Do They Really Cause Marfan Syndrome?
Marfans with fibrillin-1
without fibrillin-1 mutations
The 1997 study determined that most people with Marfan syndrome did not have fibrillin-1 mutations, yet much of the literature on Marfan syndrome created after this date, and through today, continues to state that fibrillin-1 mutations cause Marfan syndrome. How can a mutation be said to cause a disorder, if it cannot been proven that most most of the people with the disorder even have the mutation? Yet, against all logic, virtually all Marfan research is focused on fibrillin mutation research, and fibrillin research alone.
It has been suggested that perhaps the 1997 study was flawed, that the subjects did not all actually have Marfan syndrome, or that perhaps they had the mutations but that they were unable to be located. Perhaps, but the possibility of a flawed study is a mere speculation, and I have never seen any hard evidence to support either of these suggestions. Other studies have also only found fibrillin-1 mutations in a fraction of the families studied. Even if the 1997 study was found to be flawed, this possibility still would not provide scientific proof that most Marfans have fibrillin-1 mutations, or that the mutations cause the syndrome, had the study not been flawed. I believe that the most obvious conclusion to consider is that this study, and all of the others like it, are valid, and that the hypothesis they set out to test, that fibrillin-1 mutations cause Marfan syndrome, is the conclusion that should be questioned.
Update: May, 2005: I went through the Pubmed database today to see what the latest research has been on Marfan Syndrome since I first wrote this paper in 2000.
Studies continue to prove that not everyone with a fibrillin mutation has Marfan syndrome and that lots of people who do have the mutations do not have Marfan syndrome.
Yet, against all logic, many researchers still cling to the completely illogical premise that fibrillin mutations cause Marfan syndrome when their own studies, year after year after year, fail to support this premise.
Here's a couple of the more recent studies that support my original argument:
Comprehensive molecular screening of the FBN1 gene favors locus homogeneity of classical Marfan syndrome. - "The phenotype of the eight patients without proven FBN1 mutation did not differ from the others with respect to the presence of major cardiac, ocular, and skeletal manifestations or positive familial history." (italics added for emphasis)
Genotype and phenotype analysis of 171 patients referred for molecular study of the fibrillin-1 gene FBN1 because of suspected Marfan syndrome. - "Diagnostic criteria for MFS were fulfilled in 94 patients, 62 (66%) of whom had an FBN1 mutation. ...... Among the 77 patients who did not meet the criteria, an FBN1 mutation was found in 9 patients (12%).
Fibrillin-1 mutations are found in a wide variety of people with some very common conditions. Because of this, I question whether their presence in some people with Marfan syndrome is even a significant event. Fibrillin-1 mutations have been found in people who simply have tall stature. Lots of people are tall. Being tall is not even a diagnostic criteria for Marfan syndrome, according to the Gent nosology. I was told by a geneticist that it was not included in the nosology because it was such a common trait, and was not specific to Marfan syndrome. Why would a gene associated with connective tissue cause tall stature? Just because someone is tall and has a fibrillin mutation does not provide proof that the mutation played any direct part in determining the tall stature. Association does not prove cause and effect.
Fibrillin-1 mutations have also been found
in people who have aortic aneurysms, mitral valve prolapse or skeletal
features of Marfan syndrome, such as scoliosis or pectus excavatum (Robinson,
2000). Many people would either be tall, or have an aneurysm,
or have scoliosis, or pectus excavatum or MVP. Not including blood
relatives or people I met through inherited connective tissue disorder
lists or groups, I know six people with pectus deformities, seven people
with scoliosis, four people with MVP and lots of tall people. And
that doesn't mean more of the people I know do not have these conditions,
these are just the people that for some reason or another told me they
had these medical problems. If I took an actual survey, I'm sure
lots of my friends and acquaintances may easily have at least one of these
conditions, though none of them have Marfan syndrome.
While I could not find any studies that demonstrated the rate of fibrillin-1 mutations in the general public, these facts all imply to me that a big chunk of the general population may well have fibrillin-1 mutations. Yet Marfan syndrome is relatively rare, occurring in only 1 out of 10,000 people, or only 0.01% of the population. If you put all of these facts together, it means that the vast majority of the people with fibrillin-1 mutations would not be expected to have Marfan syndrome.
So what percent of the general population
actually has fibrillin-1 mutations? I asked a researcher this question
and was told this number is unknown. Well, I think it is pretty
important to know this number. What if the rate of fibrillin-1 mutations
in the general population is 28%, just like the rate the 1997 study found
in Marfans? Then there would not even be any statistically significant
association between fibrillin-1 mutations and Marfan syndrome.
Return to top
People who simply have mitral valve prolapse syndrome (MVP) have been found to have fibrillin-1 mutations. MVP is so common it is thought to occur in 3 to 10% of the population, depending upon the study. MVP has been clearly linked, in a large number of studies over the past decade, to magnesium deficiencies, which are also very common. (See my section on MVP syndrome for more on this topic.)
Magnesium supplementation is used to treat MVP. Studies show up to 85% of the people with MVP are deficient in magnesium. Studies show most people do not need gene therapy to treat MVP. They just need to correct magnesium deficiencies. This is not "new age medicine" - this last paper on magnesium appeared in the American Journal of Cardiology. One of my relatives is a Kaiser patient and gets treated for MVP with magnesium therapy, indicating to me that it is considered a part of conventional medical treatment with at least some doctors in the U.S. If fibrillin-1 mutations indeed caused mitral valve prolapse, then magnesium supplementation would have no effect. Yet, studies show it does.
It is curious that of most of the fibrillin studies I have read, the articles say fibrillin-1 mutations cause Marfan syndrome and are associated with a wide variety of other connective tissue disorders such as mitral valve prolapse syndromes. Yet there is no more association between fibrillin-1 mutations and Marfan syndrome than there is between fibrillin-1 mutations and mitral valve prolapse syndromes, so it is illogical to use the term cause for one set of features and associated for another set of features. I suspect that the reason for this is because if researchers report that fibrillin-1 mutations cause mitral valve prolapse syndrome, then it weakens their case that the mutations could logically also be the sole cause of a rare disorder like Marfan syndrome, so they stay with cause for Marfan syndrome and associated for mitral valve prolapse syndrome.
I do not believe that there is proof that fibrillin-1 mutations cause either MVP syndrome or Marfan syndrome. There may be some type of association between fibrillin-1 mutations and a wide variety of features of connective tissue disorders, but that is not in and of itself enough to prove cause and effect. There are many cases of hereditary disorders that have particular genes associated with the disorders, as has been found in spina bifida studies, but that does not mean the genes themselves had any direct or singular impact on the disorder. (See my section on Genetic Disorders: Treatable Through Nutrition for more on this topic.)
Numerous studies show that most of the people who have scoliosis have osteopenia or osteoporosis. Osteopenia and osteoporosis are highly multifactorial conditions with a large, established number of environmental factors, including diet and physical activity levels. Perhaps not coincidentally, magnesium deficiencies have also been linked to both conditions.
Fibrillin mutations may possibly have some
association to mitral valve prolapse or scoliosis, but if they
were the singular cause of the disorders, then all of these other studies
linking scoliosis and MVP to environmental factors would all have to be
Return to top
percent of all cases of Marfan syndrome occur in people with no family
history of the disorder. This makes it highly unlikely then
to be solely a genetic disorder. If we consider the possibility
that Marfan syndrome is influenced by both genes and environmental
factors, this would provide a logical explanation for why the disorder
occurs more often in some families, yet also occurs in people without
a family history of the disorder.
As you may remember from biology class, Gregor Mendel discovered that the color of pea plants was determined by inherited attributes from the parent plants. He found he could predict the color distribution of offspring pea plants based on the colors of the parent plants. The children pea plants always followed an established pattern of color inheritance. This genes-only pattern is not true for all plant color, however, and is even less likely to be true for complex medical disorders like Marfan syndrome.
If Mendel had picked another plant instead of peas to study, his results would not have been quite so tidy. The color of hydrangeas is determined in part by their genes and in part by the acid content of the soil, an environmental factor. One of my friends changes the color of her hydrangeas by dumping coffee grounds on some of them to achieve a multi-color effect. Her hydrangeas do not have "spontaneous genetic mutations". They have environmental factors influencing their colors - her used coffee grounds. That is why the colors of her hydrangeas do not follow any predictable inheritance patterns.
Instead of considering environmental
factors in Marfan syndrome, which would be the most logical and straight
forward consideration, geneticists often describe people with the syndrome
and no family history, as "spontaneous mutations", assuming that their
genes just "spontaneously mutated" to cause the syndrome. This theory
makes the assumption that environment plays no role in the disorder.
However, since there have never been any studies that ruled out environment
in Marfan syndrome (and there are plenty to suggest a connection), then
this is an illogical conclusion. If Marfan syndrome is, in fact,
caused by "spontaneously" generated genetic mutations, then anyone making
this statement should be asked to provide proof that genetic mutations
exist in the syndrome.
The large 1997 study on fibrillin mutations in Marfans showed that the fibrillin-1 detection rate was lower in sporadic cases than it was in familial cases. In sporadic cases, the detection rate for fibrillin-1 mutations was only 20%. This means that 80% of the people with sporadic cases were not found to have a fibrillin-1 mutation. This would not support the theory that the sporadic cases were caused by a spontaneous fibrillin-1 mutation, since it cannot be proven that 80% of the people with sporadic cases even have the mutation. Environmental factors as a consideration in Marfan syndrome remain a more logical and scientifically supportable reason as to why the disorder does not follow autosomal dominance patterns of inheritance. The most obvious reason Marfan syndrome doesn't follow autosomal pattern of inheritance is simply because it isn't an autosomal dominantly inherited disorder.
Note: For a more eloquent description of why spontaneous mutations are unlikely to just happen, and the role of nutrition in genetic expression, read the book The Driving Force, by Michael Crawford.
If you look at someone's genes and find
fibrillin-1 mutation, there is no way to predict whether that person is
simply tall, just has scoliosis or has Marfan syndrome. Having
a fibrillin-1 mutation does not mean you have Marfan syndrome.
Even among people within the same family who have identical fibrillin-1
mutations, one may have Marfan syndrome and one may not. This is
a pretty clear indication that the genetic mutation could not possibly
be the singular cause of the disorder. If it were, then everyone
with the genetic mutation would have the syndrome, but they don't.
Many people with the mutation do not even
have any features that are diagnostic criteria for Marfan syndrome.
It goes against all logic for
researchers to cling to their original hypothesis that fibrillin mutations
cause Marfan syndrome when their own research fails to support this conclusion.
Return to top
Researchers have some insights into fibrillin, however, they do not have a complete understanding of all the factors, such as diet or medications, that might affect fibrillin. My theory is that even if there is some type of association between Marfan syndrome and fibrillin-1 mutations, the defect may not necessarily be in the fibrillin genes. Perhaps the defect associated with Marfan syndrome is not in any gene per se, but in how peoples' diets and other environmental factors interact with their gene(s).
People with homocystinuria, have many features in common with Marfan's syndrome, i.e. Marfanoid habitus, dislocated lenses, etc., yet many cases of homocystinuria are highly responsive to nutritional therapy. Interestingly, fibrillin-1 connective tissue proteins are influenced by elevated homocysteine levels, and elevated homocysteine levels are influenced by one's diet. Elevated homocysteine levels can be reduced through nutritional therapy.
Some people with homocystinuria have been able to specifically improve their fibrillin-1 connective tissue proteins by correcting cysteine deficiencies. A 2000 research paper noted that, "...that a deficiency of cysteine and subsequent inhibition of fibrillin-1 accumulation in CBS deficient patients may be at least partly responsible for their phenotype, and suggest that maintenance of normal plasma cyst(e)ine levels may be an important therapeutic goal." In other words, they can correct their defective fibrillin-1 connective tissue proteins by changing their diets. Based on this knowledge, wouldn't the most logical assumption be that there may be other nutritional factors that can influence fibrillin-1 connective tissue proteins, including those of Marfan syndrome?
The concept that inherited disorders are caused by a singe gene is no longer widely accepted in most genetic research and literature. The $60 million Environmental Genome Project is based on the concepts that few disorders are caused by a single genetic or environmental event. The premise of the Environmental Genome Project, an outgrowth of the Human Genome Project, is that most disorders are in reality caused by complex sets of interactions between genetic and environmental factors. I don't understand why Marfan research isn't based on the same concepts as the Environmental Genome Project, especially since the fibrillin genetic research does not seem to be panning out.
There is very strong evidence to suggest that environmental factors play a role in Marfan syndrome. People with Marfan syndrome are often treated successfully with a medication called Propranolol to prevent aortic aneurysms. This proves that the course of Marfan syndrome can be altered by an environmental factor, because Propranolol is an environmental factor and it has an affect on the disorder.
There are many associations between Marfan
syndrome and copper deficiency symptoms, especially conditions such as
aortic aneurysms and the unusual condition of emphysema in nonsmokers.
aneurysm-prone turkeys and the aneurysm-prone blotchy mouse, like Marfans,
are successfully treated with Propranolol, yet separate studies show
that aneurysm-prone turkeys have also been successfully treated with
increased dietary copper. The turkeys have a hereditary
disorder, there obviously must be some genes that cause them to be aneurysm-prone.
Yet it is their diets, not just their genes, that also determine whether
or not they actually get aneurysms. If dietary copper helps the
turkeys, then would dietary copper, or perhaps some other dietary changes,
Interestingly, the aneurysm-prone blotchy mouse also develops emphysema, just like Marfans. The mouse has a problem with copper, too. It does not metabolize it correctly. Note the problem is with copper again - it's not niacin, it's not iron, it's not folic acid. It is copper - just like copper was a factor in the aneurysm prone turkeys. Copper deficiencies cause aortic aneurysms in a wide variety of animals. Copper deficiencies have also been linked to aortic aneurysms in humans. Copper deficiencies in animal studies have been linked to scoliosis, osteopenia, osteoporosis, and emphysema-like lung changes - all conditions that are features of Marfan syndrome. It seems unlikely to me that all of the similarities between features of Marfan syndrome and copper deficiencies are sheer coincidences. The role of copper in Marfan syndrome would seem to be a highly logical area to study. (See my section on Marfan Syndrome & Copper for more specifics and links to many of the studies.)
Calcification of heart calves, mitral valve prolapse, and rachitic skeletal defects such as osteopenia, pectus excavatum and scoliosis have all been linked to magnesium deficiencies and have also been linked to Marfan syndrome. Magnesium supplementation has been shown to be effective in treating many of these problems. The role of magnesium in Marfan syndrome would also be a highly promising area to research.
Why would a fibrillin mutation cause calcification of heart valves? I know of no studies that show that it would. Perhaps a more logical scenario would be to consider the possibility that fibrillin may need a biochemical like magnesium for its creation, and a shortage of magnesium affects fibrillin production and causes other conditions, like heart valve calcification, that are unrelated to fibrillin. Marfans also have abnormalities of hyaluronic acid, a magnesium dependent biochemical. A magnesium shortage simply provides a more logical explanation for many of the features we know about the disorder. It would be relatively easy and inexpensive to prove or disprove some of my theories by testing Marfans' cellular (not serum) magnesium levels and catecholamines excretion.
Among people with identical fibrillin-1 mutations, one person may have Marfan's syndrome and another may not. Logically, this then is a pretty clear indication that the fibrillin mutation alone cannot be causing the syndrome, and that other factors must be considered.
Because of these facts, common sense dictates that fibrillin mutations alone are not the only factor, if they are any factor at all, in the development of the syndrome. There is simply more evidence supporting the idea Marfan syndrome, like most other disorders, is not the result of genes alone, but rather a combination of genes and environmental factors.
Return to my page on Marfan Syndrome.