Less Is More: High-Affinity Antibodies Block Blood-Brain Barrier Conduit

A high-affinity transferrin receptor antibody (red) gets waylaid in LAMP1-positive lysosomes (green) in brain endothelial cells (nuclei, blue). Not shown: A lower-affinity antibody avoided the lysosomes. [Courtesy of Ryan Watts.]

Ever been stuck waiting for a stalled ski lift? Often the problem is hesitant newbies who flounder when disembarking at the top. A similar kerfuffle can happen when trying to shunt antibodies into the brain. The transferrin receptor (TfR) transport system, which can be co-opted to transport cargo across the blood-brain barrier, shuts down if the cargo does not readily hop off on the brain side of endothelial cells, according to a paper in the January 27 Journal of Experimental Medicine. The results have implications for researchers hoping to use the transferrin receptor to sneak therapeutic cargo across the barrier—a requisite for Alzheimer’s and other neurodegenerative diseases. The new work suggests that antibodies that hitch a ride on the receptor need to bind loosely and dissociate easily when they arrive in the brain if the antibodies are to reach the expected dose where it matters most: right near the target.

Scientists have tried for decades to use TfR to deliver antibodies and other therapies into the brain. They have been stymied by inefficient release of the cargo on the brain side of endothelial cells that line the barrier (Moos and Morgan, 2001). “Understanding the cell biology and every in and out of this approach will be essential to its success,” said Ryan Watts at Genentech in South San Francisco, who led the work.

The new study came out less than a month after Per-Ola Freskgård and colleagues at Genentech’s parent company, Hoffman-La Roche in Basel, Switzerland, published a study debuting Brain Shuttle, their version of a TfR targeting system, (see Jan 2014 news). Both Watts’ and Freskgård’s studies show that the degradation of TfR antibodies in the lysosomes—the cell’s waste disposal vesicles— prevents them from crossing into the brain. However, the authors came to different conclusions on why some antibodies get snagged.

Ever since previous work from Watts’ lab showed that decreasing the affinity of TfR-specific antibodies upped their release into the brain, the researchers have been focusing on monovalent antibodies. These bind more weakly than standard bivalent antibodies that recognize the same antigen with both arms. The researchers created an antibody that bound TfR with one arm and β-secretase-1 (BACE-1) with the other. This bispecific antibody entered the mouse brain better than a bivalent TfR-specific antibody, and it reduced amyloid levels. The researchers proposed that the lower affinity of the bispecific antibody boosted its entry (Yu et al., 2011).

To determine whether affinity or valency, i.e., the number of TfR arms, explained the differences in transport efficiency, the researchers tinkered with the anti-TfR portion of the bispecific antibody. They found that lowering its affinity produced a therapeutic molecule that crossed the blood-brain barrier more readily than their original construct (see May 2013 news). For the current study, Watts wanted to find out why. Were the high-affinity antibodies simply getting stuck, degraded, or were they dismantling the whole transport system by preventing TfR from recycling to the cell surface? he wondered.

To answer these questions, the investigators tested whether antibodies altered natural TfR dynamics. They found that four days after injecting mice with high-affinity bispecific antibodies, brain TfR levels were down by more than half. In contrast, mice that received low-affinity antibodies maintained nearly normal levels of the receptor. This suggested that the high-affinity antibody somehow triggered the receptor’s degradation.

To find out how, first author Nga Bien-Ly and colleagues first tracked the fate of antibodies within endothelial cells in culture. The high-affinity antibody comingled with lysosomes more often than the low-affinity ones (see image below). Next, the researchers peered through implanted cranial windows and found the same distribution in the vasculature of live mice. The results point to a funneling of the high-affinity antibodies into lysosomes, where they are degraded rather than being transported across the cell.

For the grand finale experiment, the group tested whether a dose of the high-affinity antibody would effectively shut down the TfR transport system. The researchers treated mice with 50mg/kg, waited two days, and then gave the animals an equal dose of the low-affinity antibody. Pre-dosing cut the subsequent brain uptake of the low-affinity antibody by 75 percent. In contrast, an initial injection with the low-affinity antibody had no significant effect on subsequent uptake. This may explain why lower-affinity antibodies are better at getting cargo into the brain. “The goal is to have sustained exposure [to the antibody] over the dosing window,” Watts said. “If the receptor starts to get degraded, whatever antibody is still onboard can’t get into the brain.”

These experiments were elegant and elucidated the mechanism behind differences in TfR transport, David Holtzman of Washington University, St. Louis, wrote to Alzforum. Holtzman was not involved in the study. “From a translational perspective, understanding this mechanism should allow for further development of even better bispecific antibodies to facilitate even greater transport of a variety of cargoes into the CNS that have therapeutic potential for diseases such as AD,” Holtzman wrote.

In their recent paper, Freskgård and colleagues showed that attaching one TfR-specific Fab antibody fragment (sFab) onto the stem of a monoclonal antibody specific for Aβ delivered the complex into the brain with high efficiency, whereas attaching two fragments (dFab) did not. The investigators also detected dFab in lysosomes at a higher rate than sFab. This indicated that, like Watts’ high-affinity bispecific antibody, dFab was being destroyed before it reached the brain side of the endothelium. The authors concluded that the valency of the antibody, rather than affinity, accounted for the poor brain delivery. However, Watts emphasized that differences in affinity might account for Freskgård’s results. Researchers from Roche were unavailable for comment.

The sFab construct from Freskgård’s study had roughly the same affinity for TfR as Watts’ high-affinity bispecific antibody; the dFab affinity was even higher. Genentech’s theory would predict that either of these antibodies would deplete TfR in the brain. The Roche study reported no change in TfR levels after adding sFab to brain endothelial cells, but did not check for TfR changes in vivo. If levels of the receptor fell, it would raise not only efficacy concerns about sustained antibody exposure, but also potential safety concerns for iron transport, Watts said.

Though technically part of the same company, the two groups generated their results independently. “In a perfect world, it would be nice to compare all these approaches in the same experimental conditions,” said Robert Bell of Pfizer, who was not involved in either study. He was impressed by both approaches, and hoped the technology will move forward into humans. “This is a very complex problem,” Bell said. “They are showing here that this [TfR transport approach] is possible.”

Watts’ group is currently wrapping up monkey experiments with a bispecific TfR/BACE1 antibody. The results look promising, he said. In addition to BACE1, the scientists have set their sights on other neurodegenerative disease targets and are optimizing the affinities of TfR antibodies for future use in human studies.


Bien-Ly N, Yu YJ, Bumbaca D, Elstrott J, Boswell CA, Zhang Y, Luk W, Lu Y, Dennis MS, Weimer RM, Chung I, Watts RJ. Transferrin receptor (TfR) trafficking determines brain uptake of TfR antibody affinity variants. J Exp Med. 2014 Feb 10;211(2):233-44. Epub 2014 Jan 27 PubMed.

Yu YJ, Zhang Y, Kenrick M, Hoyte K, Luk W, Lu Y, Atwal J, Elliott JM, Prabhu S, Watts RJ, Dennis MS. Boosting brain uptake of a therapeutic antibody by reducing its affinity for a transcytosis target. Sci Transl Med. 2011 May 25;3(84):84ra44. PubMed.

Moos T, Morgan EH. Restricted transport of anti-transferrin receptor antibody (OX26) through the blood-brain barrier in the rat. J Neurochem. 2001 Oct;79(1):119-29. PubMed.

To view commentaries, primary articles and linked stories, go to the original posting on Alzforum.org here.

Copyright © 1996–2014 Biomedical Research Forum, LLC. All Rights Reserved.

blood brain barrier Brain Shuttle disease-ad drug delivery topic-preclinical transcytosis
Share this: