A first-of-its-kind comprehensive map of the network of connections in the human immune system has uncovered previously unknown interactions and filled in gaps in understanding why approved immunotherapies work in some patients and not others.

The immune system map makes it possible to see the impact different diseases have on the immune system as a whole and to assess how drug molecules interact with cell surface receptor proteins of immune cells. This capability is expected to point the way to new drug targets.

Immunotherapies can be incredibly effective in some patients and not in others, said Gavin Wright, professor of microbial biochemistry at York University, U.K., who is senior author of a paper in Nature published on August 3, describing the drawing up of the map.

"Our research, a culmination of over two decades of work, could hold the key to understanding why these treatments are more effective in some groups and how they could be adapted to ensure as many people as possible can benefit from them," Wright said.

Although the interaction networks involving secreted proteins have been systematically catalogued, in large part the interactions between cell surface proteins have not. That is because membrane-embedded proteins have weak binding affinities and are not tractable with classic biochemical approaches.

One consequence is that for many clinically important immune receptors, the physiological ligand is unknown.

The lengthy process of compiling the map involved isolating and investigating a near-complete set of all the cell surface proteins via which immune cells physically link together, and then using computational analyses and a mathematical model to detail all the cell types and the messages flowing between them.

Interactions of the hundreds of surface proteins on each immune cell are often extremely transient and dynamic, requiring the development of specialized methods to track them and assemble an accurate map.

The researchers used a high-throughput surface receptor screening method to systematically map the direct protein interactions across a recombinant library encompassing most of the surface proteins on human leukocytes.

To enable a systemic survey of the surface protein interactions between immune cells at the scale of whole cell surface proteomes, the researchers first developed a method for testing binary interactions of all possible combinations of recombinant surface proteins, including detecting even those of low affinity.

This simultaneously addresses several limitations of previous methods to make it possible to screen hundreds of thousands of interactions while consuming minute amounts of protein.

The researchers next tested all possible protein pairings in human cells. Positive interactions from this primary screen were retested in a secondary screen to give a final matrix of reproducible interactions.

Benchmarking this against hand-curated interactions in the published literature showed the screen had captured most of these. In addition, 28 new interactions were found.

One of these was an endogenous nontumor ligand for the previously orphaned immune checkpoint receptor Vista (V-domain Ig suppressor of T-cell activation), a potent negative regulator of T-cell function that is expressed in leukocytes.

Overall, 57% of the binding pairs uncovered are unique, without either of the proteins having another binding partner. That exclusivity was particularly common among proteins that are considered to have primarily signaling roles, while the largest interconnected group featured integrins and other adhesion molecules.

The researchers validated their discoveries in a number of ways, with all of the top-ranked interactions supported by at least one additional method.

Building on these findings, a mathematical model was used to infer the relative frequencies at which human immune cells physically interact. These inferences have sufficient precision to be consistent with published empirical measurements, they say.

Organ-level interactions

The immune system traverses many different organs, each of which may be key to understanding the biological role of an interaction. To put their findings in an organ-level context, the researchers created an interactive atlas that charts where these receptors and ligand pairs have been detected across single cell expression datasets of human tissues.

The atlas allows multiple kinds of analyses, ranging from summarizing the overall cellular connectivity of different tissue immune populations, to inferring which cell-cell pairs are capable of carrying out a particular receptor interaction.

The systemic multi-organ atlas makes it possible to determine whether immune receptor interactions proceed through shared structures or are distinct between tissues.

A broad range of pathological conditions involve dysregulation of the immune system and the researchers say the tools they have developed can be used to look at how the physical interactions they have catalogued between immune cells are changed in disease.

They incorporated paired disease and available reference samples in their atlas to generate hypotheses on which interactions may differentially appear in diseased states.

As one example, phagocyte populations could be seen to shift a large fraction of their total cellular contacts within the tumor microenvironment of kidney samples.

The newly discovered interactions also could be integrated with known signaling pathways to infer cellular communication pathways that are differentially active in disease states.

The map is "an incredible tool" that can be used to highlight proteins and pathways that would be good drug targets, according to co-author Berend Snijder, of the Institute of Molecular Systems Biology at ETH Zurich, Switzerland. "It can also give insight into whether a drug will have an impact on other pathways, which can cause side effects. All of this information may help in the development of new therapies," Snijder said.

The map is now freely available, providing a template for future studies in disentangling cellular circuits in immunity.

"Meticulously isolating and analyzing every immune cell and their interactions with others has given us the first map of the conversations between all the immune cells in the human body," said first author Jarrod Shilts of the Wellcome Sanger Institute in Cambridge, U.K.

"This is a huge step in understanding the inner workings of the immune system and will hopefully be used by researchers [...] to help develop new therapies," Shilts said.