SEARCH

SEARCH BY CITATION

It is striking how often it is that a 30- to 50-year-old patient will walk into my office with a seemingly mild yet enigmatic problem such as an abnormal aminotransferase level or epigastric pain that persists despite thorough investigation and treatment from their internist. These patients invariably have complicated family histories of multiple cancers or chronic diseases such as diabetes, rheumatoid arthritis, Alzheimer's disease, and others. Recently, I began linking some of these clinical findings to potentially dysfunctional genetic pathways to effect a diagnosis, treatment, and management plan, and possibly avert disease escalation. Take one recent example of a patient who presented with abnormal aminotransferases and gastric fundic gland polyps (not on proton pump inhibitors). I noted her family history of hepatocellular carcinoma (HCC) and gastric cancer spanning three generations. Familial adenomatous polyposis coli/Gardner's syndrome with dysfunctional wnt signaling, epigenetic loss of E-cadherin, and silencing of transforming growth factor β signaling are some of the possibilities.1–3 With the right information at hand, this patient's management plan could include: a screening strategy for HCC and gastric cancer; identifying mutations in affected family members; examining affected gastric tissue for loss of E-cadherin; instituting simple preventive/disease modifying measures such as vitamin D, calcium, and aspirin that switch off pro-oncogenic activated wnt signaling.4, 5 Clearly, many patients will need to be studied carefully with genetic networks and a systems biology approach to identify key modifying factors that will effectively work on an individual basis.6 I have most definitely embarked on a steep learning curve as I continue to manage my patients. Possibly, as in hereditary non-polyposis colorectal carcinoma (HNPCC) studies, this patient and her family will eventually fall into a specific genetic group, and we will be able to look to future prevention of other cancers, such as the uterine cancers by earlier hysterectomies in HNPCC patients.7

As we progress through the 21st century, it is clear that as hepatologists who think in terms of hypothesis-driven clinical research using Oslerian principles and now a systems biology approach, we have come a long way from the days of prerandomized controlled trial use of herbal medicines. These empiric cures of liver diseases date back to the Xia dynasty in China and the Indian Vedic period (2100 B.C.), with written reports in the Indian Caraka Samhita and the Eastern Zhou dynasty of China as far back as 600 B.C. and 400 B.C., respectively.7 Around 129 A.D., the Roman physician Galen, from the bustling and vibrant city of Pergamum, argued that the liver was the principal organ of the human body, and that it emerged first of all the organs in the formation of a fetus. “The liver is the source of the veins and the principal instrument of sanguification,” states the text On the Usefulness of the Parts of the Body. For Galen, it was the liver rather than the heart where blood was most actively formed; it is a warm, moist organ. If all veins exchanged fluids through the liver, connecting only tenuously to the heart in order to provide a tiny amount of blood to mix with spirit in the arteries, then the liver would be the center of the circulation of material substances in the body. Galen's tenure as the leading authority in medical theory lasted for at least 1400 years. Fast-forward to the 20th century, when a formal study of liver diseases began with the formation of the American Association of the Study of Liver Diseases (AASLD) in 1950 by a small group of leading liver specialists (including Hans Popper, Leon Schiff, Fred Hoffbauer, Cecil Watson, Jesse Bollman, and Sheila Sherlock). Today the 7,000 members of the AASLD with scientists and health care professionals from other disciplines have successfully identified Hepatitis A through E, genetic pathways for a host of liver diseases, and new preventive/treatment strategies such as hepatitis B virus vaccination/interferon for liver disease. Furthermore, the universal role of the liver in systemic diseases offers us an even better vantage on key scientific findings and clinical management of liver disease.

Then, 2001 brought forward a landmark in science: the complete sequencing of the human genome by a consortium of laboratories led by Francis Collins and independently by Craig Ventnor, a task that took roughly 10 years to accomplish.9 Sequencing of the human genome represents far more than the effects of individual genes, giving a more integrated view that examines whole ensembles of genes as they interact in health and human disease. By coincidence, the publication dates fell during the week of the anniversary of the birth of Charles Darwin. Darwin's message that the survival of a species can depend on its ability to evolve in the face of change resonates ever more today with the newer advances.

Human genome sequencing revealed 3 billion DNA bases in the right order, conserved regulatory regions, RNA genes, and the 98% that lies outside protein-coding regions. There are an estimated 15 million places along our genomes where one base can differ from one person or population to the next. By mid-2007, more than 3 million such locations, known as single-nucleotide polymorphisms (SNPs), had been charted.9 Called the HapMap, this catalog has made the use of SNPs to track down genes involved in complex diseases—so-called genome-wide association studies—a reality. More than a dozen such studies were published this year.

Traditionally, as scientists, we have hunted down genes by tracking the inheritance of a genetic disease through large families or by searching for suspected problematic genes among patients; examples include genetic iron overload disorders such as hemochromatosis related to mutations in the HFE, hemojuvelin, transferrin receptor 2, and hepcidin genes.10 Genome-wide association studies now go much further. They compare the distribution of SNPs—using arrays that can examine some 500,000 SNPs at a time—in hundreds or even thousands of people with or without a particular disease. By tallying which SNPs coexist with symptoms, researchers can determine how much increased risk is associated with each SNP. In the past, such links have been difficult to develop and required years of analysis. This year, however, researchers linked variants of more than 50 genes to increased risk for a dozen diseases. Predicting a disease as clinicians, or organism phenotype from the genotype, remains a central question in medicine and genetics. Most importantly, we would like to find out if the perturbation of a single gene may be the cause of a disease.

How could networks help us as hepatologists? To begin with, take the example of elucidating the phenotypic expression and metabolic pathways of copper and iron. Such an analysis could improve our understanding of genetic disorders such as Wilson disease, hemochromatosis, and nongenetic iron overload syndromes through insight into dysregulation of HFE, hepcidin, or others.10, 11 A better understanding of genetic and nongenetic modifiers in ZZ α1-antitrypsin deficiency and other protein-folding mutations such as Gaucher disease and cystic fibrosis is a future to which we should aspire. Genetic predictability of risks in, for example, glycogen storage disease for hepatic adenoma formation and transformation to HCC would be important for tumor surveillance guidelines and transplantation guidelines. In the three distinct forms of familial intrahepatic cholestasis that are the result of mutations in the ATP8B1, ABCB11, and ABCB4 genes, wide variations in clinical phenotypes are observed. Mutations in ABCB11 and ABCB4 genes result in abnormalities in canalicular excretion of bile acids and phospholipids, respectively. The molecular pathophysiology of the systemic disease associated with mutations in ATP8B1 remains unclear.12, 13 Disease- and genotype-specific prognoses and therapeutic approaches would be essential for future therapeutic intervention. As for many other tumors, development of HCC occurs as a multistep process with accumulation of genetic and epigenetic alterations in regulatory genes, leading to activation of oncogenes and inactivation or loss of tumor suppressor genes. More recently, the clonal nature of HCC and its cancer stem/progenitor cell phenotype and epigenetic inactivation of (tumor suppressor) genes by promoter hypermethylation has been recognized as an important and alternative mechanism in tumorigenesis.14 Aberrant methylation of promoter sequences occurs in premalignant conditions such as chronic hepatitis B or C and cirrhosis of the liver as well as in HCC.

However, our current ability to predict the phenotypic effects of perturbations of individual genes is limited. Network models of genes are one tool for tackling this problem. In a recent study, it has been shown that network models covering the majority of genes of an organism can be used to accurately predict phenotypic effects of gene perturbations in multicellular organisms.15 A computational network covering most Caenorhabditis elegans genes was tested for its predictive capacity. Briefly, it identified six genes whose inactivation suppresses defects in the retinoblastoma tumor suppressor pathway as well as successful prediction that the dystrophin complex modulates epidermal growth factor signaling. An analogous network for human genes might be similarly predictive and thus may facilitate identification and targeting of disease-associated genes. For instance, it is possible that for iron overload disorders, a connection similar to that of the Rb pathway could give rise to a surprise; hypothetically, defensin/innate immune function could become connected, creating new therapeutic possibilities such as dietary isoleucine for the management of genetic as well as the more common nongenetic causes of iron overload. In this way, it may be possible in the near future to identify specific subgroups of patients (for example, those with chronic hepatitis C) that are at risk for HCC, through genetic profiles, genetic/epigenetic networks facilitating and generate a successful screening/therapeutic plan that will avert this lethal cancer.

It is this great and evolving tapestry of information in which, with rules that encourage exploration and reward creativity, we can find many of the answers that will help us approach and define the optimal management of liver diseases, and in turn multiple systemic illnesses.

Acknowledgements

  1. Top of page
  2. Acknowledgements
  3. References

I would like to thank Bibhuti Mishra, Viveka Mishra, Premkumar Reddy, Keith Lindor, Sandra Acheampong, and Stephen Byers for their helpful suggestions and editing.

References

  1. Top of page
  2. Acknowledgements
  3. References