will survive. But how do you get the plasmid in the first place? PCR (polymerase chain reaction). PCR allows you to make billions of copies of DNA in a few hours. You need at least one copy to begin with, some primers (short oligonucleotides) that are complementary to the region you want amplified, and a DNA polymerase. Southern and Northern blots. If you want to find out whether your favorite gene is present in a sample of DNA or RNA, you can do a Southern (DNA) or Northern (RNA) blot. You run a gel, transfer the nucleic acid to a `piece of paper' (called a membrane), incubate it with a labeled probe to your favorite gene, and see if anything lights up on your membrane. What could you say if there was a band on your Southern blot but not your Northern? RFLP analysis. Going back to restriction sites, you can have a mutation that either generates a new site or gets rid of an existing one. These different forms of sites are called restriction fragment length polymorphisms, or RFLPs. In this case the mutation does not cause a visible phenotype the different restriction pattern is the phenotype. So you can treat RFLPs just like any other allele. You can also use them to calculate recombination frequency. If you have one allele and one RFLP that are linked, you can diagram the crosses, draw out the parental and recombinant gametes, and find RF in the same way. 6 Patrick Yuh Spring 2007 VI. Eukaryotic Chromosomes During mitosis, eukaryotic chromosomes condense into those Xshaped structures you've all seen in textbooks. If you unwind a chromosome, it becomes one very long linear piece of DNA. Organization of DNA into chromosomes divides a genome into manageable units, and also helps with regulation of gene expression. Earlier in the course we talked about karyotyping, which is a way to see the chromosomal makeup of an organism by staining the DNA. It's an easy way to detect aneuploidy, which is a deviation in the normal diploid number of chromosomes. Nondisjunction is the primary cause for aneuploidy in humans. Recall that NDJ leads to gametes with either an extra chromosome or no chromosome. When combined with an okay gamete, this leads to either trisomy (3 copies of a chromosome), or monosomy (one copy). Also recall that trisomy 21 causes Down syndrome. The older the mother gets, the higher the risk for having children with trisomy 21. Why is this? Two proposed models: 1) NDJ in meiosis I from the mother. Oocytes are arrested in metaphase I until ovulation, which could last for over 40 years. So during all this time the egg cells have their homolgous chromosomes being tugged around the metaphase plate by spindle microtubules. Perhaps homologue segregation errors increase over time. Model 2: there are many interactions between the embryo and the uterus during pregnancy. If something is wrong, this can lead to spontaneous abortions. This is called maternal selection. Perhaps, as women age, this selection becomes less stringent. This model is called relaxed selection. Karyotyping can also identify chromosomal rearrangements associated with cancers. Many translocations have been correlated with different types of cancer, including the Philadelphia chromosome, a translocation between chr 9 and 22. This translocation occurs in the middle of two genes, and results in a hybrid gene containing functions of both. One of them is Abl, a protein involved in telling cells to proliferate. So now when the cell receives a signal to express one gene, both genes end up expressed, which leads to inappropriate cell proliferation, which is cancer. A little bit about how chromosomes are packaged...experiments have shown that if you unwind the DNA in a chromosome, it is 100,000 times longer than the condensed Xshaped structure. Clearly there's some packaging going on. The first level of packaging involves DNA wrapping around a group of proteins called histones. Each histone complex consists of 4 pairs of histones H2A, H2B, H3, and H4. These 4 homodimers form an octamer around which the DNA is wrapped. Think of it like a spool of thread. DNA wraps around two turns per octamer. This structure is known as the nucleosome. The Cvalue paradox and heterochromatin. Most of our DNA doesn't code for proteins. Eukaryotes seem to have a lot of DNA they don't need. This excess is called the Cvalue paradox. There are three resolutions to the paradox: 1) big genes maybe there are a lot of very large genes containing long introns; 2) many genes maybe they have a function but don't code for protein; 3) inactive DNA maybe much of the DNA isn't transcribed at all. 7 Patrick Yuh Spring 2007 It turns out that the last option accounts for much of the excess DNA. When you look at stained mitotic chromosomes under a microscope, you see some parts that have been stained more deeply (because they're more condensed). This was named heterochromatin because it looked different from the rest of the stained DNA. Heterochromatic regions were concentrated near the centromeres. Much later it was found that this heterochromatin contains highly repetitive sequence. Does this `junk DNA' or `selfish DNA' have a purpose? Perhaps. Because these regions are always condensed, they're known as constitutive heterochromatin. There's another kind of heterochromatin that is only condensed some of the time, known as facultative heterochromatin. Remember X inactivation? The X that gets turned off becomes a deeply staining Barr body, taking on a heterochromatic appearance. Isn't it cool to make connections with stuff you learned about before? This is a well known example of facultative heterochromatin, where the inactivated X chromosome isn't off all the time, but becomes inactive during development. VII. Regulation of Gene Expression (the lac operon) I'm not going to say too much about the basics of the lac operon since it should still be fresh in your minds. My advice is to draw everything out and follow what happens when you have various mutations in the genes. Cis and trans dominance. The idea here is that proteins can move around, so they act across chromosomes, or in trans. DNA binding sites don't code for protein (the DNA itself serves as a docking site for other proteins), so it can't move. They act in cis, on the same chromosome. So mutations in protein coding genes can act in trans, while mutations in DNA binding sites act in cis. Aside from understanding cis and trans acting elements, just practice following what happens when you have certain mutations. The interesting ones are Is (super repressor) and Oc (operator constitutive). Is codes for a repressor that does not have a lactose binding site, which means it cannot fall off the operator, which means transcription is always OFF. Oc is a mutation in the operator binding site, which means the repressor can no longer bind to it, which means transcription is always ON. You can have combinations of mutations as well. What if you have both Is and Oc? Does it matter if both mutations are on the same chromosome? What happens if you have a mutation in the promoter? These are all questions you should be able to answer. 8 ...
View Full Document
This note was uploaded on 04/07/2008 for the course BIO 105 taught by Professor Sullivan during the Spring '08 term at University of California, Santa Cruz.
- Spring '08