Four Harvard biochemists announced yesterday that they have artificially reproduced a mammalian gene.

The breakthrough will allow scientists to produce large quantities of most genes, aiding research on thalassemia, a form of anemia, and similar genetic diseases, one of the reasearchers said yesterday.

"Our ultimate hope is a gene therapy" that would cure diaseases like thalassemia by supplying victims with needed genetic information, the researcher, Thomas P. Maniatis, assistant professor of Biochemistry and Molecular Biology, said.

"Why does a liver cell differ from a brain cell, when they both carry the same genetic information," Maniatis asked. Artificial reproduction of genes can answer such questions, he said, because it can act as a probe into DNA organization and control in higher animals.

DNA is a large complex organic molecule. A gene consists of two sections of DNA curled around each other in a double helix.

The team of researchers began by synthesizing a single strand of DNA. They then used an enzyme to add a complementary second strand, forming a complete gene. A "hook" of DNA material at the end of the first strand served as a "primer," or base for the enzyme activity, Fotis C. Kafatos, professor of Biology and one member of the team, said yesterday.

Although the four completed the synthesis itself last spring, they still had to cut the hook connecting the two strands of DNA and prepare data "rigorous" enough for publication, Kafatos said. The additional work delayed completion of the experiment until October, he said.

Matthew S. Meselson, professor of Biology, said yesterday that the team's technique was essentially a "time-saver." "Other methods of getting genes for study are more or less of a fishing expedition" because they require removing an individual gene from a large and complex chromosome, Meselson said.

Meselson said that the new technique changed what would otherwise be a "major undertaking" into a few days' work, and added that he would soon be using it in research on fruit fly genes.

Maniatis, Kafatos, and Argiris Efstratiadis, a doctoral candidate in Biology, worked on the project for the past year and a half. Allan Maxam, research assistant in Biochemistry and Molecular Biology, joined them last April.

The gene synthesized was equivalent to the gene in rabbits that produces hemoglobin, Kafatos said. He added, though, that many genes can be produced in this manner and that the rabbit gene had been selected because it is a common subject of research.

Maxam said yesterday that as a side effect of the experiment, techniques involving the splitting of DNA and the handling of the resulting fragments had been refined and improved, benefitting future experimenters.

The team, which was funded by the National Science Foundation and the National Institutes of Health, plans to use their synthesis techniques as a base for further study of DNA, Maniatis said