April 25, 2001 11 a.m.
James Garbern MD, PhD
Textbook: Sack Chapter 6
Objectives for this lecture:
At the end of this lecture, students will:
1. become familiar with autosomal recessive patterns of disease transmission.
2. understand the roles of ethnicity and consanguinity in disease prevalence in
3. gain an appreciation of some of the more common recessive disorders.
In classical genetics, autosomal recessive traits are those in which the heterozygous
state (i.e. having one mutant allele at a locus) is indistinguishable from normal, and auto-
somal dominant traits are conditions where homozygous (i.e. identical alleles at a locus)
individuals are indistinguishable from heterozygotes. In practice, however, there may be
subtle distinctions noted in heterozygous carriers of a disorder conventionally regarded as
recessive. For example, although sickle cell anemia (i.e. the disease) is considered a re-
cessive disorder, individuals heterozygous for Hb S (the abbreviation for sickle hemoglo-
bin), are more resistant to malaria. In addition, sickle cell trait (i.e. the presence of sickle
hemoglobin in any amount) is dominant. Individuals heterozygous for the ataxia-
telangiectasia gene, ATM (see below), appear superficially normal with respect to cere-
bellar function and vascularity, but have a significantly higher risk of developing malig-
nancies. In fact perhaps 10% of breast cancers occur in ATM heterozygotes. To make
matters more confusing, there are genes where different mutations affecting the same
gene can result in either a dominant or recessive phenotype, as in the case of osteogenesis
imperfecta. Therefore, classification of a mutation as dominant or recessive must be done
on an allele (gene variant) basis, rather than a gene basis.
Recessive disorders are generally due to the complete or near total loss of functional
gene product. The great majority of disorders due to metabolic enzyme deficiencies are
autosomal recessive, and many of the remaining are X-linked recessive.
Most recessive traits and disorders begin during childhood. The typical presentation
of an individual with an autosomal recessive disorder is a child where both parents are
clinically asymptomatic and usually unaware of any familial disorder in their families.
Males and females are affected equally. This is diagrammed in the figure below. Individ-
ual 12 (by convention, siblings should be listed from left to right in birth order), is the
first known member of either family with a hypothetical recessive disease. Of course if
the child has another affected sibling, as would be the case with 15, the youngest brother
of 12, then the likelihood of a genetic condition should be immediately apparent. So, a
horizontal grouping of affected individuals in a family strongly suggests a recessive dis-
1 2 3 4
5 6 7 8 9
III 3 6
11 12 13 14 15 16
17 18 19
Pedigree of an autosomal recessive trait. Affected individuals are indicated by solid symbols, heterozygotes
are indicated by half-filled symbols. Note that all children of an affected homozygote are obligate carriers.
The following cartoon diagrams the possible genetic outcomes for children of two
heterozygous parents and also demonstrates the 3:1 ratio of normal:affected that would be
expected. Each of the four possibilities is equally likely for each pregnancy. There is a 1/4
chance for each child inheriting both abnormal alleles, a 1/2 chance of being heterozyg-
ous, and a 1/4 chance of being completely normal.
Carr ier Fat he r Carr ier Mo t her
Pa ren ts
Gam et es
Chi ld ren
Af fect ed No rmal No rmal No rmal
ho mo zy go us he te ro zyg o us he te ro zyg o us ho mo zy go us
Carr ier Carr ier
Another way of demonstrating these results is using a so-called Punnett square. If
D is the shorthand abbreviation for the normal gene, and d the symbol for the recessive
allele at this locus, then there are four possible combinations of parental gametes:
Thus, the criteria for an autosomal recessive
trait or disorder are:
1. Affected individuals in a family usually are seen
only within a sibship, not in their parents, offspring
or other relatives, i.e. horizontal rather than vertical
clustering of the condition in the family pedigree.
2. The risk to each sib of an affected individual of
showing the phenotype is 25%.
3. Consanguinity significantly increases the risk of
manifesting a recessive phenotype.
4. Males and females are equally likely to be affected.
Factors that influence the occurrence of autosomal recessive conditions
Because the occurrence of a recessive autosomal disorder depends upon two oc-
currences of bad luck, the frequency in a population of a given disorder is the square of
the carrier frequency, i.e. the frequency of heterozygotes in the population. Conversely,
the carrier frequency is calculated using an important concept from population genetics,
the Hardy-Weinberg equilibrium, to be covered in more detail in a later lecture. Because
the frequency of most recessive disorders is quite low and most American families are
small, most affected individuals will appear to be sporadic cases, with only a single case
within the family. The ascertainment bias in detecting recessive traits and disorders can
be diagrammed also. In this example, for two heterozygous parents, all 64 (43) possible
outcomes for three children are depicted on the next page.
42% of 3 child families would have no affected children, 42% would have a sin-
gle affected child, 14% would have two affected children and 2% would have all 3 child-
ren affected. Even among the families where there is at least one affected child, almost
(37 or 73 %) , would have only a single affected individual. Therefore, only a high level of
suspicion and a knowledge of genetics will enable the proper diagnosis to be made.
The formula from which the following figured is derived, for the morbidly curious, is:
n! s t
Probability = pq
n=number of children
s=number of normal children in family
t=number of affected children
p=probability of being affected
q=probability of being normal
Example: 1 affected, 2 normal children:
3! 1 3 6 1 9 27
Probability= = =
2!1! 4 4 2 4 16 64
The Hardy-Weinberg equation
Knowing just the prevalence of a rare autosomal recessive condition in a stable
population (that is not a lot of new immigrants or emigrants), it is possible to estimate the
frequency of the carrier state in the population using the Hardy-Weinberg equation.
p and q are defined as the frequencies in the population of two alleles in the ex-
ample shown in the figure (taken from Sack, page 138), where p is the frequency of the
mutant recessive allele and q the frequency of the normal allele. Note that the frequency
of the disease in the population is p2, since affected individuals have to have two copies
of the mutant allele, while q2 is the frequency of homozygous normal individuals. Clini-
cally, however, the homozygous normal people cannot be distinguished from heterozyg-
ous individuals (whose frequency is the carrier frequency and is = 2pq). The important
result is that the carrier frequency in a population can be quite significant, even though
the disease prevalence is low. Since there are thousands of recessive diseases that are
known, and quite likely many more to be discovered, it can be estimated that all of us are
likely to be carriers for several recessive, and potentially lethal, disease genes.
Hardy-Weinberg relationship for a
genetically stable population and
considering two alleles. In this ex-
ample for a rare recessive condition,
the carrier frequency can be calcu-
lated just knowing the prevalence of
the recessive condition. If the ho-
mozygous condition is present in 1
in 10,000 then p = 1/100. With this
information and the two equations,
it is possible to calculate q. The car-
rier frequency is 2pq.
Ethnicity and consanguinity
Because of the low carrier frequency of most recessive disorders among the gen-
eral population and since people as a rule do not mate with random members of the gen-
eral population, it is especially important to recognize that certain genetic conditions
occur with much greater frequency in certain ethnic or geographic populations. For ex-
ample, Tay-Sachs disease is much more common in people of Ashkenazi (Central and
Eastern European) Jewish, cystic fibrosis is more common among those of Northern Eu-
ropean descent and sickle cell anemia is more common among those of African descent.
The observed frequencies of these disorders enables the calculations that among Ameri-
cans of Northern European descent, about 4.5% carry the cystic fibrosis mutation, among
Ashkenazi Jews, about 3% are Tay Sachs carriers and among African Americans, about
8% are sickle cell carriers.
Another major influence on the prevalence of recessive disorders in a population is
the degree of consanguinity. In part B of the figure below, it should be readily apparent
that the risk of being heterozygous for an allele is higher if the parents are related, as is
the case for first degree cousins 12 and 13. The prevalence of consanguinity has dimi-
nished greatly in the US during the past 50 years, but there are still some relatively iso-
lated communities, for example the Amish of Pennsylvania, where the degree of
consanguinity is higher, and ordinarily rare diseases are much more common. Worldwide,
there are regions where the degree of consanguinity is quite high. For example, almost
50% of marriages among Arabic Israelis have been shown to be consanguineous, usually
cousin-cousin marriages. The coefficient of relationship (r) is the proportion between two
relatives of genes identical by descent from a common ancestor. First degree relatives,
e.g. parent and child or siblings have a coefficient of relationship of 0.5 (i.e. half of their
genes are identical), second degree relatives, e.g. aunt - nephew, have a coefficient of re-
lationship of 0.25, third degree relatives, e.g. first cousins have a coefficient of 0.125, and
fourth degree relatives, e.g. second cousins, have a coefficient of 0.0625. The coefficient
of inbreeding (F) is the proportion of loci at which one individual is homozygous by des-
cent, and F=r/2.
Although it occurs infrequently, but is much more likely when there is consanguini-
ty, mating between an affected homozygote and a heterozygote results in a frequency of
affected children that can superficially appear to be autosomal dominant, i.e. 50% of the
children will be affected. This pattern is referred to as pseudodominant or quasi-
If an individual has one
I mutant allele on one chromo-
some and a different mutant al-
5 6 3 4 7 8 lele on the other chromosome at
the same locus, that individual is
14 13 12 9 10 11 termed a compound heterozy-
IV gote. Since both alleles are ab-
15 16 17 18 19 20 21 22
normal, there is a high likelihood
Pedigree of family with an autosomal recessive gene. Individuals 12 and
13 are first cousins, and have one eighth of their genes in common that they will
(0.125 not be able to com-
pensate for or complement each
other, and the individual will be affected. In the absence of consanguinity, many apparent
recessive conditions will turn out to be compound heterozygotes.
Pseudo-dominant inheritance with mating between a homozgote and heterozygote for a recessive trait
From the mechanistic standpoint, the autosomal recessive disorders are the easiest
to understand in terms of disease pathogenesis. Generally speaking, these conditions arise
through mutations that prevent the expression of a functional gene product. Many of the
recessive genes encode enzymes, which because they have catalytic activity and are
usually produced in excess of the amount required for normal function, loss of one copy
of a gene does not cause disease. When both copies of an essential gene are mutated there
is very little to no functional gene activity and disease is the consequence.
Examples of some of the more common recessive disorders seen in the States fol-
low. I have also tried to present diseases where a range of mutation types have been dem-
onstrated, showing the diversity of possible genetic mechanisms that are known, and to
indicate the relative utility of molecular diagnostic testing for these disorders.
Cystic fibrosis (CF, OMIM 219700)
Cystic fibrosis is the most common autosomal recessive disorder among Cauca-
sians, affecting about 1/2500 newborns. Therefore it can be calculated that the carrier fre-
quency is about 4%. Pancreatic insufficiency and severe pulmonary obstruction due to
bronchiolar secretions that are thick and viscous characterize the disease. At birth, an af-
fected neonate may even develop intestinal obstruction from the thick intestinal secre-
tions. Death usually develops by age 25 from severe pulmonary complications.
Demonstrating an elevated chloride concentration in the sweat can make the diagnosis.
In 1989, two groups working independently identified the gene that is affected in
CF. The gene resides on chromosome 7 (band q31) and is a very large one, composed of
about 250,000 bases, with 26 exons that encodes a protein of 1480 amino acids. Initial
structural analysis and subsequent experimentation has demonstrated that the protein is,
not unexpectedly, a chloride transporting membrane-associated pump. The following fig-
ure depicts the gene, its transcript and protein product:
Genetic studies of affected patients have demonstrated that about 70% of CF chro-
mosomes have a mutation that deletes three nucleotides, corresponding to amino acid re-
sidue 508, and is abbreviated del508 or 508. At least 500 other mutations account for the
remaining 30%, precluding an easy and sensitive diagnostic test, although screening for
about 30 of the more common mutations has a success rate of over 90% among Cauca-
sians, and about 50% among African Americans. While detection of the 508 mutation is
diagnostic of a CF mutation, it may not be present on both chromosomes, or be present at
all in a CF patient. It must be found on both chromosomes to be of reliability for prenatal
counseling purposes (for example, for expectant parents who have had a prior CF child).
In a CF patient without homozygous 508 mutations, screening for the mutation(s) can be
The numerous mutations in the CFTR gene and their types indicated by a ‘tic’ mark along
the 24 exon containing gene.
Males who are compound heterozygotes, with one severe CFTR allele on one chro-
mosome and one of certain mild CFTR alleles on the other chromosome, can have con-
genital bilateral absence of the vas deferens (CBAVD, OMIM 219700). These males are
infertile, but may conceive using sperm aspirated from the epididymis. Otherwise, these
individuals are healthy and do not have the other features of CF. However, they are hete-
rozygous for a severe allele of CF and should be counseled appropriately when they are
planning to have children.
The major goal in finding the CF gene was to develop more effective treatment for
this disease. Since one generally cannot fix something unless one knows what’s broken,
identifying the protein affected by CF was a critical step, but just an early step towards
attaining this goal. Animal models of CF have been made, using a technique called homo-
logous recombination, and these animals respond to gene therapy: use of recombinant vi-
ruses engineered to provide the missing CF gene. Clinical trials with humans are contem-
plated in the near future for this fatal disease.
Sickle cell anemia (OMIM 603903)
The first human genetic disease whose molecular basis was demonstrated was sickle
cell anemia. Sickle cell anemia is an autosomal recessive disease that affects about 1/500
African American newborns. The fundamental problem is an hypoxia induced conforma-
tional change in the -globin chain in red blood cells, resulting in the characteristic bana-
na or sickle shape deformity in the red blood cells seen microscopically. The cell shape
change is due to the formation of hemoglobin fibers that form from the abnormal packing
of hemoglobin molecules. The RBC deformity results in clogging of vascular capillaries,
causing sludging of blood and hypoxia, or
even infarcts of organs and tissues of the
body. Symptomatically, the disease is charac-
terized by attacks of pain in many parts of the
body, especially the bones, hematuria, and
neurological symptoms when strokes or cere-
bral ischemia occur. Linus Pauling, the
double Nobel prize laureate, deduced that
there was a mutation which created an amino
acid substitution in the protein, and which
was confirmed by Vernon Ingraham. The -
globin gene on chromosome 11 has a point
mutation at codon 6 where the usual A is re-
placed by a T, causing a missense codon, in
turn resulting in the normal glutamate being
replaced by valine in HbS. This mutation can
be abbreviated Glu6Val, or E6V, in single letter nomenclature.
The extremely high prevalence of sickle cell trait (that is the heterozygous state) ap-
pears to be due to the resistance of RBCs of a heterozygote to malaria. This, amazingly
enough, was experimentally demonstrated by inoculating sickle cell carriers with malaria.
Many other mutations in the hemoglobins have been characterized since the elucida-
tion of sickle cell anemia, and which cause much less common anemia syndromes.
Thalassemia (OMIM 141900, beta; OMIM 141800, alpha)
The thalassemias constitute a related group of hemoglobinopathies characterized not
only by mutations that cause an abnormal protein to be made, but which affect the amount
of protein that is made. The -thalassemias are characterized by insufficiency of -globin
in RBCs, whereas the -thalassemias are characterized by insufficiency of -globin. Inte-
restingly, the thalassemias are most prevalent in the same regions of the world where ma-
laria is endemic, leading to speculation that this as well as sickle cell anemia and G6PD
deficiency provide some resistance to malaria.
There are actually 2 -globin chains on each chromosome 16, so that mutation of one
of the complement of 4 does not usually result in anemia. However, when 2 of the 4 -
globin genes is defective, the individual has -thalassemia trait. This can occur in one of
two ways: 1) both -globin genes on one chromosome are defective, or 2) one -globin
gene is defective on each chromosome. Situation 1 is more prevalent in Southeast Asia,
whereas the second is more common among those of African descent. When 3 of the 4 -
globin genes are defective, then a situation arises where there is a 2:1 ratio of -globin to
-globin chains, leading to detectable levels of 4 globin, or Hb H, which causes signifi-
cant but not lethal anemia. When all 4 -globin genes are defective, then 4 globin or he-
moglobin Bart’s predominates, and the clinical syndrome of hydrops fetalis results. This
is a fatal condition in the neonatal or even fetal period, and the name describes the tre-
mendous edema that develops in the oxygen-starved tissues. At the DNA level, most -
thalassemias are the result of one or more -globin gene deletions. Fetal transfusion of
bone marrow cells could treat this condition if recognized early enough.
The range of mutations that has been described causing -thalassemia is quite stagger-
ing and interesting, not only from the clinical standpoint, but also from the molecular bio-
logical standpoint as well. Virtually all conceivable derangements to a gene that can be
imagined have been demonstrated.
There is only one -globin gene on each chromosome 11. Mutations have been found
that affect the regulation of gene expression by altering critical sequences in non-
transcribed regulatory portions of the gene, including very distant far upstream or 5’ mu-
tations of a critical regulatory region that appears to act as a master switch for the entire
-globin gene complex. Other mutations affect critical sequences that perturb the normal
splicing patterns of the -globin gene exons, including one that creates a new splice ac-
ceptor site within an intron, and is the most common cause of -thalassemia in the Medi-
terranean region. Another mutation alters the sequence AATAAA, which is the signal for
the polyadenylation of mRNA, and results in improper processing of the RNA. In addi-
tion, there are mutations within the coding region of the -globin genes as well. This tre-
mendous variation in location and type of mutation precludes easy screening for
mutations; fortunately, it is relatively easy to examine the globin proteins directly to es-
tablish the diagnosis.
The clinical term for a heterozygous mutation affecting -globin expression is thalas-
semia minor and is asymptomatic. Homozygotes,( or as often happens with rare recessive
disorders when consanguinity is not a factor), compound heterozygotes (where the patient
carries two different defective alleles) for -globin gene mutations have thalassemia ma-
jor or Cooley’s anemia. Because fetal hemoglobin persists during the first year of life, ba-
bies with thalassemia major are usually normal, but as the switch from fetal to adult
hemoglobin is attempted, severe anemia becomes apparent. Massive hepatosplenomegaly,
marrow space enlargement, with consequent bone thinning, develop in an attempt to pro-
duce hemoglobin. Transfusions, coupled with iron chelating therapy prolong life.
Ironically, although the globin gene was the first eukaryotic gene cloned, with a major
goal of developing gene therapy. There is still no immediate prospect for gene therapy on
the horizon. This became apparent as the complex details of gene regulation became un-
raveled, and pointed out the importance of not just simple gene replacement, but getting
the replacement gene to the right cells at the right time, and in a form that can produce the
proper levels of gene expression in the intended target cells.
Metachromatic Leukodystrophy (OMIM 250100)
Although rare, the genetics of this disorder illustrates some important concepts in
clinical genetics. This autosomal recessive disorder is due most often to a defect in the
arylsulfatase A gene on chromosome 22q13, which encodes a lysosomal enzyme that de-
grades a specific class of lipids, predominantly in the cerebral white matter. Although
they develop normally after birth, children with the classic form of this disease develop
progressive spasticity, behavioral difficulties, weakness, loss of cognitive functions, and
seizures, before dying by around age 5. MRI scans show a characteristic symmetric pat-
tern of abnormal cerebral white matter. Because of the enzyme deficiency, cerebroside
sulfates build up in the brain and other tissues. This material stains aberrantly, hence the
The classic infantile form of disease is caused by different small mutations that cause
near to complete loss of enzyme activity. However, as is the case with several other reces-
sive disorders, there is allelic heterogeneity, where some alleles are milder, both from
the clinical and biochemical standpoints. In general, mutations that do not completely in-
activate the enzyme result in later onset symptoms and less rapid progression of signs.
There are now recognized juvenile and even adult onset forms of MLD which have pro-
portionately greater residual arylsulfatase A activity. As for other diseases that affect the
gene for a protein whose function is well characterized, it is more practical to screen for
the disease by direct enzyme testing, initially with blood, but it may be necessary to study
the enzyme more carefully in fibroblasts cultured from a small skin biopsy.
An interesting recessive condition has been recognized that clinically and pathologi-
cally is identical to juvenile onset MLD, but which on direct testing for arylsulfatase A,
shows normal enzyme levels. However, degradation of cerebroside sulfate in fibroblasts
cultured from patients is severely attenuated. The answer was that an essential protein
cofactor, prosaposin, is also necessary for degradation of the lipid. Patients with this auto-
somal recessive MLD phenocopy have a mutation in a gene called saposin. This then is
an example of locus heterogeneity, the production of similar phenotypes by mutations at
different genes or loci.
Further complicating the diagnosis of MLD is the recognition that there are mutations
that affect the arylsulfatase A gene that result in enzyme with about half the normal activ-
ity in the test tube, but which are not associated with disease in homozygotes. This is then
a pseudodeficiency, in part attributable to laboratory technique, but can make identifica-
tion of carriers difficult, since the frequency of the pseudodeficiency allele is about 7% in
the general population.
Finally, MLD is an example of a previously untreatable genetic disorder that can be
treated currently. Because the microglia are a major cell type in the brain for the normal
breakdown of sulfatides, and they derive at least in part from circulating macrophages,
bone marrow transplant can replace enzymatically deficient microglia. This has been
shown to result not only in halting but also reversing disease progression in patients.
Ataxia-Telangiectasia (OMIM 208900)
This condition affects about 1/40,000 and has an estimated carrier frequency of 1% in
the US. The condition is classically a childhood onset disorder of ataxia or clumsiness
associated with later mental decline, oculocutaneous telangiectases, and immunodeficien-
cy, which results in severe infections, and often premature death. Heterozygous carriers,
although asymptomatic, have an approximately 5 fold increased risk of breast cancer and
other malignancies, and are extremely radiation sensitive.
Recently, the gene on chromosome 11 (band q22) affected in ataxia-telangiectasia,
ATM, was discovered. It has sequence properties that suggest it is a member of the regu-
latory protein kinase family. One potential target of its regulation could be the important
p53 gene, which is mutated in many cancers, and acts as an anti-oncogene. It also appears
to be an important regulator of proteins involved in cell cycle control.
Friedreich ataxia (229300)
Friedreich ataxia is one of the more common recessive disorders of adolescents and
children. It accounts for about 50% of hereditary ataxia (incoordination) in the US and
Europe. The disorder generally is apparent by childhood, but it is now clear that milder
forms, with onset well into adulthood, also exist.
The classic clinical findings are:
1. Progressive limb and gait ataxia before age 25.
2. Autosomal recessive inheritance.
3. Absent tendon reflexes in the legs with Babinski signs, often with pes cavus.
4. Axonal sensory neuropathy by electrophysiologic testing.
Dysarthria, weakness, scoliosis, loss of proprioception and incoordination of eye
movements are very common. Significant complications include hypertrophic cardi-
omyopathy, cardiac arrhythmias and diabetes. Mild deafness and optic atrophy can occur.
More than 95% are chairbound by age 45 and death from cardiac disease has been re-
ported in the past by age 50.
The gene responsible for Friedreich ataxia is located on chromosome 9 and is called
frataxin. The mutation is a trinucleotide expansion that normally consists of 7 to 20 GAA
triplets lying in the first intron of the gene. Affected individuals have 200 to about 1000
repeats. There is a very rough correlation between the number of repeats and severity of
disease, including the presence of disease complications. Thus far, this is the only disease
due to triplet repeat expansion that has recessive inheritance.
Interestingly, the frataxin protein is a mitochondrial protein that has been shown to
transport iron out of mitochondria. Therefore deficiency of this protein results in accumu-
lation of iron. Excessive mitochondrial iron promotes the formation of toxic oxygen radi-
cals, leading to eventual mitochondrial dysfunction and deterioration, and which accounts
for the defective oxidative phosphorylation recently found in FA patients. The recognition
that frataxin is a mitochondrial protein explains the previously documented impairment of
energy utilization in patients, and the spectrum of clinical findings, which is quite remi-
niscent of those seen in classic mitochondrial disorders. Preliminary trials of an iron che-
lator have suggested some benefit, at least in delaying one of the most serious
complications of FA, cardiomyopathy.