Chapter 16: Applications of Molecular Genetics

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome to the Deep Dive.

We're here to break down complex source material like key chapters in genetics and pull out the essentials for you.

That's right.

And today we're diving into applied molecular genetics.

It's a field with some truly stunning applications.

Absolutely.

Let's start with a really powerful example.

The story of Corey Haas.

Ah, yes, Corey's case.

He had Labor's congenital amaurosis type 2.

It's a, well, a pretty severe inherited blindness.

Caused by faulty copies of a specific gene, right?

The RPE64 gene.

Exactly.

And that gene is crucial.

Its protein product is needed to make rhodopsin.

The pigment our photoreceptors need to function.

Right.

Without it, those photoreceptor cells just degenerate over time.

Leads to severe vision loss, usually from a very young age.

But Corey's story took a different path.

In 2008, he was part of a gene therapy trial.

A really pivotal one.

Doctors injected working copies of that RPE64 gene directly into one of his retinas.

And the results were pretty remarkable, weren't they?

They were profound, really.

He was about six at the time.

After the treatment, he could recognize faces.

He could read.

Things we take for granted.

He could even ride his bike safely.

Activities that were just impossible for him before.

It was a huge success story for gene therapy.

And that kind of success is only possible because of the tools developed through recombinant DNA technology.

That's the core of it.

So our mission today, you know, is to explore how this technology really works in practice.

We'll look at how we find disease genes, the therapies themselves,

using organisms to make useful products, and this new frontier of genome editing.

We're aiming for a clear summary, tracing the path from discovery to, well, to design, all based on the source material.

Okay, let's dive in.

Finding the gene has to come first, right, before you can even think about fixing it.

Absolutely.

And in the early days, before easy genome sequencing,

researchers used something called positional cloning.

Which means finding the gene based on its location on a chromosome, even if you have no idea what the protein it makes actually does.

Exactly.

Huntington's disease, or HD, is the classic example here.

A devastating condition.

Progressive degeneration of the central nervous system.

Autosomal dominant inheritance.

Untreatable.

Right.

And the breakthrough came in 1983.

Researchers found the HD gene was linked to a specific genetic marker.

A marker called an RFLP?

Restriction Fragment Length Polymorphism.

Yes.

Basically a variation in the DNA sequence that changes where certain enzymes, these molecular scissors called restriction enzymes, cut the DNA.

And finding that linkage, even though it was just a marker near the gene on chromosome 4,

that was the key.

That was the crucial first step.

It pointed them in the right direction.

Took another 10 years, mind you, to actually isolate or clone the gene itself, the HDT gene, for Huntington.

And when they found it, the cause was surprising, wasn't it?

This unstable repeat sequence.

Yeah.

A trinucleotide repeat.

CAG, CAG, CAG over and over within the gene.

And the number of repeats is critical.

Hugely critical.

Healthy people might have say 11 to 34 copies of CAG.

Okay.

But in people with Huntington's, that number expands dramatically.

Often 42, sometimes over 100 copies.

And that expansion causes the problems.

It does.

It leads to an abnormally long stretch of the amino acid glutamine in the Huntington protein.

These altered proteins clump together, forming toxic aggregates, especially in neurons.

And there's that chilling correlation.

The more repeats, the earlier the disease starts.

Exactly.

It's an inverse relationship.

More repeats, earlier onset.

Once they knew that sequence, the specific cause,

diagnostics must change overnight.

Oh, completely.

You go from years of painstaking linkage studies to a relatively simple DNA test.

How does that test work?

You use PCR, the polymerase chain reaction to amplify just that specific CAG repeat region from a patient's DNA sample.

Okay.

Make lots of copies of it.

Right.

And then you use gel electrophoresis, specifically polyacrylamide gel, to separate the DNA fragments by size.

That tells you precisely how many CAG repeats there are.

So a quick, accurate diagnosis.

That's powerful.

Very powerful.

And fast.

Results in an afternoon.

Let's talk about another major success story for positional cloning.

Cystic fibrosis, CF.

Right, CF.

Much more common than HD.

It's autosomal recessive, and it causes thick mucus buildup, especially in the lungs and pancreas.

And again, they didn't know the protein involved initially.

Nope.

Biochemical approaches hadn't worked.

So it had to be positional cloning.

They knew it was on the long arm of chromosome 7.

But finding it there was still a huge job.

You mentioned chromosome walks and jumps.

Yeah, those are laborious techniques to sort of, well, walk along the chromosome sequence bit by bit, looking for the gene.

They used clues to help narrow it down.

Like what kind of clues?

Things like CPG island -specific DNA sequences, often found near the start of genes.

And they did zooblots.

Zooblots?

What's that?

It's where you test if a of other animals like mice, chickens, cows.

If it is, it suggests it's important, likely part of a gene.

Clever.

And they also needed the right tissue source, right?

Yeah.

Because the gene isn't active everywhere.

Crucial point.

The CF gene product is mainly in epithelial cells, like those lining sweat glands.

So they screen a CDNA library made from sweat gland cells.

And that finally led them to the CFTR gene.

Yes.

CFTR,

Cystic Fibrosis Transmembrane Conductance Regulator, a big name for a big protein 1480 amino acids.

And what does it do?

It's an ion channel.

Specifically, it controls the movement of chloride ions across cell membranes.

This regulation is vital for maintaining the right balance of salt and water, especially in mucus.

And the most common mutation messes that up.

Right.

The most frequent one, about 70 % of cases, is called Delta F508.

It's a deletion of just three DNA bases, causing the loss of single amino acid, phenylalanine, at position 508 in the protein.

So again, knowing the sequence allows for direct testing.

Exactly.

And it doesn't always have to be a repeat or a deletion.

Sometimes it's a single letter change, a point mutation.

Like in sickle cell anemia.

Perfect example.

The mutation that causes sickle cell hemoglobin, the HBBS allele, changes just one DNA base.

But that single change happens to eliminate a cutting site for a specific restriction enzyme, MTII.

Ah.

So you can use that enzyme to test for it.

Precisely.

You use PCR to amplify the relevant part of the beta -globin gene, and then you add the MTII enzyme.

If it cuts, it's the normal allele.

If it doesn't cut, it's the sickle cell allele.

Simple, elegant diagnostics based on sequence knowledge.

Okay.

So we've gotten really good at finding the genetic culprits, which brings us back to Corey Haas and the idea of fixing them.

Let's talk gene therapy.

Right.

Gene therapy, in its most common form, means adding a functional wild type copy of a trans gene into the cells of someone who has defective copies.

And why add a gene instead of just supplying the missing protein?

Good question.

Often the proteins themselves, especially enzymes, are unstable.

Or they need to be inside the cell to work correctly.

You can't just inject them.

The cells need to make them continuously.

Makes sense.

Now there are different types of gene therapy, aren't there?

Yes.

A critical distinction.

There's somatic cell gene therapy, which targets the body cells of the individual.

Any changes are not passed on to their children.

That's where current human trials are focused.

And the other type.

That's germline gene therapy.

This would target sperm or egg cells or early embryos.

Changes would be heritable, passed down through generations.

Which raises huge ethical questions.

Massive ethical and societal concerns.

So germline therapy isn't practiced in humans.

All the focus is on somatic cells.

Okay.

So how do you get the good gene into the patient cells?

That seems like the big hurdle.

It is the major challenge.

Viruses are often used as delivery vehicles or vectors because they're naturally good at getting genetic material into cells.

Which types of viruses?

Two main types have been common.

Retroviral vectors are one.

They actually integrate their genetic material, including the transgene, right into the host cell's chromosomes.

That sounds good for long -term effect, right?

Stable integration.

It can be, yes.

Potential for long -lasting expression.

But there's a big catch.

Random insertion.

The retrovirus can stitch itself into the genome almost anywhere.

If it lands in the middle of an important gene or near a gene that controls cell growth,

well that can cause serious problems.

Like cancer.

Okay.

That's a significant risk.

What's the alternative?

Adenoviral vectors are another option.

They typically don't integrate into the host chromosome.

The transgene stays separate, sort of like a mini chromosome or plasmid in the nucleus.

Less risk of disrupting host genes, then?

Less risk of insertional mutagenesis, yes.

But the downside is often transient expression.

The effect might not last long as cells divide.

And adenoviruses can trigger a strong immune response in the patient.

So trade -offs either way.

And these risks aren't just theoretical, are they?

There have been serious setbacks in trials.

Absolutely.

We have to remember the history.

The first human trial for ADS -CID was back in 1990.

But a later trial for a different immunodeficiency, X -linked SCID, around 2000, highlighted the dangers.

What happened there?

It seemed successful at first, didn't it?

It did.

It restored immune function in quite a few young boys.

A real triumph, initially.

But then, the tragedy,

four of those patients later developed T -cell leukemia.

Oh no.

And they traced it back to the vector.

They did.

The retroviral vector, doing what retroviruses sometimes do, had inserted itself right next to a proto -oncogene called LMO2.

A gene involved in controlling cell growth.

Exactly.

The insertion essentially turned that gene on inappropriately, leading to uncontrolled T -cell proliferation leukemia.

That must have been a devastating blow to the field.

It was.

It showed that vector insertion wasn't always random.

Sometimes they preferentially integrate near -active genes.

It forced a major reevaluation of safety protocols.

So where does that leave gene therapy now?

What's the goal?

Well, most current approaches are still gene addition.

You put the gene in and hope it lands safely and works.

Not ideal.

Not ideal.

The ultimate goal, the holy grail really, is targeted gene transfer.

Using techniques like homologous recombination to swap out the defective gene with the functional copy right in its normal spot on the chromosome.

Precise replacement, not just addition.

Exactly.

That requires much safer vectors and, crucially, that precise targeting technology, which we'll get to later with things like CRISPR.

Okay.

So while gene therapy grapples with these in vivo challenges, the same core recombinant DNA tech was also being used for something maybe less dramatic, but incredibly impactful.

Making proteins outside the body.

Right.

Large -scale production of valuable human proteins.

Think human insulin or human growth hormone, HGH.

Before genetic engineering, where did those come from?

It was difficult.

Insulin came from pigs and cows, which sometimes caused immune reactions.

HGH had to be extracted from the pituitary glands of human cadavers.

Scarce, expensive, and carried risks like transmitting diseases.

Not a great situation.

Not at all.

Genetic engineering changed everything.

Now we can get bacteria, like good old E.

coli, to produce these human proteins safely and economically.

How do you convince bacteria to make a human protein?

You have to splice the human gene's coding sequence, the part that actually spells out the protein, together with bacterial control signals.

Control signals?

Like promoters?

Exactly.

You need a bacterial promoter to tell the bacterial machinery when and how much to transcribe the gene, and a bacterial ribosome binding site to tell the ribosomes where to start translating the message into protein.

So you might hook the HGH gene up to controls from, say, the E.

coli lac operon.

Precisely.

That way, the bacterium treats the human gene just like one of its own, churning out the human protein.

Basically turning bacteria into little protein factories.

This ability to manipulate genes extends beyond bacteria, right, to creating whole transgenic organisms.

Absolutely.

Transgenic animals, especially mice, are hugely important in research.

How do you make a transgenic mouse?

There are a couple of main ways.

One is direct microinjection.

You physically inject copies of your gene of interest directly into the pronucleus of a fertilized mouse egg.

The pronucleus being the nucleus before the male and female genetic material have fully merged.

Correct.

Some of those eggs will integrate the injected DNA into their genome.

You implant the eggs into a surrogate mother, and some of the resulting pups will be transgenic.

Okay.

What's the other method?

The other involves embryonic stem cells, or ES cells.

These cells are derived from early embryos,

glastocysts, and they have the potential to develop into any cell type in the body.

So you modify the ES cells first.

Right.

You introduce your gene construct into ES cells grown in a dish.

Then you select the cells that have successfully incorporated the gene.

Then you inject these modified ES cells back into a normal mouse blastocyst.

This creates a chimeric embryo, a mix of normal cells and modified cells.

A chimera meaning an organism composed of cells from different genetic origins.

Exactly.

You implant this chimeric embryo into a surrogate mother.

If the modified ES cells contribute to the formation of the germ line, the sperm or eggs, resulting chimeric mouse, then that mouse can pass the transgene onto its offspring.

Giving you a stable line of transgenic mice carrying that specific genetic modification.

Very powerful for studying gene function or modeling human diseases.

Extremely powerful.

You can study growth, like mice engineered to overproduce growth hormone or create models for complex diseases.

And what about plants?

Can we engineer them too?

Yes.

And plants have a natural genetic engineer we've learned to exploit.

A bacterium called agrobacterium tumfations.

I've heard of this one.

It causes tumors in plants.

It does.

Crown galls.

It does this using a special plasmid called the T -plasmid for tumor inducing.

How does the plasmid cause tumors?

The T -plasmid contains a specific segment of DNA called the T -DNA, or transfer DNA.

The bacterium actually transfers this T -DNA segment from the plasmid and integrates it into the chromosome of the host plant cell.

Wow.

Natural genetic engineering.

Pretty much.

The genes within the T -DNA then direct the plant cell to overproduce certain hormones causing the tumor and also to produce special compounds that the bacterium feeds on.

So researchers figured out how to use this system.

Exactly.

They disarmed the key plasmid by removing the tumor causing genes from the T -DNA.

Then they could insert genes they were interested in maybe for herbicide resistance or pest resistance into that T -DNA region.

And the bacteria still transfers this modified T -DNA into the plant cell.

It does.

And a key advantage with plants is totipodency.

Meaning a single plant cell can often be cultured and induced to regenerate into a whole complete plant.

Right.

So you modify one cell using agrobacterium, grow that cell up, and regenerate a fully transgenic plant carrying your desired trait.

It's been revolutionary for agriculture.

Okay.

This leads us towards manipulating genes not just to add them, but to understand their function by turning them off.

Reverse genetics.

Exactly.

Classical genetics goes phenotype to gene.

You see a trait, you find the gene responsible.

Reverse genetics flips that.

You start with a known gene sequence.

And you design a way to specifically inhibit its expression or create a null mutation, completely knock it out to see what happens, to figure out its function.

And knockout mice are a prime example of this.

The definitive example, really.

The process builds on that ES cell technology we discussed.

How does it work specifically to knock out a gene?

You design a gene targeting vector.

This piece of DNA contains a copy of your target gene that you've deliberately disabled, made non -functional.

Critically, you also include a selectable marker gene, usually nearer, which provides resistance to the antibiotic neomycin.

Okay.

So you introduce this vector into ES cells.

Right.

Now two things can happen.

The vector might just insert randomly somewhere in the genome, or through a rare process called homologous recombination, it might actually find the cell's original copy of the target gene and replace it with your disabled version.

That replacement is what you want to knock out.

How do you select for those rare cells?

That's the clutter part.

First, you treat the cells with neomycin.

Only cells that have incorporated the vector DNA somewhere and thus have the near gene will survive.

Okay.

That gets rid of cells that didn't take up the DNA at all.

But how do you distinguish between random insertion and the desired homologous recombination?

You add a second marker gene to the outside ends of your targeting vector, often the TKHSV gene.

If the vector inserts randomly, the whole thing, including TKHSV, gets incorporated.

If homologous recombination occurs correctly, only the disabled gene and near swap in, the TKHSV gene gets left out.

I see.

And there's a drug that targets TKHSV.

Yes.

Gansacolover.

It's toxic to cells that have the TKHSV gene.

So you treat the neomycin -resistant cells with Gansacolover.

Ah.

Cells with random insertions die.

Cells with the targeted knockout survive because they have near but lost TKHSV.

Precisely.

It's a powerful selection strategy.

You then take those doubly selected ES cells, inject them into blastocysts, create pimeric mice, and breed them to get mice completely lacking the target gene.

A lot of work, but you get a definitive answer about the gene's function.

Absolutely.

Now there's another powerful technique for silencing genes, maybe less permanent but often faster, RNA interference or RNAi.

Right.

Using double -stranded RNA to shut down gene expression.

Exactly.

The cell has natural machinery to deal with double -stranded RNA, often seeing it as a sign of viral infection.

This machinery, involving enzymes like Dyser and the RISC complex, RNA -induced silencing complex,

processes the dsRNA into small fragments.

And RISC uses these fragments.

It uses one strand of the small RNA fragment as a guide to find messenger RNA, mRNA molecules in the cell that have a matching sequence.

The mRNA carrying the instructions from the gene you want to silence.

Correct.

Once RISC finds the matching mRNA, it either triggers the mRNA's degradation or blocks it from being translated into protein.

Either way, the gene is effectively silenced.

How do scientists trigger this?

How do they get the dsRNA into cells?

Two main ways.

You can synthesize the dsRNA corresponding to your target gene in the lab and directly inject or introduce it into cells.

Okay.

Direct delivery.

Or you can be a bit more indirect.

You can introduce a trans gene, a piece of DNA engineered to produce an RNA molecule that folds back on itself into a hairpin structure.

The cell's machinery then processes this hairpin RNA into the functional dsRNA needed to trigger silencing.

So RNAi is a great tool for quickly checking the function of a gene by temporarily knocking down its activity.

A very powerful tool for functional genomics.

Yes.

Widely used.

But then came the real game changer in terms of manipulating genomes.

CRISPR Cas9.

Oh, absolutely.

CRISPR has revolutionized genetic engineering.

It's adapted from, believe it or not, a bacterial immune system.

Bacteria have immune systems.

They do.

Against viruses, bacteriophages that infect them.

The CRISPR system is like a molecular memory bank of past infections.

Cas9 is an enzyme, an endonuclease, that acts like molecular scissors cutting DNA.

And how does it know where to cut?

That's the key, isn't it?

Precision.

That's the genius of it.

The Cas9 enzyme is guided to a very specific target DNA sequence by a short RNA molecule called a single guide RNA or sgRNA.

So the researcher designs the sgRNA to match the exact spot in the genome they want to target?

Exactly.

The sgRNA directs Cas9 precisely where to make a double strand break in the DNA.

There's one small requirement.

The target site needs to be immediately adjacent to a short sequence called the PAM, the protospacer adjacent motif.

For the common Cas9 from Structococcus pyogenes, this PAM sequence is typically NGG.

So Cas9 cuts the DNA at the targeted spot.

What happens then?

The cell immediately tries to repair that break.

One major repair pathway is called non -homologous end joining, or NHEJ.

And NHEJ is messy.

It's fast, but often error prone.

It just tries to stick the broken ends back together quickly.

This process frequently introduces small insertions or deletions, we call them indels, right at the cut site.

And those indels usually disrupt the gene.

Very often, yes.

They cause frameshift mutations, leading to a non -functional protein product.

So the simplest application of CRISPR -Cas9 is targeted mutagenesis, creating a knockout much more efficiently than the older methods.

But CRISPR can do more than just break things, right?

You can edit or replace sequences, too.

Absolutely.

That's where it gets really sophisticated.

If you want to, say, delete a larger chunk of DNA,

you can use two sgRNAs to guide Cas9 to make cuts on either side of the region you want to remove.

NHEJ repair will then often join the ends, deleting the intervening segment.

Okay, deletion.

What about precise editing or replacement?

For that, you leverage a different DNA repair pathway called homologous recombination, or HR.

Remember how that worked in the knockout mice?

Right.

Using a template to guide the repair.

Exactly.

So you provide Cas9 and the sgRNA to make the targeted cut, and you simultaneously provide the cell with a donor DNA template.

This template contains the desired sequence change, or the correct gene sequence, flanked by sequences that match the DNA on either side of the Cas9 cut site.

And the cell uses this template to repair the break.

The HR machinery uses the donor template to accurately repair the double strand break, incorporating the new sequence in the process.

This allows for precise gene correction, insertion of new sequences, or making specific edits.

It's incredibly versatile.

From finding genes with RFLPs to editing them with CRISPR.

It's an amazing progression.

Before we wrap up, we should touch on one more major application that relies on these molecular tools.

DNA profiling.

Right.

Establishing identity.

Crucial in forensics and paternity testing.

It works by looking at variations, polymorphisms, in DNA sequences that are unique to individuals.

We don't look at the whole genome though, right?

We focus on specific regions.

Correct.

The standard today relies on short tandem repeats, or STRs.

These are short sequences of DNA, typically 2 to 10 base pairs long, that are repeated over and over again in specific locations in the genome.

And the number of repeats varies between people.

Highly variable.

At a given STR locus or location, you might have inherited, say, 7 repeats from your mother and 10 repeats from your father.

Someone else might have 8 and 8, or 12 and 9.

So analyzing multiple STR loci gives you a unique pattern.

A highly unique pattern.

The FBI uses a standard set of 13 STR loci, called the CODIS Panel Combined DNA Index System.

How is the analysis actually done?

It's not looking at gels anymore, is it?

Not typically visual, like the old RFLP days.

It's highly automated.

You use multiplex PCR, meaning you amplify all 13 CODIS loci simultaneously in a single reaction tube.

Using primers tagged with fluorescent dyes.

Exactly.

Different dyes for different loci, or different size ranges.

Then, the amplified, fluorescently labeled STR fragments are separated by size with extreme precision using capillary gel electrophoresis.

Like a very thin automated gel.

Kind of.

As the fragments pass a detector, lasers excite the fluorescent tags, and a sensor records the color and timing, which translates directly to the size, and thus the number of repeats at each locus.

The output is a graph, an electrophoregram, showing peaks for each allele.

Generating a unique DNA profile.

How unique are we talking?

Statistically incredibly powerful.

For example, the chance that two unrelated Caucasian individuals would happen to match at all 13 CODIS loci is estimated to be about 1 in 5 .75 trillion.

Wow.

That explains its power in forensics.

And paternity testing.

Same principle.

A child inherits one allele at each locus from their mother, and one from their father.

So any STR marker present in the child must also be present in either the mother or the alleged father.

If there are multiple mismatches, paternity can be excluded with very high certainty.

Okay, what a journey we've covered.

From painstakingly tracking down the Huntington's gene using linkage to RFLPs.

Through using RNAi to probe gene function.

All the way to the precision targeting of CRISPR -Cas9 for genome editing.

It really feels like we've gone from just observing genetics to actively engineering it.

It's a paradigm shift, no question.

And if you think back, we started with the risks seen in early gene therapy.

That tragic LMO2 insertion causing leukemia.

A consequence of imprecise, somewhat random integration.

Right.

A serious unintended consequence.

But now we have tools like CRISPR -Cas9 that offer the potential for that ideal scenario we talked about.

Precise gene replacement using homologous recombination.

Fixing the gene exactly where it belongs.

Much safer, theoretically.

Much safer, potentially.

Which leads to a really important question for you, the listener, to consider.

Given the high stakes and potential rewards of older therapies versus the increasing precision of new tools,

what's the moral calculus now?

How should scientists and society balance the drive to treat devastating diseases with the responsibility that comes with these powerful new genome editing capabilities?

That's a deep question.

Really framing the future of genetic medicine.

It's definitely something to think about.

Thank you for guiding us through all that complexity.

My pleasure.

It's fascinating stuff.

Thank you for listening.

And on behalf of the Last Minute Lecture Team, thanks for joining us on this deep dive into molecular applications.

We'll catch you on the next one.