Research Briefs

By Andreas von Bubnoff

Researchers Make Immunogen That Can Bind to Precursors of Broadly Neutralizing Antibodies

Over the past few years, researchers have isolated many new broadly neutralizing antibodies (bNAbs) from chronically HIV-infected individuals. But only a small percentage of HIV-infected individuals can make bNAbs. It also takes years for them to develop, possibly because the precursor B cells that will eventually develop into the cells that produce the mature bNAbs have to first undergo a process called affinity maturation, during which the antibody sequences a person inherits in their germline accumulate mutations that increase the affinity of the antibody for antigens such as HIV Env (see Vaccines to Antibodies: Grow Up!, IAVI Report, July-Aug. 2010).

These mutations have been shown to be required for the ability of HIV-specific bNAbs to bind to and neutralize many different HIV strains. The first evidence of this came when Dimiter Dimitrov, a senior investigator at the National Cancer Institute in Frederick, Maryland, and his colleagues found that the unmutated precursors of bNAbs such as 2F5 cannot bind HIV Env (1). This led Dimitrov to suggest that an HIV vaccine should contain immunogens that can bind the unmutated precursors of bNAbs to kick-start the affinity maturation process because HIV-uninfected people only have B cells that express such precursors (2).

Now, Barton Haynes, a professor at Duke University Medical Center, and colleagues report that they have made such an immunogen by removing part of the sugar groups from the HIV gp140 Env protein (3). Unmutated precursors of the HIV-specific bNAbs 2F5 and 4E10 can bind this sugar-depleted Env but not the normal Env that has the sugars in place.

In rhesus macaques, the sugar-depleted Env protein was also a better immunogen than the normal Env, inducing higher levels of antibodies that bind to the membrane proximal external region (MPER; the part of HIV recognized by 2F5 and 4E10) a few weeks after immunization. The study is the first, Haynes says, to show that an HIV immunogen “that bound to the unmutated ancestor better or at all just happened to be the one that was a better immunogen.”

“This is the first published report of an immunogen capable of binding the putative germline predecessor of two broadly neutralizing antibodies,” says Dimitrov, who was not involved in the study. “Even more importantly, they found this newly designed vaccine immunogen, which binds the germline, also shows high immunogenicity in monkeys.”

Haynes says the study shows that “there are ways to design immunogens that can bind to germlines that you would expect would be able to trigger and set off this pathway of B cell maturation that we are trying to induce.”

While the results are promising, immunization with the sugar-depleted Env didn’t induce HIV neutralizing antibodies in the vaccinated macaques, Haynes says, possibly because the immune systems in the macaques only had a few weeks after immunization to respond, too short for affinity maturation to take place. And even if there was enough time for bNAbs to be induced, they might be deleted by the immune system, Haynes adds, because some bNAbs are polyreactive and bind many different antigens including some that are self. “We made the hypothesis years ago that this polyreactivity might predispose these antibodies to be downregulated by the immune system because they cross-react with self,” Haynes says.

To show that this is the case, Laurent Verkoczy at Duke University and Haynes did a second study in which they made mice that only expressed the mature bNAb 2F5 in all of their B cells, and found that the immune systems of these mice indeed deleted B cells that expressed mature 2F5 on their surface from the bone marrow (4). However, when they took the antibody-expressing B cells from the bone marrow of the mice and separated them from the cells that normally eliminate B cells that make polyreactive antibodies, they found that a few of the 100 B cells they characterized in detail made 2F5 antibodies.

This suggests that in most people, even if they develop 2F5 or other types of bNAbs, their own immune systems may eliminate most of them. Why the immune systems of the people who can make these types of bNAbs don’t eliminate them is unclear, however. “Some people for whatever reason don’t effectively eliminate these cells and they are allowed to eventually go on and mature,” Haynes says, adding that they are trying to find out why.

It is also unclear how to keep the immune systems of vaccinees from eliminating bNAbs that are induced by a vaccine. Haynes hopes it might be possible to induce affinity maturation pathways that lead to types of bNAbs that are not subject to this elimination. To test this, Haynes and colleagues are immunizing animals with immunogens that can bind to unmutated precursors of bNAbs to see if the resulting antibodies are eliminated or not.

In addition, Haynes is working to understand how bNAbs develop from their precursors in HIV-infected individuals to identify ones that are not subject to elimination, and plans to isolate unmutated and partially mutated precursors of bNAbs from the bone marrow of HIV-infected individuals to use them as templates to study additional ways Env can be changed to allow these antibodies to bind. 

1. Biochem. Biophys. Res. Commun. 390, 404, 2009
2. Viruses 1, 802, 2009
3. PLoS Pathog. 7, e1002200, 2011
4. J. Immunol. 187, 3785, 2011


Stem Cell-Like Memory T Cell Identified in Humans

If a person is infected with a virus or pathogen, their immune system responds by generating effector CD4+ and CD8+ T cells that are specific for that pathogen. Many of them become memory CD4+ and CD8+ T cells, which persist long after the infection is over. Memory and effector T cells are derived from naive T cells specific for millions of different antigens, most of which will never be seen. When a naive T cell encounters the right antigen that binds to its T cell receptor (TCR), it starts dividing, and can become either a central memory T cell that circulates in blood and lymph nodes and serves as a source for additional T cells, or an effector memory T cell, which is found in tissues such as the mucosa.

Scientists have long suspected that some memory T cells can persist for an especially long time and multiply and regenerate other types of memory T cells especially well, but the exact identity of these cells was unclear, says Mario Roederer, a senior investigator at the Vaccine Research Center at the National Institute of Allergy and Infectious Diseases.

Now, Roederer and colleagues identified a new type of memory T cell in humans that can regenerate itself and other T cell types better than effector and central memory T cells, making it similar to stem cells (1). The identification of these so-called stem cell memory (SCM) cells might give researchers a better tool to develop vaccines that can induce long-lasting CD4+ and CD8+ memory T-cell response.

To identify the cells, Roederer and colleagues treated human naive CD4+ and CD8+ T cells in vitro with TWS119—a drug that keeps cells from differentiating too much and has been shown to turn naive CD8+ T cells from mice into stem cell-like CD8+ memory T cells (2). They found that these cells expressed unique cell surface markers, differentiating them from known types of memory T cells (effector memory and central memory T cells). Using these markers, Roederer and colleagues were able to identify SCM cells among white blood cells from healthy people.

Of all memory T cell types, SCM cells are most similar to stem cells because they have the best capacity to regenerate themselves and other memory T cell types, Roederer says. “When they are stimulated, they can divide many times and generate a very large population of T cells,” he says. “And they can persist for a very long time without needing any antigenic stimulation.”

“What’s unique is that they found a subset that seems to be more stem-like than the remaining memory cells,” says Louis Picker, a professor of pathology, molecular microbiology, and immunology at Oregon Health & Science University, who was not involved in the study. “[Their] function appears to be long-term self renewal of the memory response. This population may underlie the ability of memory responses to last for decades.”

The identification of SCM cells could explain previous observations that suggested memory T cells can regenerate themselves and other T cell types surprisingly well, Picker says, referring to a 1997 study where he and his colleagues described a man who had lost all of his T cells as a result of immunosuppressive therapy after a liver transplant. Once the immunosuppressive drugs wore off, the man lived for many years with all of his memory T cells derived from the liver transplant. This means that all of his memory T cells had been regenerated from memory T cells in the donor liver, without any help from naive T cells (3).

Because SCM cells can multiply better than other memory T cell types and can develop into CD8+ T cells that can kill other cells, Roederer and colleagues also investigated whether SCM cells could kill cancer cells. They made SCM cells that produced a mesothelin tumor-specific antibody connected to their TCR. Recognition of a mesothelin tumor cell by this antibody could activate the TCR and the SCM cell could then kill the tumor cell. Indeed, they found that injecting these modified SCMs into mice with a human mesothelin tumor caused the tumor to shrink. All of the mice survived, whereas mice that were treated with central or effector memory T cells that had been modified in the same way died after a few weeks. The reason SCM cells were better at killing the tumor is because they generated 10-50 times more cells than the other memory T cell types, Roederer says. “The reason they protected the mice is simply because they were able to expand to much larger numbers and then differentiate into effector cells and then kill the tumor,” says Roederer.

Roederer says identification of SCM cells is relevant for the development of vaccines that induce a life-long vaccine-specific CD4+ and CD8+ T-cell memory response in the absence of antigen, which would translate into the ability to control the virus long after vaccination. “To generate life-long immunity, you need to generate these stem cell memory cells that are specific to the vaccine,” Roederer says. “They are the ones that can persist for what we think is forever, whereas the central memory and effector memory cells are much more dependent on the presence of antigen, and their numbers wane if antigen is not present.”

Currently, Roederer is studying how many SCM cells are generated in rhesus macaques using different vaccine strategies. “I would hypothesize that a vaccine that generates a good number of SCMs would generate better durable long-term memory and so that’s the hypothesis that we are going to test,” Roederer says. He also plans to look for SCM cells in humans who received different HIV vaccine candidates.

1. Nat. Med. 17, 1290, 2011
2. Nat. Med. 15, 808, 2009
3. Exp. Hematol. 25, 147, 1997