Biotechnology: PCR, Gel Electrophoresis & Genetic Engineering
Learn how restriction enzymes, plasmids, PCR, gel electrophoresis and CRISPR cut, copy, sort and edit DNA — plus how to read gel bands and DNA fingerprints.
What you'll do in this lesson
A voice-first session with the Crimsora tutor on Biotechnology: PCR, Gel Electrophoresis & Genetic Engineering, then targeted practice and FRQs — with the tutor adapting to where you get stuck.
What this lesson covers
You already know DNA's structure, how it is replicated, and how genes are transcribed and translated into proteins. Biotechnology is what happens when humans take those same molecular tools out of the cell and use them on purpose. With a handful of enzymes and a machine that heats and cools test tubes, a lab can cut a gene out of human DNA, paste it into a bacterium, make a billion copies of a single molecule overnight, sort fragments by size on a gel, and rewrite a single base in a living genome.
This lesson organizes those techniques by their job: cut (restriction enzymes), carry (plasmid vectors), copy (PCR), sort (gel electrophoresis), and edit (CRISPR-Cas9). Then you will practice the skill teachers ask for most on this topic — looking at a pattern of bands on a gel and explaining what it says about who a sample came from and how closely two organisms are related.
This lesson organizes those techniques by their job: cut (restriction enzymes), carry (plasmid vectors), copy (PCR), sort (gel electrophoresis), and edit (CRISPR-Cas9). Then you will practice the skill teachers ask for most on this topic — looking at a pattern of bands on a gel and explaining what it says about who a sample came from and how closely two organisms are related.
Cutting DNA: Restriction Enzymes and Sticky Ends
Restriction enzymes (restriction endonucleases) are bacterial proteins that cut double-stranded DNA at specific short sequences called recognition sites. Bacteria evolved them as a defense against viral DNA; molecular biologists borrowed them as molecular scissors.
Most recognition sites are 4–8 base pairs long and palindromic — the sequence reads the same on both strands. The enzyme EcoRI recognizesand cuts between the G and the first A on each strand. Because the two cuts are offset, each fragment ends with a short single-stranded overhang (AATT) called a sticky end. Sticky ends matter enormously: any two pieces of DNA cut with the same enzyme have complementary overhangs, so they can hydrogen-bond to each other regardless of what organism they came from. The enzyme DNA ligase then seals the sugar-phosphate backbone permanently. Some enzymes cut straight across and leave blunt ends, which join less specifically.
A counting rule your teacher will use: cutting a linear DNA molecule at sites produces fragments, but cutting a circular molecule (like a plasmid) at sites produces exactly fragments. Students routinely miss this and report one fragment too many for plasmids.
The key misconception here is that restriction enzymes "know" where genes are. They do not. They cut wherever their recognition sequence appears, which may be inside a gene, between genes, or thousands of times in a large genome. Choosing an enzyme whose site flanks — but does not interrupt — the gene of interest is part of experimental design.
Most recognition sites are 4–8 base pairs long and palindromic — the sequence reads the same on both strands. The enzyme EcoRI recognizesand cuts between the G and the first A on each strand. Because the two cuts are offset, each fragment ends with a short single-stranded overhang (AATT) called a sticky end. Sticky ends matter enormously: any two pieces of DNA cut with the same enzyme have complementary overhangs, so they can hydrogen-bond to each other regardless of what organism they came from. The enzyme DNA ligase then seals the sugar-phosphate backbone permanently. Some enzymes cut straight across and leave blunt ends, which join less specifically.
A counting rule your teacher will use: cutting a linear DNA molecule at sites produces fragments, but cutting a circular molecule (like a plasmid) at sites produces exactly fragments. Students routinely miss this and report one fragment too many for plasmids.
The key misconception here is that restriction enzymes "know" where genes are. They do not. They cut wherever their recognition sequence appears, which may be inside a gene, between genes, or thousands of times in a large genome. Choosing an enzyme whose site flanks — but does not interrupt — the gene of interest is part of experimental design.
Carrying and Copying: Plasmid Vectors, Transformation, and PCR
A cut-out gene is useless unless it can be replicated. Two strategies do that.
Recombinant DNA in a plasmid vector. A plasmid is a small circular DNA molecule found in bacteria that replicates independently of the chromosome. A useful vector carries an origin of replication, at least one restriction site, and a selectable marker such as an antibiotic-resistance gene. The plasmid and the donor DNA are cut with the same enzyme, mixed, and joined by ligase to form recombinant DNA. Bacteria then take the plasmid up during transformation (often triggered by calcium ions plus heat shock). Growing the bacteria on antibiotic-containing medium kills every cell that failed to take up a plasmid, so only transformed colonies survive — that is how the marker gene lets you "select" success. Because the central dogma works the same way in bacteria, the cells transcribe and translate the inserted gene, which is how human insulin and human growth hormone are manufactured.
PCR (polymerase chain reaction) copies DNA in a tube instead of a cell. Each cycle has three temperature steps.
Taq polymerase comes from a hot-spring bacterium, so it survives the 95-degree step. Each cycle doubles the target, so from one starting molecule you get copies after cycles — 30 cycles yields over a billion. PCR needs a template, primers, free nucleotides, and buffer; it does not need living cells, helicase, or RNA primers, which is where students often confuse it with replication in the nucleus.
Recombinant DNA in a plasmid vector. A plasmid is a small circular DNA molecule found in bacteria that replicates independently of the chromosome. A useful vector carries an origin of replication, at least one restriction site, and a selectable marker such as an antibiotic-resistance gene. The plasmid and the donor DNA are cut with the same enzyme, mixed, and joined by ligase to form recombinant DNA. Bacteria then take the plasmid up during transformation (often triggered by calcium ions plus heat shock). Growing the bacteria on antibiotic-containing medium kills every cell that failed to take up a plasmid, so only transformed colonies survive — that is how the marker gene lets you "select" success. Because the central dogma works the same way in bacteria, the cells transcribe and translate the inserted gene, which is how human insulin and human growth hormone are manufactured.
PCR (polymerase chain reaction) copies DNA in a tube instead of a cell. Each cycle has three temperature steps.
| Step | Temperature | What happens |
|---|---|---|
| Denaturation | about 95 degrees Celsius | Hydrogen bonds break; strands separate |
| Annealing | about 50–65 degrees Celsius | Short primers base-pair to sequences flanking the target |
| Extension | about 72 degrees Celsius | Taq polymerase adds nucleotides |
Sorting DNA: How Gel Electrophoresis Works
Gel electrophoresis separates DNA fragments by size. A slab of agarose gel — a mesh of tangled polysaccharide fibers — sits in buffer between two electrodes. Samples are loaded into wells at one end, and the power supply is turned on.
Every nucleotide carries a negatively charged phosphate group, so all DNA fragments are negatively charged and all of them migrate toward the positive electrode (the anode). The separation therefore is not caused by differences in charge. It is caused by friction: short fragments slip through the pores of the gel easily and travel far from the wells, while long fragments snag and stay near the top. After the run, a stain (or fluorescent dye) reveals the fragments as bands, each band a huge population of identically sized molecules.
To convert band position into a number, one lane is loaded with a DNA ladder (size standard) containing fragments of known length in base pairs. You compare an unknown band's distance to the ladder to estimate its size.
Two errors show up constantly. First, reversing the size relationship — writing that the largest fragment travels farthest. Second, treating band thickness as size; a thick or bright band usually means more copies of that fragment, not a bigger fragment. Also remember that gels reveal length, not sequence. Two completely different 2,000 base-pair fragments produce bands at the same place. A gel narrows possibilities; sequencing identifies.
Every nucleotide carries a negatively charged phosphate group, so all DNA fragments are negatively charged and all of them migrate toward the positive electrode (the anode). The separation therefore is not caused by differences in charge. It is caused by friction: short fragments slip through the pores of the gel easily and travel far from the wells, while long fragments snag and stay near the top. After the run, a stain (or fluorescent dye) reveals the fragments as bands, each band a huge population of identically sized molecules.
To convert band position into a number, one lane is loaded with a DNA ladder (size standard) containing fragments of known length in base pairs. You compare an unknown band's distance to the ladder to estimate its size.
Two errors show up constantly. First, reversing the size relationship — writing that the largest fragment travels farthest. Second, treating band thickness as size; a thick or bright band usually means more copies of that fragment, not a bigger fragment. Also remember that gels reveal length, not sequence. Two completely different 2,000 base-pair fragments produce bands at the same place. A gel narrows possibilities; sequencing identifies.
Reading Gels: DNA Fingerprints, Identity, and Relatedness
DNA fingerprinting (DNA profiling) exploits the fact that non-coding regions of the genome contain short tandem repeats (STRs) — sequences like GATA repeated a variable number of times. Two people rarely have the same repeat number at many loci simultaneously. If you PCR-amplify several STR loci and run the products on a gel, each person produces a distinctive pattern of bands. Alternatively, genomic DNA can be cut with restriction enzymes; because mutations create or destroy recognition sites, different individuals yield different fragment lengths from the same region.
Interpretation rules to apply carefully:
Identity: a suspect or reference sample matches an evidence sample only if the band patterns are the same at every locus examined. One clear mismatched band is enough to exclude, and exclusion is a stronger logical conclusion than a match.
Parentage: a child inherits one allele of each locus from each parent, so every band in the child's lane must appear in the mother's lane or the father's lane. Bands present in a child but absent from both alleged parents rule that pairing out. A band shared with the mother tells you nothing about the father.
Relatedness among species or individuals: more shared bands means more shared sequence, which means a more recent common ancestor. This links directly to the heredity ideas from the previous unit — molecular evidence is just another line of data for the same family trees that anatomy suggests.
Students often over-claim. A gel showing two shared bands out of six does not prove siblings; it shows partial similarity. Always state how many loci were compared before drawing a conclusion.
Interpretation rules to apply carefully:
Identity: a suspect or reference sample matches an evidence sample only if the band patterns are the same at every locus examined. One clear mismatched band is enough to exclude, and exclusion is a stronger logical conclusion than a match.
Parentage: a child inherits one allele of each locus from each parent, so every band in the child's lane must appear in the mother's lane or the father's lane. Bands present in a child but absent from both alleged parents rule that pairing out. A band shared with the mother tells you nothing about the father.
Relatedness among species or individuals: more shared bands means more shared sequence, which means a more recent common ancestor. This links directly to the heredity ideas from the previous unit — molecular evidence is just another line of data for the same family trees that anatomy suggests.
Students often over-claim. A gel showing two shared bands out of six does not prove siblings; it shows partial similarity. Always state how many loci were compared before drawing a conclusion.
Editing DNA: CRISPR-Cas9
Restriction enzymes cut only at their own fixed recognition sequences, so they cannot be aimed at an arbitrary gene. CRISPR-Cas9 solves that. It comes from a bacterial immune system that stores fragments of past viral invaders in Clustered Regularly Interspaced Short Palindromic Repeats.
The engineered tool has two parts. Cas9 is a nuclease that cuts both strands of DNA. A guide RNA (gRNA) carries roughly 20 bases that are complementary to the target DNA sequence. The gRNA leads Cas9 to the matching site by base pairing — the same A-T/G-C rules from the replication lesson — and Cas9 cuts there. What happens next depends on the repair pathway. Sloppy repair inserts or deletes bases and usually produces a frameshift that knocks out the gene. If researchers also supply a DNA template with the desired sequence, the cell can copy it in during repair, correcting or inserting a specific allele.
The programmability is what makes CRISPR different: change the 20-base guide and you change the target, so the same protein can edit any gene.
Limitations and ethics belong here too. Off-target cuts at similar sequences can cause unintended mutations. Editing somatic cells affects one patient; editing germline cells (eggs, sperm, embryos) changes every future descendant, which is why it is heavily restricted. Evaluating those tradeoffs — benefit, risk, cost, who decides — is engineering-design thinking, not just biology.
The engineered tool has two parts. Cas9 is a nuclease that cuts both strands of DNA. A guide RNA (gRNA) carries roughly 20 bases that are complementary to the target DNA sequence. The gRNA leads Cas9 to the matching site by base pairing — the same A-T/G-C rules from the replication lesson — and Cas9 cuts there. What happens next depends on the repair pathway. Sloppy repair inserts or deletes bases and usually produces a frameshift that knocks out the gene. If researchers also supply a DNA template with the desired sequence, the cell can copy it in during repair, correcting or inserting a specific allele.
The programmability is what makes CRISPR different: change the 20-base guide and you change the target, so the same protein can edit any gene.
| Tool | Targets what | Reprogrammable? | Typical use |
|---|---|---|---|
| Restriction enzyme | Its own fixed 4–8 bp site | No | Cutting DNA for cloning |
| Cas9 + gRNA | Any sequence matching the guide | Yes | Editing a gene in a living cell |
Key terms
- Restriction enzyme.
- A bacterial enzyme that cuts double-stranded DNA at a specific palindromic recognition sequence, often leaving single-stranded sticky ends.
- Sticky end.
- A short single-stranded overhang left by an offset restriction cut; complementary sticky ends from the same enzyme allow DNA from different sources to be joined.
- Plasmid vector.
- A small circular bacterial DNA molecule engineered to carry a foreign gene into a host cell; usually contains an origin of replication, restriction sites, and a selectable marker.
- Recombinant DNA.
- A single DNA molecule assembled from segments of two or more sources, typically joined by DNA ligase.
- PCR.
- Polymerase chain reaction — a cyclic heating and cooling process using primers and heat-stable Taq polymerase that doubles a target DNA sequence each cycle, giving copies after cycles.
- Gel electrophoresis.
- A technique that pulls negatively charged DNA fragments through an agarose gel toward the positive electrode, separating them by size because small fragments migrate farther.
- DNA fingerprint.
- An individual-specific pattern of bands produced by analyzing variable regions such as short tandem repeats, used to compare identity or parentage.
- CRISPR-Cas9.
- A programmable editing system in which a guide RNA base-pairs with a chosen DNA sequence and directs the Cas9 nuclease to cut it, allowing gene knockout or targeted correction.
Worked example
A lab tests parentage. Genomic DNA from a mother, her child, and two men (Male A and Male B) is amplified at one variable locus and run on a gel with a ladder. The observed band sizes, in base pairs, are: Mother 4,000 and 2,500; Child 4,000 and 1,800; Male A 3,000 and 1,200; Male B 1,800 and 900. Which man could be the biological father, and how would the child's lane look on the gel compared with the mother's lane?
Step 1 — Apply the inheritance rule. At any locus the child carries two alleles, one from each biological parent. So every band in the child's lane must be matched by a band in the mother's lane or in the biological father's lane.
Step 2 — Assign the maternal band. The child's 4,000 bp band also appears in the mother's lane, so that allele came from her. (Her other allele, 2,500 bp, simply was not passed on.)
Step 3 — Identify the paternal band. The remaining child band is 1,800 bp. It must have come from the biological father, so the father's lane must contain an 1,800 bp band.
Step 4 — Test each man. Male A has 3,000 and 1,200 bp — neither is 1,800, so Male A is excluded. Male B has 1,800 and 900 bp, and his 1,800 bp band matches the child's unexplained band. Male B is not excluded and is consistent with being the father.
Step 5 — Describe the gel positions. Smaller fragments travel farther from the wells. In the child's lane the 1,800 bp band sits noticeably lower (closer to the positive electrode) than the 4,000 bp band. Compared with the mother's lane, the child shares the high 4,000 bp band but lacks her 2,500 bp band and shows an extra, lower band at 1,800 bp.
Step 6 — State the conclusion with appropriate caution. Only one locus was examined, so this evidence excludes Male A but does not prove Male B is the father; any unrelated man carrying an 1,800 bp allele would also match. Real casework compares 13 or more loci to make a match statistically convincing.
Step 2 — Assign the maternal band. The child's 4,000 bp band also appears in the mother's lane, so that allele came from her. (Her other allele, 2,500 bp, simply was not passed on.)
Step 3 — Identify the paternal band. The remaining child band is 1,800 bp. It must have come from the biological father, so the father's lane must contain an 1,800 bp band.
Step 4 — Test each man. Male A has 3,000 and 1,200 bp — neither is 1,800, so Male A is excluded. Male B has 1,800 and 900 bp, and his 1,800 bp band matches the child's unexplained band. Male B is not excluded and is consistent with being the father.
Step 5 — Describe the gel positions. Smaller fragments travel farther from the wells. In the child's lane the 1,800 bp band sits noticeably lower (closer to the positive electrode) than the 4,000 bp band. Compared with the mother's lane, the child shares the high 4,000 bp band but lacks her 2,500 bp band and shows an extra, lower band at 1,800 bp.
Step 6 — State the conclusion with appropriate caution. Only one locus was examined, so this evidence excludes Male A but does not prove Male B is the father; any unrelated man carrying an 1,800 bp allele would also match. Real casework compares 13 or more loci to make a match statistically convincing.
Practice questions
During gel electrophoresis, why do all DNA fragments move toward the positive electrode, and what determines how far each one travels?
- DNA is positively charged, and fragments with more charge travel farther
- DNA is negatively charged because of its phosphate groups, and shorter fragments travel farther
- DNA is negatively charged because of its nitrogenous bases, and longer fragments travel farther
- DNA has no net charge, so travel distance depends only on the amount of dye added
Answer: DNA is negatively charged because of its phosphate groups, and shorter fragments travel farther
Every nucleotide has a phosphate group carrying a negative charge, so all DNA — regardless of length or sequence — is pulled toward the anode. Because the charge-to-length ratio is essentially the same for all fragments, charge cannot be what separates them. The agarose mesh does the sorting: short fragments thread through the pores quickly and end up far from the wells, while long fragments are held back near the top. The choice naming nitrogenous bases is a common mix-up; bases carry the genetic information but not the negative charge.
A circular plasmid 6,000 base pairs long contains exactly two EcoRI recognition sites, located 1,500 base pairs apart. Predict how many bands appear when the digested plasmid is run on a gel and give their sizes. Then explain how the result would differ if the same DNA were linear with those two sites.
Answer: Two bands, at 1,500 bp and 4,500 bp. A linear molecule with two cut sites would instead give three fragments (three bands).
Cutting a circle at two points releases two arcs, not three pieces: one is the 1,500 bp stretch between the sites and the other is the remaining bp. So the gel shows two bands, with the 1,500 bp band farther from the well because it is smaller. For a linear molecule, each cut adds one piece, so cuts give fragments — here three. The general rule to remember is circular DNA gives fragments from cuts, linear DNA gives , and forgetting the circular case is the most common error on plasmid problems.
Both restriction enzymes and CRISPR-Cas9 cut double-stranded DNA. Explain the key difference in how each one is targeted, and why that difference makes CRISPR suitable for correcting a disease-causing mutation in a patient's cells.
Answer: A restriction enzyme recognizes only its own fixed palindromic sequence, while Cas9 is aimed by an interchangeable guide RNA that base-pairs with any chosen sequence — so Cas9 can be pointed at one specific mutated gene.
Targeting in a restriction enzyme is built into the protein's shape; EcoRI will always cut at GAATTC and nowhere else, and that sequence occurs at many random places in a genome, so it cannot single out one gene. With CRISPR, the cutting protein stays the same and the roughly 20-base guide RNA supplies the address through ordinary complementary base pairing. To fix a mutation, researchers design a guide matching the mutant sequence and supply a repair template with the correct bases; the cell copies the template while repairing the cut. The practical caveat is off-target cutting at sequences similar to the guide, which is why guides are designed and tested carefully.
FAQ
- What is the difference between PCR and DNA replication in a cell?
- Both build new strands using a template, base pairing, and a DNA polymerase working . But PCR happens in a tube, uses heat (about 95 degrees Celsius) instead of helicase to separate strands, uses short DNA primers instead of RNA primers, uses heat-stable Taq polymerase, and copies only the small region defined by the primers rather than the whole genome. A cell also needs ligase and many other proteins that PCR does not.
- Why do smaller DNA fragments travel farther on a gel?
- The agarose gel is a mesh of tiny pores. The electric field pulls every fragment equally hard because all DNA carries the same charge per unit length, so the only difference is friction. Small fragments wind through the pores easily and cover more distance in the same run time; long fragments get tangled and stay close to the wells.
- Can a DNA fingerprint prove that two people are related?
- It can strongly support or definitively rule out a relationship, but the strength depends on how many loci are compared. A single mismatched band that cannot be explained by inheritance excludes a relationship outright. A match at one or two loci could easily occur by chance, so labs compare many independent variable loci; sharing patterns at all of them makes a coincidental match extremely unlikely.
- Why do scientists insert human genes into bacteria instead of just making the protein chemically?
- Proteins such as insulin are long, precisely folded chains that are impractical to synthesize chemically at scale. Because the genetic code and the machinery of transcription and translation are nearly universal, a bacterium given the human gene on a plasmid will read it and manufacture the protein itself, dividing rapidly to produce large quantities cheaply and consistently.
Learn this with a teacher, not a page
The Crimsora tutor teaches Biotechnology: PCR, Gel Electrophoresis & Genetic Engineering live — explaining on a whiteboard, asking you questions, and adapting to where you get stuck.