Chapter 24: Medical 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.

You know this show where we try to unpack some really dense info and hopefully pull out the essential bits, the surprising stuff, just for you.

So today, medical genetics.

It's a field that's, well, it's not just future talk, right?

Not at all.

It's happening now.

It's actively changing healthcare.

Yeah,

like hundreds of genetic tests are already being used clinically.

And this idea that your whole genome might be a routine diagnostic tool must be not so far off.

It really isn't.

And for this Deep Dive, we've really gone into chapter 24 of Brooker's Genetics Analysis and Principles.

Our mission, basically, is to unravel how these mutant genes, how they actually lead to human disease.

We'll look at the inheritance pattern, sure, but also the therapies, the cutting edge stuff,

and well, personalized medicine.

Right.

Understanding the roots of illness, how we study them, what it all means for patients.

Exactly.

The implications are huge.

So expect quite a journey.

We'll cover how scientists even know if a disease is genetic to start with.

Yeah, and how they find the specific genes responsible.

Plus the ethics, which are always tricky.

Genetic testing.

Oh, definitely.

And then some surprises, like those protein -only infectious agents, prions.

Wild.

And gene therapy, the potential, but also the hurdles.

Big hurdles sometimes.

And finally, yeah, that personalized medicine frontier.

Okay, let's dig in.

First things first.

How do geneticists even suspect a genetic link for a disease?

You can't do human breeding experiments, obviously.

Right, that's the fundamental challenge.

So they have to be detectives, really.

They rely on observation, looking for patterns in populations and families.

Okay, so what are the clues?

What's the detective kit look like?

Well, one big one is seeing if the disease clusters in families.

So if someone has it, are their close relatives more likely to have it than just random people?

Think cystic fibrosis.

Makes sense.

Runs in the family.

Exactly.

Another really powerful tool is twin studies.

If identical twins, who share almost all their DNA, both get the disease much more often than non -identical twins.

That points to genetics.

Strongly.

But, interestingly, even with identical twins, it's often not 100 % concordance for single gene disorders.

Why not?

If they have the same gene.

It could be incomplete penetrance, meaning you have the gene, but the disease doesn't fully manifest for some reason, or maybe a new mutation occurred very early on.

Environment can play a role, too, even for strongly genetic conditions.

Okay, so it's not always a simple on -off switch.

What else?

If the disease isn't contagious, you know, doesn't spread like a cold or flu that leans towards genetics.

Right, not environmental transmission.

Also look at different populations.

Is it way more common in one ethnic group, like sickle cell disease in people with African or some Asian ancestry?

That's a big clue.

Age of onset can be characteristic, too.

Some genetic disorders show up at birth, others, like Huntington's, might not appear until middle age.

And comparing to animals.

Yeah, sometimes a human condition looks remarkably similar to a known genetic disorder in, say, mice or dogs, like albinism that pops up across species.

But the real smoking gun must be looking at the DNA itself.

Absolutely.

The most convincing evidence is finding a direct link.

This specific mutation or this chromosomal abnormality consistently shows up in people with the disease.

That's where DNA sequencing and karyotyping come in, actually seeing the change.

Okay, so you suspect genetics.

Now, how does it get passed down?

This is where the family trees, the pedigrees come in, Mendelian patterns.

Where should we start?

Autosomal recessive.

Let's do it.

Autosomal recessive.

Tay -Sachs disease is a really tragic classic example.

Tell us about it.

Okay, so babies seem healthy at first, but around four, six months, things go wrong.

Rapid neurodegeneration, blindness, loss of movement, usually fatal by age three or four.

That's awful.

And it's more common in certain groups.

Yes, particularly in Ashkenazi Jewish populations, about one in 3 ,600 births.

Much rarer elsewhere.

What's happening molecularly?

What's the genetic insight?

It's a defect in the hexa gene.

This gene makes an enzyme, hexavinidase A.

Think of it as a cellular recycling crew breaking down fatty substances called GM2 gangliosides, especially in the brain.

So if the enzyme is broken?

Right.

If hexa is mutated, the enzyme doesn't work properly.

These gangliosides build up, become toxic, and basically overwhelm the nerve cells.

And the key features of this inheritance pattern?

Recessive.

Key features.

Affected kids often have two unaffected parents.

Those parents are carriers, heterozygotes.

If two carriers have children, there's roughly a 25 % chance each child will be affected, and it hits males and females equally.

Got it.

Other examples.

Oh, sure.

Albinism, cystic fibrosis, PKU, sickle cell, all typically autosomal recessive.

Okay.

Let's switch gears.

Autosomal dominant.

Sounds like a different ballgame.

It is.

Huntington disease is the poster child here, usually hits in middle age.

Progressive neurodegeneration, movement problems, personality changes, dementia leads to early death.

And the molecular cause is fascinating, right?

Something about repeats.

Exactly.

It's caused by a trinucleotide repeat expansion.

Basically, a three -letter DNA sequence, CAG, gets repeated too many times within the Huntington gene.

Like a stutter in the gene.

Kind of.

This leads to a faulty Huntington protein with a long tail of glutamine amino acids.

This altered protein misfolds, clumps together aggregates, and damages neurons.

So for dominant disorders, the inheritance pattern is?

Usually an affected person has at least one affected parent.

If one parent is affected, and heterozygous, each child has a 50 % chance of inheriting it.

And again, equal frequency in males and females.

What happens if someone gets two copies of the faulty dominant gene?

Often, that's much more severe, sometimes lethal.

Homozygos are rare for many dominant disorders because it's so detrimental.

Why do dominant mutations cause problems?

Are there different ways?

Yeah, generally three main mechanisms.

First, haplone sufficiency.

One good copy of the gene just isn't enough to do the job properly, like needing two engines, but one is broken.

Second, gain of function.

The mutated gene product does something new or abnormal or it's overly active.

A chondroplasia, a form of dwarfism, is like this, a receptor protein is stuck in the on position.

And the third.

Dominant negative, this is kind of insidious.

The faulty protein doesn't just fail to work, it actively interferes with the normal protein produced by the good copy.

Marfan syndrome is an example,

defective fibrillin protein weakens connective tissue, like one bad support beam compromising the whole structure.

Okay, now on to the X chromosome.

X -linked recessive inheritance.

Hemophilia A.

Yep.

Classic hemophilia.

Caused by a faulty gene on the X chromosome that codes for clotting factor 8, essential for blood clotting.

And because males only have one X chromosome.

Exactly, they're hamozygous.

If they inherit that faulty X, they have the disease.

Females have two Xs, so they can be carriers, usually without symptoms, because their other X is a working copy.

This is the royal disease, right?

That's the one.

Traced back to Queen Victoria, it spread through European royalty because daughters were carriers and passed it to their sons.

So key features, males affected way more often.

Mothers of affected sons are often carriers and might have affected brothers or fathers.

Daughters of affected men are carriers.

And those carrier daughters have a 50 % chance of having an affected son.

That's the pattern.

What about X -link dominant?

You hear less about that.

Much rarer.

And often very severe, especially in males, sometimes lethal before birth, so you mostly see it in females.

Rett syndrome is an example.

Often these cases arise from new mutations, not necessarily inherited from a parent.

This all seems complex enough.

But then there's locus heterogeneity.

What's that?

Sounds like it could mess up figuring out inheritance patterns.

It absolutely can.

Locus heterogeneity means the same disease phenotype, the same set of symptoms can be caused by mutations in different genes.

So it looks like one thing, but it's genetically diverse.

Precisely.

Hemophilia is a perfect example again.

Hemophilia A and hemophilia B are both X -linked recessive, but caused by mutations in two totally different genes on the X chromosome, factor VIII and factor IX genes respectively.

And then there's hemophilia C that's autosomal recessive caused by a mutation on chromosome IV.

So if you just looked at a family with various members having hemophilia, you might struggle to see a clear single pattern if different types are present.

It's a crucial insight.

Don't assume one disease name means only one gene is involved.

Right.

That's a really important point for diagnosis.

OK, so we see the patterns, but the big hunt is finding the actual gene, the specific spot on the DNA.

How on earth do they narrow that down from billions of base pairs?

Yeah, it's a huge challenge.

One really powerful concept here is the haplotype.

Haplotype.

Define that for us.

OK.

Think of it as a specific set of genetic markers, variations like SNPs, that are located close together on the same chromosome and tend to be inherited as a block, like a unique barcode for a segment of your chromosome.

And these blocks stay together over generations because the markers are close, less likely to be broken up by crossing over.

Exactly.

The closer they are,

the tighter the linkage.

And this leads to haplotype association studies.

The idea is, if a disease -causing mutation first arose in one person, a founder, generations ago, it probably arose on a chromosome that had a specific haplotype, a specific barcode.

And because the mutation and those nearby markers are linked, that same haplotype is likely still found in most people who have the disease today,

descendants of that founder.

Ah, so you look for a haplotype that consistently travels with the disease.

That's the core idea.

Find the shared barcode among affected people.

And the Huntington disease story with Nancy Wexler is a prime example of this.

A landmark example.

Truly incredible work.

Nancy Wexler, driven by Huntington's and her own family, led this massive study in Venezuela tracing the disease through generations involving thousands of people.

Her team eventually found a specific DNA marker called G8C on chromosome 4.

This marker was almost always inherited, along with the Huntington's mutation itself.

This strong association linkage disequilibrium was the key breakthrough that pinpointed the gene's neighborhood.

So they knew roughly where to look on chromosome 4.

Right.

Back then, they used laborious techniques like chromosome walking.

Now researchers can analyze maybe a million -base pair region flagged by association, look at the 5, 10 candidate genes within it based on function, and then sequence those candidates in affected people to find the actual mutation.

That's amazing detective work, but that focuses on families.

What about finding genes for common, complex diseases across populations?

For that, we have genome -wide association studies, or GWAs.

GWAs.

How does that work?

Instead of just families, GWA looks at huge numbers of people, some with a disease, say type 1 diabetes, and some without the controls.

It scans across their entire genomes, testing millions of common genetic variations, usually to see if any specific variants are significantly more frequent in the group with the disease.

So you're looking for statistical associations across the whole genome.

Exactly.

And the results are often shown in these plots called Manhattan plots.

Why Manhattan?

Because the significant associations show up as tall peaks, or skyscrapers, against a baseline of non -significant points.

It plots the statistical significance, the t -value, for each SMP against its position on the chromosomes.

The higher the peak, the stronger the association.

But association isn't causation, right?

Crucial point.

A SMP that pops up in a GWAS might not be the cause of the disease itself.

It might just be physically close on the chromosome to the actual causative variant and gets inherited along with it due to that linkage disequilibrium, that founder effect again.

So GWAs points you to regions, but more work is needed to find the functional culprit.

And projects like the International HapMap Project helped lay the groundwork for GWAS.

Immensely.

HapMap cataloged common human genetic variations, like SMPs, and mapped how they are distributed in different populations.

It created this essential reference map that researchers used to design and interpret GWAS and other genetic studies.

A huge collaborative effort.

Okay, so we found the genes, maybe.

Now let's bring it into the clinic.

Genetic testing versus genetic screening, what's the difference?

Really important distinction.

Genetic testing is done on an individual, usually because they have symptoms, or a family history suggests they're at risk.

It's often diagnostic.

Genetic screening is different.

It's testing offered to a whole population, or a subset like newborns, regardless of symptoms, to identify individuals at risk for a specific disorder where early intervention can help.

Got it.

What kinds of methods are used for these tests?

You can test at the protein level, remember Tay -Sachs.

You can do an enzyme assay, measure hexaactivity directly using fluorescence.

Different levels mean normal, carrier, or affected.

But mostly it's DNA, though.

Yeah, mostly DNA or chromosome level.

Direct DNA sequencing if you know the gene.

Karyotyping to look at whole chromosomes, number, structure.

Surprisingly common to have alterations, maybe 1 in 200 births.

Then there are fishies, DNA microarrays, lots of tools.

And screening has had some major public health successes.

Oh, absolutely.

Newborn screening for PKU, fetal kinonuria, is a classic.

Find it early, modify the diet, and you prevent severe intellectual disability.

Huge impact.

And the Tay -Sachs screening programs in Ashkenazi Jewish communities.

Education and carrier screening dramatically reduce the incidence, like 90 % in a generation.

Amazing success.

But for most rare diseases, screening everyone isn't feasible or cost -effective, so testing is usually targeted.

We can also test before birth, right?

Prenatal testing.

Yes.

Amniocentesis samples fetal cells from the amniotic fluid, usually around 14, 20 weeks.

Cells need to be cultured, so results take a bit.

And the other one, CVS.

Chorionic villus sampling, or CVS, done earlier, 10, 12 weeks, takes a tiny sample of the placenta chorionic villi.

Results are faster because you often don't need to culture cells as long, but there's a slightly higher risk of miscarriage compared to amnio.

And then there's the really cutting -edge stuff with IVF, PGP.

Pre -implantation genetic diagnosis.

This is done alongside in vitro fertilization.

Before implanting an embryo, you can carefully remove one or two cells when it's at the 8 -cell stage.

Just one or two cells.

And test those cells for specific genetic problems, like the Huntington's mutation or for chromosomal abnormalities like aneuploidy.

Then you select only the embryos without the condition for implantation.

That's incredibly powerful technology, which brings us squarely to ethics.

What keeps you up at night, ethically speaking, with all this?

Oh boy.

It's complex.

A huge one is testing for conditions we can't yet treat or prevent, especially late -onset ones like Huntington's.

Knowing you'll likely develop a devastating disease decades from now, what do you do with that information?

The psychological burden can be immense.

I'm right not to know.

Exactly.

Then there's privacy, who gets to see your genetic blueprint.

And the potential for discrimination by employers, insurance companies.

That's a massive societal concern we're still wrestling with.

How do we make fair rules?

It's tough.

Really tough questions.

Okay, shifting gears again slightly.

Prions.

This still blows my mind.

A disease agent made only of protein.

No DNA, no RNA.

It seems to break the central dogma, doesn't it?

But yeah, prions, prokynaceous, infectious particles, they cause fatal neurodegenerative diseases.

Kuru, mad cow, scraping in animals, Creutzfeldt -Jakob, fatal familial insomnia in humans.

It's all about protein shape.

So how does a protein replicate itself?

It doesn't replicate in the traditional sense.

There's a normal version of the prion protein, PRPC, that everyone has.

The infectious form, PRPSC, has the same amino acid sequence but is folded into a different abnormal shape.

When this abnormal PRPSC encounters a normal PRPC molecule, it somehow acts as a template or catalyst, causing the normal protein to refold into the abnormal shape.

So it converts the good ones into bad ones.

Exactly.

It's a chain reaction.

The abnormal PRPSC accumulates, forms, aggregates clumps in the brain, leading to neuron death and the characteristic sponge -like holes.

It can be infectious, transmitted, or arise from inherited mutations in the PRK gene that make the protein more likely to misfold spontaneously.

Truly bizarre and fascinating.

Now, from rogue proteins to fixing faulty genes,

human gene therapy.

This feels like the ultimate goal for many genetic diseases.

It's the dream for many, yes.

Gene therapy is basically introducing functional genes into a patient's cells to treat disease or modifying existing faulty genes.

The potential is enormous thousands of single gene disorders, maybe even cancer, AIDS.

And we're seeing actual approved therapies now.

We are.

The FDA has approved some, for instance, for certain leukemias and lymphomas, often involving genetically modifying the patient's own immune cells.

How do you get the good genes into the cells?

Two main strategies.

Non -viral methods use things like liposomes, fatty bubbles, to carry the DNA into cells.

Less efficient sometimes, but they don't trigger an immune response.

And the other way, viruses.

Right, using viruses as delivery vehicles, retroviruses, adenoviruses.

They're engineered so they can't replicate and cause disease, but they're very good at getting genes into cells.

That's the main advantage efficiency.

But viruses can cause problems, right?

Immune reactions.

That's the big challenge.

The body can mount a strong immune response against the viral vector itself.

With early adenoviruses, this was sometimes severe, even fatal.

So there's a constant search for safer, less immunogenic vectors or strategies to manage the immune response.

Let's talk about the first real human gene therapy trial for ADA deficiency, a type of SCID.

Right.

ADA SCID.

Severe combined immunodeficiency due to adenosine -demonase deficiency.

Without this enzyme, toxic metabolites build up and destroy T and B lymphocytes.

Kids have virtually no immune system, usually fatal very early.

A devastating disease.

What'd they do in that first trial?

It was an ex vivo approach, meaning outside the body.

They took T cells from the patients, grew them in the lab, used a modified retrovirus to insert a working copy of the ADA gene into these T cells, and then infused the genetically corrected cells back into the patient.

Did it work?

The patients definitely showed increased ADA enzyme levels, which was promising, but they were also receiving a background treatment, an enzyme replacement therapy called PEG -ADA.

So it was hard initially to tease out exactly how much benefit came purely from the gene corrected cells versus the background therapy.

Okay, but there was another SCID trial, SCID X1, that had clearer initial benefits, but then a serious complication.

Yes, this is a really critical part of the story.

The SCID X1 trial, targeting a different gene defect, also causing severe immunodeficiency, used a similar retroviral vector approach.

Initially, it looked like a remarkable success.

Many boys had their immune systems restored, T cell counts normalized, a huge breakthrough.

But then came the devastating news.

In one of the French trials several years later, some of the treated children developed leukemia.

Leukemia, why?

It turned out the retrovirus, when it inserted the therapeutic gene into the patient's DNA, had unfortunately landed near a gene involved in regulating cell growth, an oncogene.

This insertion accidentally activated the oncogene, leading to uncontrolled growth of T cells leukemia.

Oh no.

It was a major setback.

Trials were halted.

It highlighted the very real risk of insertional mutagenesis with these early retroviral vectors.

It underscored the immense complexity and the need for even safer delivery systems.

A sobering lesson about balancing hope and risk.

A really crucial lesson.

So all this progress, the gene finding, the therapy attempts, it's all converging towards personalized medicine.

What does that really mean for us, for the listener?

It means moving away from treating diseases based just on symptoms or average responses.

It's about using your specific genetic information, along with other clinical data, to tailor treatments, medications, even preventative strategies, just for you.

And this is already happening in cancer care with molecular profiling.

Absolutely.

Molecular profiling looks inside the tumor cells, identifying the specific genetic mutations driving that cancer.

It goes way beyond just how the tumor looks under a microscope.

DNA microarrays, for example, can show which genes are overactive or underactive in your tumor compared to normal cells.

How does that help treatment?

Take diffuse large B -cell lymphoma, DLBCL.

Molecular profiling showed there were different subtypes based on gene expression patterns.

And crucially, patients with different subtypes had very different survival rates with standard therapy.

This helps doctors choose more aggressive or different treatments for those with poor prognoses based on their tumor's specific profile.

So it guides therapy choice.

Exactly.

And it drives drug development, too.

We now have targeted therapies drugs designed to hit the specific abnormal proteins made by cancer -causing gene mutations.

Think imatinib for CML or tamoxifen for certain breast cancers.

Precision targeting.

And it's not just cancer.

Tailoring drug dosages based on genetics, pharmacogenetics.

Yes.

Pharmacogenetics studies how your genes affect your response to drugs.

It influences how you absorb, transport, metabolize, and interact with medications.

Liver enzymes, like the cytochrome P450 family, are huge players here.

You mentioned warfarin, the blood thinner, as an example.

Perfect example.

Warfarin is tricky to dose correctly.

Too much, you bleed.

Too little, you clot.

It's metabolized mainly by an enzyme called CYP2C9.

And people have different versions of the CYP2C9 gene.

Right.

There are common variants that make the enzyme work faster or slower.

So you can be an ultra -rapid, extensive, intermediate, or poor metabolizer of warfarin based on your CYP2C9 genotype.

So a genetic test can tell you which type you are.

Exactly.

And that helps doctors choose a much better starting dose for you, reducing the risks of getting it wrong early on.

It's a clear win for personalized medicine, improving safety and effectiveness based on individual genetics.

So the big picture, the grand vision, is that getting your whole genome sequenced will just become routine.

That's certainly the direction things are heading.

As sequencing gets cheaper and faster, having your full genome as part of your medical record could become standard.

This would make pharmacogenetics and personalized approaches central to everyday health care.

A truly individualized approach to staying healthy and treating illness.

What an incredible journey we've taken.

From basic inheritance to gene hunts, therapies, and this personalized future, it really hammers home how understanding our genes is just fundamentally changing medicine.

It really is.

Diagnosis, treatment, even prevention genetics is becoming core to all of it.

Which leads us to a final thought for you, our listeners.

As this personalized medicine era unfolds, as we learn more and more about our own genetic makeup,

what are the rules we need?

Yeah, what ethical guidelines do we need as a society for accessing this information, for preventing discrimination, for the choices we make about health based on our genes?

It's not just a scientific question anymore, is it?

Not at all.

It's a question for all of us.

Thank you so much for joining us on this Deep Dive.

We hope exploring medical genetics has been insightful and maybe even sparked some new questions for you.

Thanks for being part of the Deep Dive family.