Two discoveries that will shape the 21st century Deciphering the - - PowerPoint PPT Presentation

Two discoveries that will shape the 21st century

Deciphering the Human Genome Lifebook.dat – the database required to make a human being Production of Human Embryonic Stem Cell Lines Formlife.exe – the executable program to make a human being Cloning of Human Beings (a negative utopia) Copy.exe – a possibility of pertuating ourselves?

Cell Nucleus Chromosom Gene Gene-product: Protein DNA

Blueprint of Life

DiakoniePublik 3/2001

Informationsübertragung bei der zellteilung

Cell division: Transfer of Information!

C in phosphate ester chain C and N in bases

Base pairs

Sugar phosphate backbone

Minor groove Major groove H O P

5‘ 3‘

DNA as carrier of Information

Human ß-Globin, Segment

... TAAGCCAGTG CCAGAAGAGC CAAGGACAGG TACGGCTGTC ATCACTTAGA CCTCACCCTG TGGAGCCACA CCCTAGGGTT GGCCAATCTA CTCCCAGGAG CAGGGAGGGC AGGAGCCAGG GCTGGGCATA AAAGTCAGGG CAGAGCCATC TATTGCTTAC ATTTGCTTCT GACACAACTG TGTTCACTAG CAACCTCAAA CAGACACCAT GGTGCACCTG ACTCCTGAGG AGAAGTCTGC CGTTACTGCC CTGTGGGGCA AGGTGAACGT GGATGAAGTT GGTGGTGAGG CCCTGGGCAG GTTGGTATCA AGGTTACAAG ACAGGTTTAA GGAGACCAAT AGAAACTGGG CATGTGGAGA CAGAGAAGAC TCTTGGGTTT CTGATAGGCA CTGACTCTCT CTGCCTATTG GTCTATTTTC CCACCCTTAG GCTGCTGGTG GTCTACCCTT GGACCCAGAG GTTCTTTGAG TCCTTTGGGG ATCTGTCCAC TCCTGATGCT GTTATGGGCA ACCCTAAGGT GAAGGCTCAT GGCAAGAAAG ...

Human ß-Globin, Exon1, Segment

...GTG CAC CTG ACT CCT GAG GAG Val His Leu Thr Pro Glu Glu AAG TCT GCC GTT ACT GCC CTG Lys Ser Ala Val Thr Ala Leu TGG GGC AAG GTG AAC GTG ... Trp Gly Lys Val Asn Val + 126 further AS ! TGA Stop !!!

Genomic Library of Mankind

• 46 Chromosomes 2 * 3,2 billion letters
• 30 000 – 40 000 Genes
• ca. 99 % not protein-coding (excess of void information)
• Man/Chimpanzee 1-2% global text difference

(ca. 120 Mio Letters)

i.e. in every line of the lifebook about 1-2 „misprints“

• enormous repetitive segments
• Retroviral traces (hundreds of thousand items) –

the human genome is a museum of virus infections !

Genomic Library of Individual Person

• ca. 2 Mio differences (SNPs)

(between non-related persons)

• ca. 60 000 of them in coding regions
• ca. 10 000 genetic defects

(each individual carries disposition for about 5 defects)

• every 500 -2000 letters a variation
• (i.e. on every page of the life book a few „misprints“

Evolutionary Traces in the Genome

• 25 % of the human genome are „deserts"
• ca. 50 % are repetitions
• among them ca. 45 % „jumping copies",

(silent since millions of years)

Genomic Non-sense

but important identification tag!

Person 1 : CA CA CA CA CA CA CA 7 repeats no!! (Father ?) CA CA CA CA 4 repeats Person 2: CA CA CA CA CA CA CA CA 8 repeats (Father ?) CA CA CA 3 repeats Person 3: CA CA CA CA CA CA CA CA 8 repeats (Father ?) CA CA 2 repeats !! Person 4: CA CA CA CA CA CA CA CA 8 repeats (Mother) CA CA CA CA CA CA CA 7 repeats !! Person 5: CA CA CA CA CA CA CA 7 repeats (Child) CA CA 2 repeats

Genetic prediction: Scope

It can establish a diagnosis It can predict the future It provides implied information on related individuals

Genetic Diagnosis and Prediction: Scale

„Within the next decade genetic testing will be used widely for predictive testing in healthy people and for diagnosis and management of patients.“ Bell J (1998) „New Genetics in Clinical Practice“, Brit. Med. J. 316, 618-620

Genetic Prediction: Upper Limit of Determination

Concordance of traits in monozygotic twins: Rare mendelian diseases: up to 100% Frequent complex diseases: 30 – 70% Relevant prediction is in terms of probability rather than of certainty (maybe useful for the insurance company, but of limited use for the individual) There is always considerable non-genetic variability!

Genetic Diagnosis and Prediction: Tests of What ?

Non-inherited genetic traits: Chromosomal Aneuploidy e.g. Down syndroma, Klinefelter syndroma Mendelian diseases: about 1500 out of 5000 may be diagnosed dominant mode of inheritance: in every generation of family recessive mode of inheritance: in one family “out of the blue“ Complex (multifactorial) diseases: about 40 genes may contribute (example of cardiovascular disease)

Genetic Diagnosis and Prediction: Tests on Whom ?

• Partners
• Embryo in vitro
• Embryo in utero
• Fetus before birth
• Newborn
• Child
• Adult

Purposes (with example): Prediction of disorder: Huntington Selection between alternatives: immune Prevention of disease: PKU

Examples of Mendelian Diseases

Huntington (D) BRCA (breast cancer, D) Cystic fibrosis Thrombophilia Porphyria Haemochromatosis Myotonic dystrophy Duchenne muscular dystrophy (sex-linked) Phenylketonuria Galactosemia Thalassemia Congenital hypothyrioidism

Examples of Complex Diseases with Genetic Contribution

Diabetes type 1 Diabetes type 2 Breat cancer Colon cancer Prostate cancer Alzheimer´s dementia (early onset form) Multiple sclerosis Bipolar disorder Schizophrenia Autism Familial Parkinson disease

Risk prediction of Complex Disease:GRR

Genotype Relative Risk (GRR) = Frequency of disease in carriers of variant allele Frequency of disease in carriers of normal allele GRR > 50 single gene disorder with high penetrance 4-50 oligogenic disorder <4 polygenic factors (complex disease)

Risk prediction of Complex Disease: Bad Test

Not even a high GRR guarantees a good test if Frequency (of risk allele) > frequency (disease in population)

HLA-B27

• ther allele

Sum Healthy persons 985 8 898 9 983 Spondylitis ankylosans (Mb. Bekhterev) 15 2 17 All persons 1 000 9 000 10 000 Frequency of disease: 17 / 10 000 = 0.17% Frequency of gene variant: 1000 / 10 000 = 10% GRR = 15/17 divided by 985/9983 = 8.9 Frequency of gene variant in disease: 15/17 = 88% Frequency of gene variant in healthy: 9845 / 9983 = 10% Test prediction on risky gene : 15/1000 diseased persons = 1.5% Rate of false positives: 98.5% Rate of false negatives: 0.02%

Is genetic prediction of disease possible?

Monogenic case , rare, high GRR, high penetrance: yes, if genotype is moderately specific Polygenic case and multifactorial causation, low GRR:

• nly statistical prediction in large populations samples

Population screening instead of individual diagnosis:

• Screening for heterozygotes for recessive disorder
• Screening of fetuses or newborns for necessary therapy
• Screening for predictive genotype in frequent disease (breast

cancer)

• Any screening runs the following risks:

False positives, if genotype is not the specific causative factor False negatives, if genotype is not the only causative factor

HDL2

IDL

LDL

Chylo

Chylo remnants

VLDL

HDL3

LDLRec CETP/HL CETP/HL

scavenger SRB1

LCAT (LPL)

LPL LPL

liver

intestine

• periph. cholesterol

HL

disc

LCAT

nasc

surface remnants surface remnants HDLrec cubulin

LRP

HL

scavenger

(on LPL def.) (LRP&HSPG)

HL

LDLRec

• 15/03/01

A C H B 48 A C E A C E A C E A C E E B 100 B 48 B 100 B 100

• periph. cholesterol

Metabolic Network of Lipoproteins

Cholesterol as risk factor

Heritability is 50% - even the best prediction can reduce the variation of individual values by 50% Variation in more than 20 genes are responsible for the normal status

• f cholesterol (clinical LDL/HDL ratio). Each can contribute around 2%

Individual prediction will be useless Statistical prediction for groups will be possible provided that the complexity of genetic causation is not too high Test makes sense only if there are genotypes with normal cholesterol levels in young, but high levels at advanced age

Risk and individual disease

Only part of all individuals with risk status will actually become ill

Every person belongs to several risk groups for a chronic disease with genetic component. Thus in principle everybody should pay a risk surcharge For all persons this will cancel Provided that equity of information is established between insurance and client!