Genome Discovery Shocks Scientists

Genetic blueprint contains far fewer genes than thought

DNA's Importance Downplayed

Tom Abate / SF Chronicle 11feb01

[ Analysis shows it's proteins not genes that count  Maggie Fox / Reuters 11feb01 below ]

Deepening the mystery of what makes us human, two scientific teams will publish the first maps of the human genome this week, revealing that the human genetic blueprint contains only a third as many genes as had been supposed.

The publication of the maps is a milestone in the decade-long, multibillion- dollar effort to decipher the DNA that carries the set of instructions, passed on from parents to children, for making a human being.

Until recently, scientists had expected to find as many as 100,000 genes in the genome. But the two scientific teams, reporting their findings this week in the journals Science and Nature, independently found only about 30,000 genes.

The paucity of genes left scientists struggling to understand how humans could be so much more complex than other animals with essentially the same number of genes.

"We have only 300 unique genes in the human (genome) that are not in the mouse," said Craig Venter, president of Celera Genomics, the Maryland firm that led one of the mapping teams. "This tells me genes can't possibly explain all of what makes us what we are."

Francis Collins, leader of the U.S. contingent to the Human Genome Project, a consortium of publicly funded scientists from around the world, said the findings will force scientists to look for other factors to explain many aspects of health, disease and behavior.

"We had a hard time explaining the (genetic) control mechanism when we thought there were 100,000 (genes)," Collins said. "Now we have only a third as many."

Celera and the public scientists of the Human Genome Project have been rivals in the race to map the genome. In June, the two teams announced at a White House ceremony that they had read most of the 2.91 billion chemical letters found in each strand of human DNA.

This week, in dozens of articles in the journals Science and Nature, Celera and the public team will publish their first efforts to unscramble the genetic secrets captured in that jumble of chemical letters.

In an unusual move signaling the importance of the findings, Nature and Science planned to hold a press conference tomorrow. However, after some British newspapers violated a publishing embargo, the journals hastily gave the media the green light to run their stories today. Venter's team will publish its findings in Science, and the public project will publish theirs in Nature.

The publications include other surprising findings, such as the discovery that vast stretches of noncoding regions in human DNA -- what has been called "junk DNA" -- may actually play an important role in driving and recording evolution.

Eric Lander, a geneticist at the Massachusetts Institute of Technology and a scientific leader of the Human Genome Project, sounded awestruck as he summarized the article he and his scientific allies published in Nature.

"I think the junk is the biggest surprise in the genome," Lander said.

Lander said half of our genome seems to consist of repeating elements of DNA that have copied and inserted themselves into the sequence. Lander said scientists used these repetitive sequences to study how the genome evolved.

They analyzed how the repeated sections became garbled as they were copied over and over, much as the original message gets garbled in a children's game of telephone. The more garbled the sequence, the older the repetition. Based on computer analyses of this garble, Lander said, scientists have dated when various sections of the genome were created.

"It's as if we've found this ancient history that contains many legendary stories," Lander said. "We have just dug up the fossil record for 700 million years of evolution."

The genome consists of 23 pairs of chromosomes, which are present in the nucleus of almost every human cell. Each chromosome hosts thousands of genes, strings of DNA letters that carry instructions for making one or more of the proteins constitute the body's chemical workforce.

But genes are few and far between. Less than 1.5 percent of the genome seems to code for proteins, Lander said. Scientists thought it was twice as much before the map was done.

Both scientific teams painted the genome as a landscape of vast dark stretches of repetitive chemical letters, interspersed with genes that seem as rare as city lights seen from an airplane at night.

"The good news is that these two papers, which are the result of different approaches, offer pretty consistent findings, which increases our confidence that we're all on the right track," Collins said.

Key findings include:

-- How earlier gene counts went wrong: Years ago, scientists noticed that genes seemed to average 3,000 letters. Given early and accurate estimates of a genome roughly 3 billion letters long, that meant 100,000 genes -- provided genes were evenly distributed.

Among the surprises in the maps, however, is just how unevenly genes are sprinkled throughout the 23 pairs of chromosomes that comprise our DNA. Venter's Science paper reports that chromosome 19 has 23 genes for every million DNA letters. Chromosome 2 has only five genes in the same expanse.

Venter said that when his team sampled chromosome 19 and another relatively gene-rich region on chromosome 4 not long ago, they extrapolated their findings to a gene count for the whole genome -- and got it wrong. "It was a statistical fluke," he said.

-- Uses for "junk DNA": Human DNA is constructed from two chemical strands that take the shape of a long, twisted ladder. The four chemical letters that make up each rung of the ladder follow a strict rule: chemical A always stands opposite T, and G pairs with C.

But these four letters are not evenly distributed. The gene maps show that ATs outnumber GCs roughly 60-40.

For reasons not completely clear, genes seem to cluster in GC rich regions, while so-called junk or repetitive DNA generally confines itself to AT zones.

Lander said the one exception to the rule is a string of repetitive letters called the Alu sequence. Many Alu sequences cluster in the GC regions of the genome, alongside genes.

Lander said that in 1998, Carl Schmid, a molecular biologist at the University of California at Davis, advanced what seemed like a nutty idea to explain Alu's unusual affinity for genes. Schmid suggested Alu sequences resided near genes because they weren't junk, but rather a mechanism to help cells repair themselves.

With the entire genome map in front of them, showing so many instances of Alu sequences around genes, scientists are beginning to take Schmid seriously. "It looks pretty convincing," Collins said.

Collins said scientists have found evidence that other repeat sequences have carried bits of useful DNA with them when they jumped around the genome, creating or modifying genes in the process. All of this suggests, as Collins put it, "that some of our junk isn't junk after all."

-- Men, women and mutations: Public gene mappers made one finding that could thrust the genome into the battle of the sexes.

The public scientists theorize that men pass on mutations to their offspring twice as often as women. They surmised this by comparing the genetic sequences of the X and Y chromosomes. Women have two Xs, men an X and a Y. The scientists analyzed the repeat elements on the Y chromosome and measured how often these repeats became garbled. After performing the same analysis on the X chromosome, they concluded "most mutation occurs in males."

The public scientists suggested that the higher male mutation rate is due to the fact that men make billions of sperm, while women are born with far fewer eggs, only a few hundred of which mature in a lifetime. These numbers favor mutations being introduced when sperm-producing cells copy DNA on the Y chromosome. Whether this is good or bad is debatable, since mutations can both introduce disease and pass on advantageous traits.

"Moms can look at dads and say, 'Two-thirds of the time you're responsible, " Lander quipped. "Dads can look at moms and say, 'Yes, but I'm doing two- thirds of the heavy lifting for evolution.' "

-- Other explanations for complexity: As scientists struggle to explain human complexity with so few genes, they note that our genes are far better at multi-tasking than the genes of other species scientists have studied.

"The average human gene can make three proteins, which is more than most people expected," Collins said.

Even so, having come up so short on genes, scientists can't explain how humans can be so much more complicated than fruit flies, which have roughly half as many genes as humans.

Part of the answer could lie in the makeup of human proteins. "In general the proteins in the human are more complex than the proteins in other organisms we've studied," said Mark Adams, vice president for genome research at Celera and a co-author on Venter's paper. "They are capable of more interactions, they do more things."

This is important because molecular biologists see the human body as a machine, in which proteins serve as the gears, motors and pulleys that perform every task from flexing muscles to firing nerve synapses. Before the mapping project, genes were considered the control software in this analogy. But now that metaphor seems dated.

The emerging view is that much of our complexity must derive from proteins, which interact to build physical systems somewhat independent of direction from the genome.

Venter said all these findings undermine the concept of genetic determinism,

the notion that genes determine everything from our behavior to our propensity toward illness.

"It's become part of the common language to say we'd like to have the gene for this or the gene for that, but the common language is wrong," Venter said.

"I believe all of our behaviors, all of our sizes and functions clearly have a genetic component but genes only explain a part of any process," he said. "We are around as a species because we have an adaptability that goes beyond the genome. If everything was hard-wired, we wouldn't have survived."

On a whimsical note, scientists will wait two more years to decide who won their "gene pool."

Last summer, more than 200 geneticists tossed a dollar into a hat and predicted how many genes would eventually be found. Estimates ranged from 28, 000 to 200,000. From the start it was decided to put today's estimates to more tests. The winner will be announced at a genetics conference in 2003, to coincide with the 50th anniversary of the discovery of the structure of DNA.

E-mail Tom Abate at tabate@sfchronicle.com

UNDERSTANDING THE HUMAN GENOME

Geneticists have found about 30,000 genes in the human genome  only a third as many as had been thought.

What is DNA?
Found in every living cell of every living thing, DNA is the instruction manual for life. The human body contains about 5 trillion cells, and within each one lies a nucleus that contains DNA strands. DNA holds the information to create proteins that cells need to grow and replicate themselves. Each DNA strand consists of a combination of four molecules that make up base pairs, the building blocks of an individual's entire genetic information or genome. 

Gene count
Laboratory mouse: About 50,000 genes
Fruit fly: 13,600
Thale cress, a plant: 25,500
Human: About 30,000
Rice: About 50,000

New York Times; Associated Press; Incyte Genomics, John Blanchard / The Chronicle

Analysis shows it's proteins not genes that count  

Maggie Fox / Reuters 11feb01

WASHINGTON -- Our future may not lie in our genes, after all.

Two separate teams of researchers will report Monday that they have taken the first in-depth look at the human genetic code and found about half what they expected to find. Instead of 60,000 to 80,000 genes, we have only 30,000 to 40,000.

Both teams agree this means that, in humans anyway, it is proteins that matter -- much more so than genes.

"Those who are looking for forgiveness of responsibility for their own lives in the genetic code will be very disappointed," Craig Venter, president and chief scientific officer of Celera Genomics Inc., the private company that did one of the studies, said in a telephone interview.

The human body, it seems, is set up to adapt to its environment, by cutting up and recombining the protein "products" of genes to make a protein suitable for the circumstance.

Each gene makes one protein -- this is the basic function of any cell. Researchers had known that proteins often have to be sliced in a certain way, a process known as cleaving, before they do anything useful.

"Most of biology happens at the protein level, not the DNA level," Venter said.

What had not been known was the degree to which this is true. The implications could be profound for medical science, which had hoped to find easy genetic answers to disease and to how people will respond to drugs.

GENE PATENTS "IRRELEVANT"

"This shows how irrelevant human gene patents are," Venter said. "The drug industry has been saying 'one gene, one patent, one drug'. But the uses for this approach can be counted on fingers."

Both teams, who publish their findings in the rival scientific journals Nature and Science, are fairly certain.

"Given all the tools that we threw at this problem, we cannot imagine that there are many more genes," Mark Adams, vice president at Celera, told a briefing for journalists.

"We only have twice as many genes as a fruit fly. But we are more complex. We can think more thoughts. Our bodies can do more things."

Humans have 3.1 billion base pairs of genetic code. A base pair is a joining of two nucleotides -- known by the letters A,C,T and G. These repeat over and over in various combinations to make amino acids, which in turn combine to make proteins.

"The size of the genome, the number of base pairs, is irrelevant to biology," Venter said.

"Corn has the same number of genes as humans. The lily plant has 91 billion pairs of genetic code."

Each protein equals a gene, but there are long stretches of base pairs that do not code for proteins, areas once known as junk DNA. These areas may help control genes.

Only just over one percent of the genome is accounted for by protein-expressing genes. Venter says all this means genes, per se, are just a small part of the story.

"Genes don't determine whether you get colon cancer," he said. "They determine whether you have an increased risk for colon cancer. We get a set of probabilities from our genetic code, a sort of range of parameters that we can work within."

KIND OF HUMBLING

"It's kind of humbling, isn't it?" Ari Patrinos of the U.S. Department of Energy, which funded much of the public effort, said in a telephone interview.

"There are very, very few few traits or diseases that are monogenic (caused by a single gene). It's been an emerging consciousness over the past five years, and the recognition that ... our genes don't control everything."

It also means the so-called "junk DNA" may be more important than at first thought.

"We just don't know. We don't call it junk," Venter said.

Eric Lander, who heads the genome lab at Massachusetts Institute of Technology's Whitehead Institute, said the "alleged junk" provides a history.

"The junk is amazing. Every piece of junk in the genome represents a transposable element," he said.

In other words, it is genetic material that people got from elsewhere, such as from bacteria the readily lend their DNA out, retroviruses that inject their genetic information into cells, or by a cut-and-paste process done by genetic elements known as transposons. If it stayed there through generations, it might do something useful.

Lander thinks some of the "junk" may help regulate genes --a role that is more important the fewer genes there are.

And some of the genes are borrowed, too. Lander said his team found that the gene for monoamine oxidase, an enzyme implicated in depression and targeted by drugs called MAO inhibitors, came from bacteria.

NOT EVERYONE AGREES

Not everyone agrees with all the conclusions.

"We know that they have missed very, very many genes that we know exist," William Haseltine, head of Rockville, Maryland-based Human Genome Sciences Inc., said in a telephone interview.

"They have missed at least half the genes, maybe more," added Haseltine, whose company holds more than 100 gene patents. "They have no medical discovery and they only found a third of the genes. That's a bore."

Haseltine, whose company looks for "expressed" genes --those that actually make a protein -- by using bits of DNA called expressed sequence tags (ESTs), says he believes there are 120,000 human genes.

Another company that says it has explored the genome, Palo Alto, California-based Incyte Pharmaceuticals Inc., maintains there are 140,000.

If you have come to this page from an outside location click here to get back to mindfully.org