The team, led by Dr. Peter Lee, focused on lactobacilli bacteria that naturally live in the vagina.

Other studies have found that women with low levels of lactobacilli may be more likely to contract the AIDS virus, which is mostly transmitted by sex between men and women.

Lactobacilli in the digestive tract are key to normal health and can also fight off dangerous invaders.

Lee and colleagues looked at whether the microbes could be made stronger to specifically fight off HIV.

Writing in the Proceedings of the National Academy of Sciences, the researchers said they gave the bacteria an extra boost by adding the gene for CD4 -- one of the molecular doorways that HIV uses to get into cells.

If HIV latched onto CD4 on a lactobacillus before it reached a human cell, it might remain harmless, Lee said. Then the bacteria might destroy the virus with naturally produced lactic acid and hydrogen peroxide.

Tests on lab dishes full of cells showed the genetically engineered bacteria reduced the rate of HIV infection in cells by at least half. Preliminary studies using monkeys showed the engineered bacteria grew well and were safe, the researchers said.

If it works, the bacteria could be made into a vaginal suppository a woman could quietly use to protect herself. Such an approach is called a microbicide and many groups are pressing for their development.

"It would be as discreet as can be," Lee said. It might also be adapted to protect a woman against a range of sexually transmitted diseases.

Condoms are considered the best way to protect against HIV, which infects 43 million people worldwide and has killed 28 million since the AIDS pandemic started in the 1980s. There is no cure and experts agree that prevention offers the best hope of battling the killer virus.

But not everyone has access to condoms, and many men refuse to wear them.

Half of all HIV-infected adults are women, with the numbers higher in Africa, the continent worst-hit by HIV. According to UNAIDS, two million women were infected with HIV in 2002.

By Maggie Fox, Health and Science Correspondent

WASHINGTON (Reuters) - The smallpox vaccine may help protect people against the AIDS virus, U.S. researchers said on Thursday.

A team at Virginia's George Mason University said they had shown, in lab dishes, that blood cells from people vaccinated against smallpox were four times less likely to become infected by the AIDS virus.

They are now negotiating with Acambis Plc, which makes vaccines against smallpox and other diseases, to test their idea further.

"Our outcomes are very encouraging," Ken Alibek, a bioterrorism and smallpox expert at George Mason, said in a statement. "Additional studies that may lead us to more definitive conclusions already are under way."

Many researchers have proposed links between protection against smallpox and against the AIDS virus. Some studies have noted that older people who were vaccinated against smallpox were also less likely to contract HIV.

A study published in 1999 showed that a relative of smallpox, called the myxoma poxvirus, uses the same cellular doorway -- the CCR5 receptor -- to infect a cell as AIDS does.

And studies have noted that people with certain mutations in CCR5 are resistant to HIV infection.

Alibek, Raymond Weinstein and colleagues at George Mason used these studies as the basis for their experiment, said Jerry Coughter, director of life science management at the university.

They took the blood of 10 people vaccinated against smallpox and 10 people who had never been vaccinated against smallpox, and exposed both batches to the HIV virus that causes AIDS.

"They saw an average four-fold reduction in infectivity," Coughter said in a telephone interview.

The study results have not been subjected to a peer review, in which other experts examine the data and find potential flaws. Coughter said the findings had been submitted to a scientific journal as a first step in this process.

HIV infects an estimated 43 million people worldwide and has killed 28 million. Work on a vaccine so far has failed to produce a shot that protects against infection.

Smallpox was eradicated in 1979 through a global vaccination effort. Hundreds of millions of people have been vaccinated against the virus and the United States and other countries have resumed limited vaccination campaigns because of fears the deadly virus could be used as a biological weapon.

By Jeremy Lovell

MANCHESTER, England (Reuters) - A British researcher said on Tuesday many young gay British men were flouting safe sex and trying to catch HIV/AIDS in their search for identity.

But Europe's biggest HIV and AIDS charity dismissed the claim, saying there was no evidence to support it and that it would only serve to further demonize gay men.

Melissa Parker of Brunel University said the booming gay sex pub, club and sauna scene in London had been growing since the mid-1990s, with unprotected sex with multiple partners the rule.

Catching the AIDS precursor HIV was not just a risk, it was a goal, she told reporters at the annual meeting of the British Association for the Advancement of Science.

"Being diagnosed with HIV is a badge of recognition of being truly gay," she said. "There is a sizeable number of young gay men new to the gay scene, exploring their sexuality but wanting to belong. HIV is seen as a bonus."

The combinations of therapies now available for people with HIV might even have exacerbated the problem.

"They have made HIV less frightening," Parker said.

Admitting she had no figures to support her claim, Parker said her research consisted of several years of interviews with a large number of sexually active gay men.

But Will Nutland, head of gay men's health promotion at the Terrence Higgins Trust, said the research was unreliable.

"Her assumptions will only serve to further demonize gay men," he said.

"Her comment that unprotected sex with multiple partners is the rule, rather than the exception is not backed up by evidence," he said. "On-going surveys of gay men's sexual behavior since 1997 show that most men use condoms most of the time with most of their partners."

He said that while levels of unprotected sex have increased, gay men still viewed HIV as a serious medical condition.

The Terrence Higgins Trust, named after one of the first people to die of AIDS in Britain, is the leading HIV and AIDS charity in the UK and the largest in Europe.

MERAFONG, South Africa (AFP) - Tuberculosis is the number one killer among South Africans suffering from HIV and AIDS, an international expert said.

Patrick Bertrand, regional coordinator of the Swiss-based Massive Effort Campaign, a non-profit organisation specialising in fusing efforts to combat TB, AIDS and malaria, said the likelihood of HIV/AIDS sufferers contracting TB was 10 times higher than for those who are HIV-negative.

"Tuberculosis is now the most important killer of people with HIV/AIDS in South Africa," he told AFP after a three-day meeting of experts to look at ways of creating a greater awareness of the disease.

"Yet, it is an easily treated and curable disease -- the drugs and diagnosis are free and accessible everywhere in the country," Bertrand said during a visit to clinics at Merafong, a mining community about 80 kilometres (50 miles) southwest of Johannesburg.

South Africa is the country with the highest co-infection rate of HIV and TB in the world, with more than half of those suffering from TB also suffering from HIV/AIDS, Bertrand added.

It is ranked as the country with the seventh-highest number of TB cases globally, following India, China, Indonesia, Bangladesh, Nigeria and Pakistan, according to the UN World Health Organisation's 2003 report on the global spread of the disease.

Health workers at Merafong, surrounded by some of the country's richest gold mines and which has one of the country's highest rates of TB infection, said they were facing an uphill battle against the disease.

Sister Thembi Kariyeni, a health worker in the dusty Khutsong township, told AFP the rate of patients defaulting on their treatment was high.

"Because you deal with migrant mine workers, they often go back to their homes elsewhere in southern Africa and then for various reasons don't take their medication, including giving it to other relatives who suffer from the same disease.

"This gives rise to strains of drug-resistant tuberculosis, which is a lot more expensive and a lot more difficult to treat," she said.

Treating a patient with the drug-resistant strain could cost an estimated 10,000 dollars a year, as opposed to around 10 dollars a year for treating a patient with normal TB, experts said.

Field workers said they believed that the infrastructure already put in place by South Africa's health authorities to treat TB could also be used in the rollout of anti-retroviral drugs (ARVs) to treat AIDS.

The South African government ended years of denial last month by recognising the efficacy of ARVs and launched a plan to provide the drugs to its five million HIV-positive citizens, but has yet to give details on how the rollout will take place.

"We believe that because there is such a strong link between HIV/AIDS and TB, they can both be treated at the same time," said one field worker, who asked not to be named.

But she added: "With HIV/AIDS there has been a lot of publicity and we haven't seen a lot of drugs. With TB, there are a lot of drugs and treatment available for free, but there has not been a lot of publicity around it."

By Will Boggs, MD

NEW YORK (Reuters Health) - AIDS drugs alone cannot control the spread of HIV in established epidemics of the infection, new research suggests.

Dr. Ronald H. Gray, from Johns Hopkins University in Baltimore, and colleagues used data from studies in Uganda to develop a model that would predict the impact of AIDS drugs or an HIV vaccine on the course of the HIV epidemic there.

The findings are reported in the medical journal AIDS.

Under current treatment standards, AIDS drugs alone could not control the epidemic, the authors found. Still, treatment of 75 percent of eligible patients could result in a 7 percent decline in the HIV-infected population. Only if all HIV-infected persons were treated could the epidemic actually be controlled.

An HIV vaccine alone could stop the epidemic, the researchers note, but it would have to be very effective. However, a less effective vaccine could be combined with AIDS drugs to bring the epidemic under control.

None of these strategies would work if people started engaging in risky behaviors because they felt their risk of infection was low.

"We cannot rely on treatment to control the epidemic," Gray told Reuters Health, "and we need to plan for an ever increasing population of persons requiring treatment in the future."

Resources assigned to HIV treatment and prevention are inadequate, Gray said. "If there is investment in treatment at the cost of prevention, the epidemic cannot be controlled. We need to balance humanitarian considerations against public health priorities," he concluded, "and I am concerned that emphasis on treatment may distort this balance."

SOURCE: AIDS, September 5, 2003.

PARIS (AFP) - A ban on life insurance for people with the AIDS virus is in many cases unjustified, according to the first study to provide hard actuarial evidence about the benefits of anti-retroviral drugs.

Life insurance companies, scared by the mortality rates seen in the early years of the AIDS epidemic, routinely deny insurance to people with the human immunodeficiency virus (HIV) and this can cause big problems for those individuals, in their business and personal lives.

But a Swiss study says the automatic ban fails to take into account the success of antiretrovirals -- the drug "cocktail" that emerged in the mid-1990s and which, for many people with HIV, has otherwise prolonged their survivability.

People who respond well to HIV and who do not have hepatitis C have a short-term mortality rate that can be even lower than people who have been successfully treated for cancer and who are usually able to get life insurance, the study says.

Amongst the HIV group, the excess death rate -- a figure that compares patients to people without the disease -- was below five per thousand patient-years.

Amongst the cancer group, the rate varied from five to 20 per thousand patient-years.

The study, published in this Saturday's issue of the British medical weekly The Lancet, "provides preliminary evidence that life coverage could be considered under specific conditions," the authors say.

However, short-term mortality was significantly higher among HIV patients who had failed to respond to antiretrovirals or who had hepatitis C, a disease linked to intravenous drug use.

The research was led by Bernard Hirschel of Geneva University Hospital, and drew on figures from a six-year-old, ongoing study of Swiss patients with HIV and from national mortality statistics.

Antiretrovirals contain the spread of HIV virus so that the body's immune system remains intact and does not reach the stage of full-blown AIDS, when death can be caused by opportunistic diseases.

These drugs are not a cure, however, and can have toxic side effects.

Their cost, too, is a barrier to treatment. Only recently has the price been lowered sufficiently for developing countries to contemplate distributing the treatment on a wide scale.

PARIS (AFP) - A study says a frontline anti-HIV drug, nevirapine, can provide a cheap, lasting shield to infants in poor countries who are at risk of catching the AIDS virus from their mother's breast milk.

A nevirapine trial involving Ugandan children found the vast majority of those who did not have the human immunodeficiency virus (HIV) after birth and in their early weeks of life remained HIV-negative 18 months later, it says.

"This simple, inexpensive, well-tolerated regimen has the potential to significantly decrease HIV-1 perinatal transmission in less-developed countries," it says.

The trial, HIVNET 012, enrolled 637 mothers in Kampala from 1997 to 1999.

The mothers were given either nevirapine or zidovudine at the onset of labour, and their baby was given a dose shortly after birth, while 19 were given placebos.

In 1999, the HIVNET 012 reported that the risks of HIV transmission to these infants were 8.1 percent (nevirapine) and 10.3 percent (zidovudine) at birth, which rose to 13.5 percent and 22.1 percent by the time they were 14-16 weeks old.

In a followup study, they determined the risk at the age of 18 months stood at 15.7 percent for nevirapine and 25.8 percent for zidovudine.

In 2002, an estimated 800,000 children became infected with HIV through mother-to-child transmission. More than 90 percent lived in developing countries.

Giving the mother the drug at the start of labour, and to the child for a short period post-natally, is believed to reduce the risk of HIV during delivery and through breast feeding.

AIDS campaigners say this technique is inexpensive but also simple to administer -- a vital factor in countries where people may live in remote rural areas where there is no access to followup care.

By Megan Rauscher

NEW YORK (Reuters Health) - Rapid HIV-1 testing performed at the bedside in labor wards by non-laboratory hospital staff gives accurate results much faster than lab testing.

This means anti-HIV drugs can be given promptly to the mother and newborn, measures proven to reduce the odds of the baby being infected with HIV.

The findings come from the Mother Infant Rapid Intervention at Delivery (MIRAD) study, funded by the Centers for Disease Control and Prevention, and published in the CDC's Morbidity and Mortality Weekly Report.

Four Chicago hospitals with the city's highest caseload of HIV-positive women of childbearing age participated in the study, conducted during the first 6 months of 2002.

In three of the hospitals, nurses, midwives, and physicians performed the FDA-approved OraQuick Rapid HIV Antibody Test onsite, using blood specimens from consenting women with unknown HIV status who were in labor. In the fourth hospital, blood samples from women in labor were sent to its 24-hour hospital laboratory for OraQuick rapid testing.

A total of 225 women were tested at the three hospitals using point-of-care HIV testing and 155 were tested at the hospital relying on the laboratory for testing.

The average turnaround time for getting the results was 45 minutes at the three hospitals using point-of-care testing. This was four times shorter than the 3.5-hour turnaround time at the hospital using laboratory testing. The OraQuick test itself takes just 20 to 30 minutes to run.

OraQuick rapid test results at each hospital were confirmed to be 100% accurate using standard lab tests, when needed.

With rapid testing, three pregnant women were identified as HIV-infected and drug therapy was initiated during labor and delivery. None of the infants became infected with HIV.

"The proportion of women now tested for HIV in prenatal care is rising," the CDC's Dr. Marc Bulterys told Reuters Health.

"Still there is a subset of women who never get tested before they come to labor and delivery and this is really our last opportunity to do testing and start antiretroviral prophylaxis to try to prevent transmission," he said.

The results of the MIRAD study show that point-of-care testing is "feasible," the investigators conclude. They urge hospitals to "assess the costs and benefits of implementing point-of-care HIV testing within their institutions."

SOURCE: Morbidity and Mortality Weekly Report, September 12, 2003.

NEW YORK (Reuters Health) - Screening HIV-positive patients for sexually transmitted diseases (STDs) may help protect their sex partners from getting the virus, new research suggests.

Although the Centers for Disease Control and Prevention(CDC) recommends STD screening for HIV patients, few HIV clinics in the US seem to practice such screening regularly.

In order to estimate the number of HIV infections that could be prevented by treating STDs, Dr. Thomas A. Farley, from Tulane University in New Orleans, and associates first determined how common gonorrhea and Chlamydia infection were among HIV patients at a public clinic.

One third of the patients were tested for STDs at least once between June 2000 and May 2001, the authors report in the Journal of Acquired Immune Deficiency Syndromes. Among those tested, 1.7 percent had gonorrhea and 2.1 percent had Chlamydia infection.

Based on statistical calculations of how STD treatment reduces HIV spread, the researchers estimated that treatment of every patient with gonorrhea or Chlamydia would prevent nearly 10 sex partners from becoming infected with HIV.

"HIV-infected persons in care should be screened routinely for STDs," Farley told Reuters Health. "Physicians should assume HIV-infected persons remain sexually active, and physicians cannot rely on patients' complaints of symptoms to determine whom to test."

Many opportunities to prevent HIV infection are being missed, by not counseling patients to reduce risky sexual behavior "and by not screening them for curable STDs," Farley added.

"STD screening and counseling should be a central part of medical care for HIV-infected persons, because of the benefits to their partners," he stressed.

SOURCE: Journal of Acquired Immune Deficiency Syndromes, August 15, 2003.