류크 오닐(생화학, 면역학자) Cell에 선천면역 50년사 기고 논문

[문형철]

https://pmc.ncbi.nlm.nih.gov/articles/PMC11060700/

**Susan Carpenter·Luke A.J. O’Neill의 Cell 리뷰(2024)**는

Cell 창간 50주년을 기념해,

지난 50년간 선천면역이 ‘주변부’에서 면역학의 중심 무대로 올라온 과정을 정리한 논문입니다.

적응면역(림프구의 클론 선택·재조합)만 보면 면역 이해가 한계(asymptote)에 부딪힌다는

Janeway의 경고에서 출발해,

PRR·사이토카인·인플라마좀·핵산 감지·훈련된 면역·면역대사까지 한 줄기로 연결합니다.

핵심 메시지:

선천면역은 더 이상 “비특이적 장벽”이 아니라,

병원체·손상 신호를 읽고 적응면역을 켜고, 항상성과 병리를 동시에 조절하는 시스템이며,

이제 감염·자가염증·암에서 치료 표적이 되고 있다.

---

1. 왜 “코페르니쿠스적 전환”인가

1970년대까지 면역학의 중심은 항체·T세포였다.

피부·점막 장벽, 리소자임, 위산, 보체 정도는 “거친 1차 방어”로 취급됐다.

1989년 Charles Janeway가

**pattern recognition receptors(PRRs)**가

**pathogen-associated molecular patterns(PAMPs)**를 인식한다고 가설을 세우면서 판이 바뀐다.

“어쥬번트의 더러운 비밀”도 여기서 풀리기 시작한다.

이후 Polly Matzinger의 danger 모델이 더해져,

미생물 산물(PAMP)뿐 아니라 손상 조직 산물(DAMP)도 같은 수용체 체계를 탄다.

지금은

골수계 세포(대식세포, DC, 호중구)와 상피·내피 세포까지

PRR을 갖고,

항원제시·공동자극분자·사이토카인으로 T/B세포를 직접 조형한다.

박스 1. 선천면역 기억(Innate immune memory)

선천면역 기억은

선천면역계가 일정 기간 동안 제한된 형태의 기억을 갖는 것처럼 보이는 현상이다.

적응면역 기억이 보통 개체 평생 유지되는 것과 달리,

선천면역 기억은 수개월에서 1년 정도 지속된다.

β-glucan과 LPS처럼 자극의 종류가 다르면 골수계 세포에서 서로 다른 훈련 프로그램이 유도된다.

이 변화는 적응면역 세포에서 보이는 유전자 재조합이 아니라 후성유전적 변화로 일어난다.

훈련된 면역 개념은, 결핵용으로 만든 BCG 백신이 바이러스를 포함한 여러 감염원에 대해 더 넓은 방어를 제공하는 현상을 어느 정도 설명한다. BCG는 적응면역이 없는 SCID 마우스에서도 치명적인 칸디다 감염을 방어하는데, 이는 이 기억이 선천면역 축 안에서 일어난다는 점을 보여 준다.

마우스 연구에서 단핵구와 대식세포가 훈련을 주도하며, 훈련은 말초를 순환하는 세포뿐 아니라 골수 전구세포에서도(중심 훈련, central training) 일어난다.

훈련에 기여한다고 보는 요소는 세포의 3차원 구조, 인핸서의 H3K4me1, 프로모터의 H3K4me3 같은 후성유전 재프로그래밍, 그리고 lncRNA 발현 유도이다. 대사 변화도 관여한다. β-glucan 훈련 때는 호기성 해당과정이 증가한다. 리간드마다 올라가는 대사체가 달라서, glucan은 푸마르산, LPS는 숙신산을 유도하며, 각각 훈련 신호를 다르게 만든다.

식이, 환경 노출, 병적 노출 등 많은 요인이 훈련에 영향을 준다. 이 경로를 치료적으로 쓰려면, 훈련을 지배하는 정확한 분자 기전을 더 밝혀야 한다.

---

C Type Lectin ReceptorsReceptorLocalizationLigandFunctionReferenceInflammasomesReceptorLocalizationLigandFunctionReferenceNucleic Acid SensorsReceptorLocalizationLigandFunctionReferenceNod Like ReceptorsReceptorReceptorReceptorReceptorReceptorToll like receptorsReceptorLocalizationLigandFunctionReference

Dectin 1 and 2	Cell surface	B-Glucan and a-Mannan	Recognition of unique components in fungai	32
Mincle	Cell surface	Spliceosome associated protein (SAP130) and Malassezia (fungi)	SAP130 is a component of U2 snRNP released from necrotic cells and activates Mincle	32,95
DC-SIGN	Cell surface	mannose oligosaccharides or fucose-containing Lewis-type antigens	Functions as a PRR against microbes in addition to functioning as a cell adhesion receptor	96

2. 50년 타임라인 (논문 Figure 1의 골격)

대략 이런 흐름이다.

• 1973: 수지상세포 발견
• 1980년대: IL-1, TNF 클로닝, NF-κB, 호중구 respiratory burst(NADPH oxidase)
• 1988–1991: IL-1R 클로닝 → Drosophila Toll과 TIR 도메인 상동성
• 1996–1998: 포유류 TLR4, LPS 수용체로 확정 (C3H/HeJ 돌연변이, Beutler; 2011 노벨상 Hoffmann·Beutler)
• 2000년대 초: NLR/인플라마좀(Martinon·Tschopp 2002), AIM2, RIG-I/MDA5
• 2013: cGAS가 세포질 DNA 센서, cGAMP–STING–I형 IFN
• 2010년대 이후: 훈련된 면역(trained immunity), 면역대사(Warburg, succinate, itaconate), NLRP3 억제제 등 약물

논문은 이 발견들이 Cell에 많이 실렸다는 점도 강조한다.

---

3. PRR 가족 — 어디서 무엇을 보는가

PRR은 유전체에 고정된(germline-encoded) 수용체다. 위치별로 정보가 다르다.

계열대표위치리간드 예시결과

TLR	TLR4, 2/6, 5 / TLR3,7,8,9	막·엔도좀	LPS, 지질단백, flagellin / dsRNA, ssRNA, CpG	NF-κB, IRF3, I형 IFN
NLR / 인플라마좀	NLRP3, NLRC4–NAIP, AIM2, Pyrin	세포질	K⁺ 유출·미토 손상, flagellin·T3SS, dsDNA, RhoA 불활성	caspase-1 → IL-1β/IL-18, gasdermin D 구멍 → pyroptosis
RLR	RIG-I, MDA5	세포질	5′-ppp RNA, 긴 dsRNA	I형 IFN
cGAS–STING	cGAS	세포질 (핵에도)	dsDNA	cGAMP → STING → IFN
CLR	Dectin-1/2, Mincle, DC-SIGN	막	β-glucan, mannan 등 진균	Syk–Card9 등

마우스 전용 TLR11/12/13(profilin, 23S rRNA)과 NAIP–NLRC4는 이후 정정에서 표가 보강됐다.

TLR4 신호가 가장 자세하다.

LBP–CD14–MD2–TLR4 복합체가 LPS를 잡고, MyD88/MAL → IRAK–TRAF6 → NF-κB(Myddosome)와, 엔도좀에서 TRIF/TRAM → TBK1–IRF3(Triffosome)가 동시에 돈다. LPS 치사에는 TLR4가 caspase-11(사람 caspase-4/5)을 유도해 pyroptosis를 켜는 경로가 핵심이라는 점도 짚는다.

---

4. 인플라마좀과 “세포질 신성”의 붕괴

인플라마좀은

세포질에서 조립되는 다단백질 복합체다.

활성화 caspase-1이 pro-IL-1β/IL-18을 자르고,

gasdermin D를 잘라 막을 뚫어

pyroptosis를 일으킨다.

NLRP3는 가장 연구가 많다.

이온 흐름, 리소좀·미토콘드리아 손상, 핵산, 삼투 스트레스 등 “너무 많은 것”이 켜서,

직접 리간드가 아니라 세포 항상성 붕괴의 통합 센서에 가깝다.

2단계(프라이밍 = NF-κB로 구성 성분 유도 + 2nd hit)가 전형적이다.

CAPS 같은 유전성 발열 증후군이 활성화 돌연변이의 임상 증거다.

Vishva Dixit의 표현대로, DNA가

“세포질의 신성(sanctity of the cytosol)”을 깨면 선천면역이 켜진다.

AIM2는

DNA → 인플라마좀, cGAS는 DNA → cGAMP → STING → I형 IFN. 자기 DNA 유출은 SLE 등과 연결된다.

---

5. 보체, 훈련된 면역, 면역대사 — “오래된 것”의 재발견

보체: 고전/대체/렉틴 경로뿐 아니라, 세포 안 complosome(C3, C5)이 대사·생존·IL-1β에 관여한다. 장 상피가 국소 C3를 만들어 병원체는 막고 공생균은 살린다는 최근 Cell 논문도 인용한다.

훈련된 면역(Box 1): 골수계 세포의 후성유전 기억(H3K4me1/me3, lncRNA, 해당과정·푸마르산/숙신산). BCG가 SCID 마우스에서도 칸디다 방어를 주는 이유가 여기 있다. 수명은 수개월~1년 수준이며, 말초 단핵구뿐 아니라 골수 전구세포(central training)에서도 일어난다.

면역대사: LPS 자극 대식세포는 해당과정↑, PKM2 이량체·HIF-1α, 숙신산 산화 → 복합체 I 역전자전달 → mtROS, 푸마르산 수화효소 억제 → mt-dsRNA → RIG-I/MDA5, IRG1 → itaconate(항염증). 선천면역 = 대사 재배선이다.

진화 박스: 성게·식물의 TLR/NLR 폭증, 박테리아 cGAMP/STING, 박쥐의 약한 NLRP3, 말 TLR4의 독특한 LPS 인식 등 “종마다 다른 선천면역 설계도”를 보여 준다.

---

6. 임상·치료 — 이미 쓰이고, 앞으로

더 쓸 자리

이미 수백만 명이 TNF·IL-1·IL-6 차단으로 RA, IBD, 건선 등을 치료받는다.
다음 단계는 수용체와 바로 아래 신호다.

• NLRP3 억제제
• STING/cGAS 조절(암 면역 vs 자가염증)
• TLR 아고니스트/길항제(백신 어쥬번트, 패혈증)
• 훈련된 면역을 백신·감염·암에 이용
• itaconate·숙신산 등 대사 노드

저자들은 “이 분야는 아직 유아기”라고 본다. 후성유전 조절, 비암호화 RNA, 조직 상피의 PRR, 어쥬번트 설계가 남은 큰 숙제다.

Cell

. Author manuscript; available in PMC: 2025 Apr 25.

Published in final edited form as: Cell. 2024 Apr 25;187(9):2030–2051. doi: 10.1016/j.cell.2024.03.036

From Periphery to Center Stage: 50 years of Advancements in Innate Immunity

Susan Carpenter 1, Luke AJ O’Neill 2

• Author information
• Copyright and License information

PMCID: PMC11060700  NIHMSID: NIHMS1987583  PMID: 38670064

The publisher's version of this article is available at Cell

This article has been corrected. See Cell. 2024 Oct 18;187(23):6780.

This article has been corrected. See Cell. 2024 Jul 16;187(16):4429.

Abstract

Over the past fifty years in the field of immunology, something of a Copernican revolution happened. For a long time, immunologists were mainly concerned with what is termed adaptive immunity, which involves exquisitely specific activities of lymphocytes. But the other arm of immunity, so-called “innate immunity”, had been neglected. To celebrate Cell’s 50th anniversary we have put together a review of the processes and components of innate immunity and trace the seminal contributions leading to the modern state of this field. Innate immunity has joined adaptive immunity in the center of interest for all those who study the body’s defenses, as well as homeostasis and pathology. We are now entering the era where therapeutic targeting of innate immune receptors and downstream signals holds substantial promise for infectious and inflammatory diseases and cancer.

In Brief:

Understanding of innate immunity has expanded enormously in recent history. In light of Cell’s 50th Anniversary, this review traces the development of the innate immunity field from the 1970s through to today, highlighting how advances in knowledge now lead to therapeutic strategies.

Introduction

The term ‘innate’ is defined as something you are born with, but in the context of immunity, it has been used to denote the part of the immune system which is present from the start of an organism’s life, and which doesn’t undergo genetic rearrangement in the course of an infection. Historically, this contrasts with the term ‘adaptive immunity,’ which involves the processes conducted by the specialized immune cells that do undergo genetic rearrangement to “adapt” and address threats to the body, known as lymphocytes. In adaptive immunity the immune system responds to the invading pathogen by allowing clones of lymphocytes called B and T cells to expand, such that there are more of them present after the infection. This is also how vaccines work, by driving expansion of antigen-specific T and B cells in a controlled and safe way, leaving the immune system ready for a rapid response should an infection occur later. The term ‘immunity’ itself refers this capacity for adaptation, coming from the Latin word ‘immunis’, meaning ‘exempt’ from further infection.

We now know that the body’s defense to pathogens involves many cells and factors in addition to lymphocytes, and these have generally been grouped under the term “innate immunity”. However, the so-called “innate” processes can generate a distinct type of memory in myeloid cells that is largely epigenetic1.This has been called “trained immunity” to distinguish myeloid cell memory from bona fide lymphocyte memory (see Box 1). This has required us to recognize that all white blood cells, known generally as leukocytes, hold roles in immune responses, with lymphocytes conducting “adaptive immunity” and myeloid cells conducting “innate immunity”. The seeming oversimplification of terms reflects the fact that back in 1974, the component parts of innate immunity seemed crude and unspecialized. Unlike the specificity that was becoming apparent in antibodies and then T cell receptors, innate immune factors were much broader in their protective effects, and not specific to one pathogen or antigen. They involved such things as the barrier function of skin and epithelia at mucosal sites (which keep many types of microbes from penetrating tissues), and such substances as mucus to trap invading organisms, lysozyme in fluids to break down bacteria, and the acidity of the stomach. The most sophisticated component of innate immunity on the radar in the 1970s was complement, a series of proteins activated in response to bacteria or an antigen/antibody complex, which leads to the lysis of bacteria.

Box topic 1: Innate immune memory.

Innate immune memory is a phenomenon in which the innate immune system appears to hold a type of limited memory for a defined period of time. Unlike adaptive immune memory, which typically lasts for the lifetime of the organism, innate memory lasts more on the order of months to a year (reviewed here218,219). Different stimuli, for example β-glucans and LPS, induce different programs of training in myeloid cells, which are induced through epigenetic changes rather than gene recombination observed in adaptive immune cells. The concept of trained immunity goes some ways to explaining how vaccines like BCG (bacillus Calmette–Guérin), which is designed against tuberculosis, can provide broader coverage to a host against a range of infectious microbes, including viruses. The BCG vaccine has been shown to provide protection against lethal candidiasis in SCID mice that lack an adaptive immune system, highlighting the importance of this type of memory occurring within the innate arm of the immune system218. Many studies indicate that monocytes and macrophages play key roles in driving the training in mouse studies, with training occurring in progenitor cells in the bone marrow, resulting in central training in addition to cells circulating out in the periphery. The cells’ 3D architecture, epigenetic reprogramming including H3K4me1 marks at enhancers, H3K4me3 marks at promoters, and induced expression‎ of lncRNAs are all believed to contribute to the processes of training. Altered metabolism also plays a role, with an increase in aerobic glycolysis present during training with β-glucans. Different ligands induce different metabolites with glucans inducing fumarate and LPS inducing succinate. Each can influence the overall signals induced during training. Many things can influence training, from diet to environmental and pathological exposure. Much work remains to be done at unraveling the exact molecular mechanisms governing training in order to be able to gain clinically impactful insights into harnessing these pathways for therapeutic benefit.

While complement has roots at the very beginning of the field of innate immunity there has been a resurgence of interest in recent years. There are three distinct pathways of the complement system, the classical, alternative, and lectin pathways. The components of the complement system are produced by the liver and are involved in the detection of blood borne pathogens, activation of inflammation and clearance (reviewed in2,3). Recently the description of the so-called “complosome” has put a spotlight back on the basic functions of the complement system. The complosome represents intracellular components of the complement system that are involved in basically all physiological processes inside cells of the immune system, including metabolism, cell survival and gene regulation4. For example, cell intrinsic expression‎ of C3 and C5 in monocytes and macrophages is involved in the production of IL1β5,6. Just in the past year there have been two studies published in Cell providing new insights into this system. The first study by Desai et al., shows how the C5a component of complement plays critical roles in driving phagocyte survival and effector functions during fungal infections7. The second study by Wu et al., implicates complement as a key regulator of gut health. They show that cells of the gut locally produce complement component C3, which provides protection against invading microbes while saving commensal microbes and ensuring a healthy gut8. The complement system was discovered over a century ago and we are still only learning about its regulatory properties. This work emphasizes the importance of eval‎uating, reassessing, and putting into context of the bigger picture what we know and appreciating what we have still to learn about the complexities of the innate immune system.

Back in the early days of innate immunity, biologists interested in inflammation began revealing other complex components and processes beyond the complement system. The role of the neutrophil in host defense had been defined, including a description of the respiratory burst driven by NADPH oxidase leading to bactericidal hydrogen peroxide production9. From the 1980s on, a large number of intercellular messenger molecules called cytokines were described, including the pro-inflammatory cytokines IL1 and TNF10, which were shown to drive a profound increase in inflammatory gene expression‎ in target cells, via such transcription factors as NF-κB11. Cytokines were shown to induce and control physiological processes such as fever and vasodilation, as well as processes like leukocyte adhesion and migration. Cytokine names are generally descriptive, with the name interleukin signifying inter for “between”, and leukin referring to leukocyte, followed by a number to designate each messenger. Chemokines, which are chemotactic cytokines that attract immune cells to site of infection, are named in a similarly standardized and numbered way. Importantly, cytokines were found to be a key link from innate immune cells such as macrophages and dendritic cells to adaptive cells, with the messages from myeloid cells driving both the differentiation and anti-pathogen effector functions of particular subsets of T cells.12.

A true ‘innate immunity revolution’ began circa 1989 when Charles Janeway hypothesized the existence of what he termed “pattern recognition receptors” (PRRs), defined as receptor proteins that recognize “pathogen associated molecular patterns” (PAMPs), essentially a system by which the host organism’s proteins can detect classes of molecules that are not usually present, and thus indicative of an infection13. The terms PRR and PAMP are now firmly embedded in the immunology lexicon and refer to a large variety of host receptors and pathogen-associated factors (see Table 1). In the article he wrote in 1988, Janeway used the phrase ‘approaching the asymptote’, in which he said if immunologists only concerned themselves with adaptive immunity, their knowledge of the immune system would reach an asymptote or limit. This acted as something of a rallying cry for immunologists, many of whom moved into innate immunity seeking PRRs and how they might work. Janeway also stated that immunologists had no idea how adjuvants worked, the ‘dirty little secret’ needed for vaccines to elicit their effects. It had been known for decades that simply injecting an antigenic protein into an animal to raise an antibody didn’t provoke a strong immune response. Things like complete Freund’s adjuvant, (heat-killed mycobacteria in paraffin oil) or alum (aluminum hydroxide) needed to be combined with the antigen to garner a strong response, and no-one knew why. The identification of PRRs and PAMPs began to clarify how microbial components in adjuvants could stimulate immune responses. But how did non-microbial adjuvants such as alum have similar effects?

Table 1:

Families of Pathogen Recognition Receptors

C Type Lectin ReceptorsReceptorLocalizationLigandFunctionReferenceInflammasomesReceptorLocalizationLigandFunctionReferenceNucleic Acid SensorsReceptorLocalizationLigandFunctionReferenceNod Like ReceptorsReceptorReceptorReceptorReceptorReceptorToll like receptorsReceptorLocalizationLigandFunctionReference

Dectin 1 and 2	Cell surface	B-Glucan and a-Mannan	Recognition of unique components in fungai	32
Mincle	Cell surface	Spliceosome associated protein (SAP130) and Malassezia (fungi)	SAP130 is a component of U2 snRNP released from necrotic cells and activates Mincle	32,95
DC-SIGN	Cell surface	mannose oligosaccharides or fucose-containing Lewis-type antigens	Functions as a PRR against microbes in addition to functioning as a cell adhesion receptor	96

Open in a new tab

The answer involves the idea promoted by pioneering innate immunologist Polly Matzinger, who posited that what the immune system actually responds to is danger.14 This was a revolutionary and somewhat controversial departure from the traditional view that the immune system exists entirely to discriminate “self” from “non-self”. PRRs turned out to be danger sensors, with “danger” in the form of microbial products (i.e., PAMPs) or the products of damaged tissue, which inflammation biologists had been studying for decades. These came to be called danger/damage associated molecular patterns (DAMPs), and today PAMPs, DAMPs, and other similar classes of molecules that serve as ligands for innate immune receptors represent a central concept in immunity.

By the 1990s, therefore, the scene was set for a major advance in immunology with the description of multiple PRRs, starting with the Toll-like receptors (TLRs),15,16 which in turn led to the discovery of NOD-like receptors (NLRs)17–20, C-type lectin receptors (CLRs)21, RIG-I-like receptors (RLRs)22,23, AIM2-like receptors (ALRs)24 and cGAS25, amongst others26 (Fig.1 and 2). Some were shown to drive multiple anti-microbial processes against bacteria, viruses, fungi and parasites, including promoting the all-important process of antigen presentation by myeloid cells to lymphocytes, thus providing the link between “non-specific” and “specific” branches of the immune response.

Fig. 1: Key discoveries in innate immunity over the past 50 years.

Open in a new tab

Here we outline the key discoveries in innate immunity through the decades. It begins with the discovery of dendritic cells in 1973 and moves through the key events of cloning cytokines (TNF) identification of the receptors (TLRs, inflammasome, cGAS) and mechanisms of action to the final development of drugs targeting the pathways (NLRP inhibitors).

Fig. 2: Molecular patterns induce a variety of cellular responses via distinct sensing pathways.

Open in a new tab

TLR signaling: plasma membrane bound or endosomally localized TLRs signal through adaptors and kinases situated in the cytosol to drive inflammatory cytokines and type 1 interferons (IFN) responses. Inflammasome: The inflammasome is a multiprotein complex formed in the cytosol in responses to microbial ligands or host danger signals. Activation leads to caspase 1 activity, cleavage, and release of IL1b or IL18. Nucleic Acid Sensing. DNA is recognized by cGAS within the cytosol, leading to the production of the second messenger cGAMP, which signals through STING to induce type 1 IFNs. RNA is sensed by RIG-1 or MDA5 to induced type I IFNs Noncoding Regulation. Many families of noncoding RNAs including lncRNAs, tRNA fragments, miRNAs and circRNAs have emerged as key regulators of biological including transcription, splicing and translation which can impact immunity.

While transient activation of these signaling cascades is critical for protection against infection, any persistent activation of these pathways can be detrimental and is associated with autoimmune and autoinflammatory conditions. The advances made in understanding the molecular and cellular players of immunology have in turn led to substantial clinical advances, notably in the targeting of cytokines in a range of autoimmune and autoinflammatory diseases, with many millions of patients benefiting in common diseases such as rheumatoid arthritis, inflammatory bowel disease, psoriasis and atopic dermatitis27.

In this review, we describe the main features of innate immunity uncovered in the past 50 years, with a general audience in mind. Many of the pioneering studies were published in Cell. We also speculate on whether the targeting of these processes will lead to further therapeutic advances. In many ways this field is still in its infancy, and we have a lot to learn about the intricacies of regulation within these pathways, particularly epigenetic regulation. Extra components critical to innate immunity might yet be discovered, and we have yet to harness the power of innate immunity for vaccine development and more effective therapeutics for infectious and autoimmune diseases.

Families of PRRs

PRRs are encoded within the genome like most proteins are, differentiating them from the lymphocyte receptors that undergo somatic recombination. These germline encoded receptors recognize conserved components critical and unique to microbes, such as components of bacterial and fungal cell walls, allowing for discrimination between “self” and categories of pathogens, as well as cell damage. A catalog of PRR families is outlined in Table 1. Once these receptors become activated, they initiate complex signaling pathways that result in the production of proinflammatory cytokines and antiviral genes28. These signals also trigger dendritic cell maturation, induce costimulatory molecules, and increase antigen presentation. This enables the innate immune system to directly activate and shape the downstream adaptive immune responses13. PRRs are expressed in various subcellular locations, and thus transmit information on whether a stimulus comes from the cell surface, within intracellular compartments, within the cytosol, or the nucleus. They can also be secreted into bodily fluids, and serve intercellular functions29. Cells of the innate immune system such as macrophages and dendritic cells undergo dramatic changes as a result of activation of their PRRs, and non-myeloid cells, such as epithelial and endothelial cells, are increasingly understood to respond to PAMPs and DAMPs in ways that relate to host defense30,31. Cells expressing PRRs mature, produce cytokines, are involved in phagocytosis, trigger a plethora of cell death pathways as well as impacting other arms of the immune response including opsonization, complement activation and adaptive immune activation. Here we will focus on 3 broad categories of PRRs, which cover the breath of recognition from the cell surface to the cytosol and which brought great insight into innate immunity in the past 50 years.

Toll like receptors

TLRs are Type 1 transmembrane glycoproteins and are structurally characterized by the extracellular leucine rich repeat (LRR) motifs required for ligand binding and the intracellular cytoplasmic Toll-interleukin 1(IL1) receptor-resistance (TIR) homology domain required for downstream signaling32. To date there are 10 TLRs identified in the human genome (TLR1–10) and 13 in mice (TLR1–13) although TLR10 is not functional in the mouse due to the presence of a stop codon in the sequence. Each TLR is triggered by unique PAMPs as outlined in table 1. TLRs 1, 2, 4, and 5 are localized to the cell surface, while TLR3, 7, 8, and 9, which all play roles in nucleic acid sensing, are found on intracellular compartments32.

In order to begin to describe the discovery of this family of receptors in humans and mice we first have to cover the early work on the IL1 receptor (IL1R1) and the initial discovery of the protein Toll in the fruit fly Drosophila. A number of labs in the 1980s characterized the functions of the pleiotropic proinflammatory cytokine IL1 as a critical regulator of T cell activation, an inducer of fever in addition to the acute phase response, which involves induction of proteins such as C-reactive protein in the liver33. The gene encoding the receptor for IL1, IL1R1 was first cloned in 1988 but curiously the predicted sequence did not contain any recognizable motifs to indicate its mechanism of action34. That was until 1991 when the protein Toll was characterized in D. melanogaster and shown to have a homologous cytosolic domain to the IL1R (now termed the TIR domain)35. Toll was first identified as being involved in dorsoventral polarity in the fly (reviewed in36). Interestingly, a pattern we have seen emerge over time is that proteins have a ‘double job’ and can play crucial roles in developmental processes in addition to processes within the immune system, at least in Drosophila. The Drosophila Toll receptor binds to the ligand Spätzle, signaling through an adaptor protein called Tube, resulting in activation of the kinase Pelle and subsequent activation of the NFκB transcription factor family member Dorsal, which is inhibited by the protein Cactus. Meanwhile in 1994, Barbara Baker and colleagues reported on N protein in tobacco plants, which conferred resistance to tobacco mosaic virus. It was Barbara Baker who coined the term ‘TIR’ domain, which stands for Toll-IL1R-Resistence domain, given the homology between Toll, IL1RI and N protein, a disease Resistance protein37. All these signaling components were then found to have homologues within the mammalian system. When loss of function mutants for Toll were generated in the fly, researchers were surprised to find these flies were highly susceptible to fungal infections yet resistant to gram negative bacterial infections38. It was found that activation of the Toll pathway by fungal infection induced the production of the antimicrobial peptide drosomycin downstream of the NFκB family member DIF (drosophila immunity factor), while diptericin is the antimicrobial peptide produced in flies in response to gram negative infection which signals through the IMD (immune deficient) pathway39–42.

Bioinformatic analysis revealed more mammalian proteins with TIR domains, leading to the description of the TLR family, which all have TIR domains, but unlike IL1RI which has Ig domains, have leucine-rich repeats (LRR). TLR4 was the first to be identified as a mammalian homologue of Toll. Medzhitov et al. were the first to show that an active form of TLR4 was capable of inducing the expression‎ of the co-stimulatory molecule B7 (CD80), a critical finding as it provided a link to T cell activation from an innate immune receptor15. Following this work, genetic mouse models identified TLR4 as the critical receptor responsible for the gram-negative bacterial product and driver of sepsis lipopolysaccharide (LPS)16,43. In the 1960s a spontaneous mutation had occurred in the C3H/HeJ mouse colony at the Jackson Laboratory, rendering the mice resistant to LPS toxicity. It was the work of the Beutler lab that traced the missense mutation to exon 3 of the TLR4 gene (previously referred to as the Lpsd gene)16. Jules Hoffmann and Bruce Beutler were awarded the Nobel prize in medicine in 2011 for their work on uncovering Toll as an innate sensor in flies, and TLR4 as the receptor for LPS in mice, respectively16,38. In 1999 the Akira lab developed TLR4 knockout mice and showed they failed to respond to LPS, again confirm‎ing that indeed TLR4 is the signaling receptor for LPS44. One important aspect, however, in LPS lethality was the discovery that LPS could induce caspase-11 and promote a type of cell death called pyroptosis via caspase-1 (discussed further below)45. LPS was then shown to bind Caspase-11 and activate this process, which in fact was key to LPS lethality46,47. Importantly, low dose poly:IC was able to bypass the requirement of TLR4 for LPS lethality in mice. The main effect of TLR4, therefore, in mice with regard to lethality is to induce caspase-11, which mediates the effect of LPS.

TLR Signaling

TLR family members signal through similar intersecting pathways, and because TLR4 represents the best studied family member, it will be our focus here. TLR4 is made up of leucine rich repeat sequences on the extracellular N terminus and the TIR signaling domain that lies inside the cell membrane and forms the platform for downstream signaling cascades. It does not operate alone in the recognition of LPS, but instead it works with a number of co-receptors including LPS binding protein (LBP), which binds to LPS in micelles allowing another co -receptor cluster of differentiation 14, (CD14) to interact (reviewed in48). CD14 increases the sensitivity to LPS by over 1000-fold49 and forms a complex with MD2 and TLR4 on the cell surface. 5 out of the 6 lipid chains within LPS are buried within the hydrophobic pocket of MD2, which bridges the dimerized complex together to form the “m” structure, solved by crystallography in 2009 by Park et.al50. These conformational changes that occur once LPS is bound initiates the downstream signaling cascade inside the cell (reviewed here51). TLR4 has the most complicated downstream signaling of all the TLRs as it has the ability to interact with multiple adaptor proteins. In 1997, myeloid differentiation primary-response protein 88 (MyD88) was shown to signal downstream of the IL1R1 receptor to activate NFκB52,53. It was then found to contain a TIR domain and signal through homotypic interaction with the TIR domains in TLRs48. In the case of TLR4 signaling, MyD88 functions alongside the adaptor protein MAL (also known as TIRAP) to drive NFκB54,55. MyD88 also contains a death domain which mediates its interactions with IL1 receptor activated kinase-4 (IRAK4) which then activates IRAK1 and 2 through autophosphorylation51,56,57. TNF receptor-associated factor 6 (TRAF6) is a ubiquitin ligase recruited to the complex and initiates the formation of K63 ubiquitin chains forming scaffolds for the recruitment of transforming growth factor (TGF)B-activated kinase 1 (TAK1) and TAK-binding proteins TAB2 and 358. Next, the IκBα kinase complex is activated through phosphorylation, and undergoes K48 linked ubiquitination and degradation, releasing NFκB to translocate to the nucleus and activate proinflammatory genes51,59. While most TLRs use a similar MyD88-dependent signaling pathway, TLR4 is unique in that it also engages in parallel with the adaptor proteins TIR domain–containing adapter-inducing IFN-β (TRIF) and TRIF-related adaptor molecule (TRAM)60,61,62. Downstream signals include TRAF3, which recruits IKKε /TANK-binding kinase (TBK1), which then phosphorylates and activates IFN regulatory factor 3 (IRF3). This in turn moves to the nucleus where it induces the production of antiviral proteins including type I IFNs62–64. These higher order complexes that form downstream of the adaptor protein engagements are sometimes referred to as the “Myddosome” and “Triffosome” complexes, which act as supramolecular organizing centers (SMOCs), promoting these signaling events58. While MyD88 signaling occurs at the plasma membrane, activation of the Triffosome requires endocytosis of the receptor complex and subsequent activation from the endosomal compartment of the cell64.

TLR4 activation has been shown to drive profound metabolic changes in macrophages, enhancing glycolysis and promoting what is termed the ‘Warburg effect,’ meaning a shift to aerobic glycolysis and a change in mitochondrial metabolism65. This is critical for the macrophage response to LPS, since inhibiting glycolysis decreases production of the key pro-inflammatory cytokine IL1β65. This process requires dimerization of the glycolytic enzyme pyruvate kinase isozyme M2 (PKM2), which translocates to the nucleus promoting the expression‎ of HIF-1alpha-dependent genes, including those encoding enzymes in glycolysis but also IL1β itself66. Profound metabolic rewiring occurs in the LPS activated macrophages with the accumulation of the Krebs cycle intermediate succinate67. This in turn has been shown to be oxidized by the Krebs cycle enzyme succinate dehydrogenase (SDH) leading to reverse electron transport through complex I in the mitochondria, driving production of reactive oxygen species (ROS) and further promoting HIF-1alpha activation68. LPS has also been shown to increase fumarate production via repression of the Krebs Cycle enzyme fumarate hydratase69. This disturbs the mitochondria via an increase in mitochondrial membrane potential, leading to the release of mitochondrial double-stranded RNA, which is detected by the RNA sensors RIG-I and MDA-5, promoting the expression‎ of interferon-beta. Finally, LPS has also been shown to increase expression‎ of the enzyme aconitate decarboxylase-1, encoded by the gene Irg-1. This converts aconitate to itaconate, which has a wide range of anti-inflammatory effects, acting to limit inflammatory macrophages70. These studies contributed to the field of immunometabolism, which began to emerge in earnest from 2013 (reviewed in71,72). Further work is needed on the complexities of immunometabolism in innate immunity. A whole multitude of metabolites are changing dynamically and we’re only at the start of the effort to determine their roles in the regulation of immune cell effector functions.

All of these discoveries happened over a roughly 20-year period and are a triumph of molecular immunology. While many of us in the field were still scratching our heads as we considered the emerging complexities of TLRs, other pattern recognition receptors entered the picture and further widened the view.

Discovery of the inflammasome

The second area within innate immunity that has seen a frenzy of activity over the past 20 years or so concerns inflammasomes. ”Inflammasome” is a term for a multiprotein complex involving one of several PRRs, that forms in the cytosol and plays critical roles in the activation of the cytokines IL1β and IL18, as well as processing gasdermin family proteins that mediate an inflammatory type of cell death called pyroptosis. The NLR sensors that comprise the inflammasome respond to PAMPS and DAMPS73. Here we provide a brief overview of this field, and for a more in depth account we direct you to the following reviews74,75. Each known PRR-associated inflammasome and its activation are outlined in Table 1. One of the dominant PRR families is the nucleotide-binding domain (NBD), leucine-rich repeat (LRR)-containing (NLR) protein family. These come in two flavors, those containing a pyrin domain (PYD) in the N terminus, referred to as NLRPs, or those containing a CARD domain (CYD) in the N terminus, referred to as NLRCs. Activation of the inflammasomes leads to proteolytic cleavage of pro-caspase 1 into its catalytically activated form. Caspase-1 processes the pro forms of IL1β and IL18 into their active forms, which are released through pores formed in the cell by gasdermin D76,77. Formation of the gasdermin D pore also causes a specialized form of cell death referred to as pyroptosis, which is associated with inflammation and downstream activation of adaptive immune cells74,75 (Fig.2). There are a number of additional gasdermin family members, all of which show specific cell and tissue expression‎ patterns with emerging roles in human health and disease. All gasdermins form pores but the exact mechanisms driving their activation remains under investigation (reviewed in78).

NLRP3 inflammasome

NLRP3 represents the best studied of the inflammasomes with a wide array of activating processes including ion fluxes (K+ efflux, Ca 2+ flux), metabolic changes (mitochondria and lysosome dysfunction, fatty acid synthesis, hyperosmotic stress) and even nucleic acids (dsDNA, viral RNA and oxidized DNA)79–81. How so many varying molecules can activate a single sensor is unclear and continues to be a focus of intense research. Activation of the NLRP3 inflammasome occurs in two steps. Step 1 involves upregulation of the components of the pathway typically through activation of NFκB, and signal 2 involves direct activation of the downstream sensor82,83. The requirement for priming differs between cell types and specific inflammasomes. While NLRP3 is advantageous to the host in responding to bacterial and viral infections and a range of noxious stimuli, it is worth noting that excessive activation of this pathway has been associated with a number of inherited inflammatory conditions including cryopyrin-associated periodic fever syndrome (CAPS)84.

Nucleic acid sensing, uncovered

Another perhaps unexpected development in innate immunity was the uncovering of sensors of DNA, which provoke innate immunity. DNA therefore moved from not only being the information molecule of life, but also a key driver of immunity and inflammation, particularly if it showed up in the wrong place outside the nucleus. Or, to use the phrase of a central figure in the field of innate immunity, Vishva Dixit, if it ‘breached the sanctity of the cytosol’. In fact, back in 1928 even before DNA was shown to be the genetic material scientist Frederick Griffith famously showed a “transforming principle” activating the immune system which later turned out to be DNA85. The reason we do not respond to our own DNA is that it is encased within the protective walls of the nucleus or mitochondria and is therefore hidden from PRRs. We now know this process is error prone, and escape of self-DNA into the cytosol is associated with autoinflammatory conditions such as systemic lupus erythematosus (SLE). From the mid-2000s there was a major push to identify and characterize the main players involved in the direct sensing of DNA within the cytoplasm. TLR9, which is localized to endosomes, was shown to be a receptor for CpG DNA, common in bacteria86. AIM2 was identified as a DNA sensor capable of binding directly to DNA and inducing the formation of an inflammasome leading to IL1 and IL18 release24,87. However, less was known about how DNA was sensed to lead to the production of type 1 interferons (IFNs). A number of potential DNA receptors were shown to play some role including IFI16, DEAD-box helicases and HNRNPA2B188. From early studies it was clear that STING played a key role in this pathway and at first was considered to not only be the adaptor but potentially the direct sensor. However, the field made rapid progress with the discovery of cyclic dinucleotides and important role for the adaptor STING in regulating this pathway25,89,90.

Cyclic GMP-AMP (cGAMP) synthase (cGAS) is a cytosolic DNA sensor that activates type I interferons through production of the second messenger cyclic GMP-AMP (cGAMP), which activates the adaptor STING25(for a thorough review of the field please read the following91,92). cGAS is present in the cytosol under physiological conditions in an autoinhibited state. Once cGAS binds DNA it undergoes conformational changes leading to the production of cGAMP from the cell’s stores of ATP and GTP25. cGAMP then functions as a second messenger to bind and activate STING. STING had been known to bind and respond to bacterial second messengers but cGAMP was the first example of a host derived second messenger activating this pathway89. STING activates the kinase TBK1, which phosphorylates the transcription factor IRF3, leading to its translocation to the nucleus and subsequent activation of Type 1 IFNs. STING can also activate NFκB, leading to the production of proinflammatory cytokines.

The main receptors that recognize RNA within the cytosol are RIG-1 and MDA5 (Fig.2 and Table 1). These cytosolically localized receptors are important in the recognition of single stranded and double stranded RNA, respectively22,23. Interestingly, our own RNA contains modifications including adenosine to inosine changes (A to I edits), which helps protect our own RNA from activating these receptors93. These receptors play key roles in protection against viral RNA infections including influenza, hepatitis, and West Nile virus (reviewed in94). RIG-1 can recognize key structures in RNA including 5′ppp- while MDA5 favors long double stranded RNA. When the receptors are activated, they result in the robust induction of Type I IFNs (Fig. 2).

How activation of the innate immune pathways controls adaptive immunity

For many immunologists the most important feature of innate immunity was how it promotes adaptive immunity, and the discovery of that connection is perhaps the most critical finding in immunology over the past 50 years. The 2nd half of the 2011 Nobel prize for Medicine was awarded to Dr. Ralph Steinman for his discovery, almost exactly 50 years ago, of dendritic cells116. Dendritic cells are often referred to as professional antigen presenting cells as they are instrumental in capturing antigens from tissue sites throughout the body and presenting them to T cells within the immune system’s specialized lymph nodes. Dendritic cells reside in the periphery and express high levels of innate immune PRRs. Once activated, dendritic cells undergo maturation involving increased expression‎ of proinflammatory cytokines, migratory chemokine receptors, and upregulation of surface proteins that interact with T cells and activate downstream adaptive immune responses. The critical role for the TLR signaling pathway in DC maturation was demonstrated when DCs from MyD88 knockout mice failed to undergo maturation.117,118. Without that critical adaptor protein used by most TLRs, only stimulation via the alternative TRIF adaptor downstream of TLR4 could activate the antigen presentation activities of DCs. Furthermore, early work from the MyD88 deficient mice highlighted the importance of the TLR signaling pathway for driving inflammatory T cell responses in particular, as less inflammatory T cell subtypes and B cells retained most functions in the absence of the bulk of TLR signaling118. Dendritic cells are not a single cell type and over the last number of years many DC subtypes have been characterized, including those that are tissue resident within each organ (reviewed in119). DCs engage with cytotoxic T cells (CTLs) for the removal of viral pathogens and tumors. Intracellular pathogens and protozoa are dealt with by CTLs and Th1 (type 1 subtype) of T cells while Th17 (type 3 subset) cells play a critical role in control of extracellular pathogens. For more in-depth discussion of DCs and adaptive immunity we refer you to the following reviews119,120.

Discovery of the Innate Lymphoid cells (ILCs)

Members of the innate lymphoid family of cells were first described in the mid ‘70s but the nomenclature utilized today was only proposed in 2013121. ILCs play key roles in the regulation of the innate immune responses. ILCs act as innate counterparts to T cells, with the functional subtypes of these cells seemingly mirroring the functional subtypes of T cells, but without the antigen-specific T cell receptor. The founding member of the ILC family, natural killer (NK) cells act similarly to CD8 cytotoxic T cells, while ILC1s are Th1 like, ILC2 are Th2 like, and ILC3 are similar to type 3 responding T cell subtypes (TH17, TH22). NK cells were first named in 1975122 and shown to be important in early responses to viral infections. ILCs generally function within mucosal tissues, where they are typically present at low numbers, and are involved in activation of inflammation, tissue remodeling, metabolic control and influence on adaptive immune responses. For in-depth review of the ILC literature we recommend the following reviews121,123.

Controls on Innate immunity

Another very fruitful area is the past 50 years has been the elucidation of multiple controls on innate immune pathways. Transient activation of the complex signaling cascades downstream of PRRs is critical to the maintenance of homeostasis. Therefore, it was important to understand the key players involved in controlling the timing of these immune cascades. Like all aspects of the immune response, there are layers of regulation that contribute to the exquisite timing observed in innate immune cells. The speed of protein turnover is one simple layer of regulation. For example, the transcription factor p65 is retained in the cytoplasm by IκB and only travels to the nucleus following activation of the PRRs. Following stimulation, IκB undergoes phosphorylation and subsequent degradation within the 28S proteasome. However, the turnover of IκB is rapid as it itself is induced by NFκB and therefore once IκB is translated it resumes its role of retaining p65 within the cytoplasm contributing to the transient nature of the response48.

There have been many processes identified that play roles at various stages of the innate immune response. We have created a summary table outlining some of the key players (Table 2) and will focus on two key controllers here: A20 and ubiquitination, and non-coding RNAs, both of which have seen an extensive body of findings in the past 20 years.

Table 2:

Negative regulators of Innate Immunity

Protein NameMode of RegulationKnockout phenotypeSNPsReference

ABIN1	Partner of A20	Knockout mice have similar phenotype to A20 KO mice	SNPs associated with SLE and psoriatic arthritis	145–147
A20	Ubiquitin modifying enzyme Deubiquitylates TRAF6	Develop autoimmunity	SNPs associated with rheumatoid arthritis, SLE, psoriasis, coeliac disease, Crohn’s disease, type 2 diabetes, atheroscleros is and lymphomas	148
CYLD	Inhibits NFkB activity	KOs are sensitive to chemically induced tumors and impaired fear memory	Mutations in CYLD have been identified in patients with polycythemia vera	149–151
DUBA	Negatively regulates interferons through deubiquitination of TRAF3	Mice with DUBA-deficient T cells develop excessive inflammation in the small intestine after challenge with anti-CD3 antibodies	Variants associated with X-Linked Intellectual Disability and Congenital Malformation	152,153
IL10	Blocks induction of proinflammatory cytokines downstream of ligands such as LPS	Knockout mice develop colonic inflammation beginning at 3 weeks of age	GWAS studies associate IL10 SNPs with inflammatory conditions and cancer	154
IL1R2	Decoy receptor for IL1 signaling	KO mice show increased susceptibility to collagen induced arthritis	SNPs association in a cohort of Chinese patients with cervical cancer	155–157
IRAK-M	Expressed in monocytes and macrophages. It is induced by TLRs and negatively regulates the pathway through inhibition of the formation and activation of the IRAK1/4/TR AF6 complex	KO mice show abnormal osteoclast development and increased inflammatory responses to infection	SNPs associated with early onset asthma	158,159
LLRC25	Inhibits TLRs by promoting autophagic degradation of p65. Inhibits IFN by promoting degradation of RIG1	No data	No data	160,161
Metallothionein 3 (MT3)	Negative regulates caspase11 through regulation of zinc levels	KOs show abnormalities in psychological behavior and show accelerated onset and progression of ALS	SNP associations with autism	162–165
MyD88S	Myd88S arises from alterative splicing of the MyD88 gene and behaves as a dominant negative of IlL and LPS signaling	No data	No data	166
NLRX1	Negatively regulates RIG1 signaling by binding MAVS. Also negatively regulates TLR signaling by targeting TRAF6 and IKK	KOs produce higher levels of IFNB and IL6 following influenza infection. MAVs constitutively interacts with RIG1 in the KO mice	No data	167–169
NLRC5	Negatively regulates NFkB by inhibiting IKK phosphorylation. Regulates type 1 IFN by blocking RIG1 and MAVs interactions. NLRC5 is also an MHC-class 1 trans activator	Reports of increased TLR signaling and IFN production in KO mice. In addition to the KO mice having impaired CD8T cell responses due to loss of MHC Class I	SNPs associated with susceptibility to pulmonary aspergillosis	170–175
NLRP11	Targets traf6 for degradation via the ligase RNF19A and is primate specific	No data	SNP associated with susceptibility to Crohn’s disease, and a gene duplication was found in juvenile idiopathic arthritis patients	176,177
Nod2 (Card15)	Suppresses NFkB	KOs show enhanced TH1 cytokines IL12, IFNy and IL18 following stimulation with peptidoglycan	SNPS associated with excessive Th1 responses and Crohn’s like disease	178
PHLDA1	Negatively regulates TLR4 signaling through interactions with TOLLIP	Information from Mouse genome informatics (MGI) indicates that KOs are viable with no obvious defects in immune function	No data available	162
PUM1	Negatively regulates LGP2. Can suppress TLR4 mRNA translation	KO exhibits reduction in body and organ size	Deletions associated with Pumilio1-Associated Development al Disability, Ataxia and Seizure; PADDAS	179–182
RP105	Interacts with TLR4 complex to inhibit interactions with LPS	DCs from RP105 KO mice produced increased cytokines in response to LPS. Mice were more susceptible to LPS challenge producing higher levels of TNF post LPS injection	One SNP in cow associated with a mycobacteriu m infections	183,184
SOCS-1	Suppresses IRAK1 and inhibits Type 1 IFN	KO mice die within 3 weeks due to multi organ failure. They are more susceptible to endotoxic shock	SNPs associated with rheumatoid arthritis and early onset autoimmunit y	185–189
sST2	ST2 is the receptor for IL33 and the main form of the receptor functions to promote NFKB signaling while the soluble version inhibits. sST2 acts as a decoy receptor and binds IL33	No Data	SNPS in the distal promoter which impacts the full gene is associated with atopic dermatitis	156,190,191
SIGIRR (IL1R8)	Interacts with TRAF6 and IRAK Competes for binding to MyD88 and acts as a coreceptor for anti-inflammatory cytokine IL37	KO mice show enhanced responses to IL1 and LPS but not TNF and show increased suscesptibility to endotoxic hock and colitis	SNPs associated with infectious diseases including tuberculosis	192–194
Soluble TLRs (sTLRs)	They block interactions between the TLRs and their agonists.	No Data	No Data	195
Tollip	Autophosphorylates IRAK1	IL-13-treated Tollip KO mice significantly increased lung eosinophilic inflammation	Mutations associated with development and/or prognosis of idiopathic pulmonary fibrosis (IPF)	196–199
TRIAD3A	Ubiquitylates TLRs	KOs have microglial defects	Recessive mutations in RNF216/T RIAD3 cause Gordon Holmes syndrome (GHS)	200,201
TRAILR	Stabilizes IkBa	KOs show enhanced immune responses with increased levels of IL12, IFNα and γ.	SNP is associated with enhanced responses to IFNβ treatment in MS patients	200,202
USP38	Through altering ubiquitination through interactions with KDM5B. USP38 also negatively regulates IFN signaling	KO mice are more susceptible to endotoxic shock and acute colitis producing higher levels of inflammatory genes compared to wild type mice. KO mice also have increased K33-linked ubiquitination and higher expression‎ of TBK1	SNPs linked to susceptibility to asthma and malaria	203–205

Open in a new tab

A20 and control of ubiquitination

A20 (also known as TNFAIP3) is a universally expressed ubiquitin modifying protein that is itself induced downstream of NFκB signaling. A20 is unique in that it can work to add ubiquitin chains or remove them. A20 functions to negatively regulate NFκB signaling in addition to inhibiting cell death124. It became clear that A20 is a critical component for the maintenance of homeostasis following the generation of A20 knockout mice125. While the mice are born at expected Mendelian ratios they die quickly after birth due to multiorgan inflammation,125 and the main contributor of the dysregulated signaling appears to stem from the TLR pathway126. A20 is strongly associated with a number of inflammatory diseases, such as SLE, due to single nucleotide polymorphisms (SNPs) identified in the gene,127. A20 has also been implicated as a tumor suppressor, as SNPs are associated with lymphoma124.

Many arms of the innate immune signaling pathways are controlled through the process of ubiquitination. Specific ubiquitin marks determine if a protein is removed or activated. Lysine 48 (K48) linked ubiquitin chains mark proteins for degradation through the proteasome while lysine 63 (K63) linked ubiquitin chains act as activation scaffolds for downstream signaling. A20 mediates negative regulation of the TLR128 and NOD129 signaling pathways through deubiquitination of K63 linked proteins and has also been shown to control the NLRP3 inflammasome129.

Emerging Roles for noncoding RNA in the regulation of innate immunity

While much of the early work on innate immune signaling focused on protein cascades and cellular phenotypes, the development of next generation sequencing has opened a Pandora’s box of RNA transcripts with apparent regulatory function. The majority of the human genome is actively transcribed, but an important question has been, how much of the RNA transcripts that don’t encode proteins are functionally relevant? The most advanced area of understanding as it relates to noncoding RNA in innate immunity is the field of microRNA research. MicroRNAs (miRs) were first identified in the early 90s130,131 and their specific roles in the innate immune system emerged in the late 2000s, with the description of miR155 and miR146a as regulators of NFκB132–134. miRNAs are small RNAs (23 nucleotides in length), transcribed mostly from RNA polymerase II, with a few being RNA polymerase III transcripts. They can be encoded as independent genes or emerge from the introns of protein coding genes. A primary transcript is transcribed then processed within the nucleus, before being exported to the cytoplasm where it undergoes cleavage by Dicer to form a duplex. One strand of the RNA is then loaded onto the RNA-induced silencing complex (RISC) which then guides the complex to the 3’ untranslated regions (UTRs) or target mRNAs leading to repression of the target protein.

miR155 and miR146 represent the two best characterized miRNAs within the innate immune system and evidence suggests that they can even counterbalance each other. Both miR155 and miR146 are highly inducible following inflammatory activation with TLR ligands or following infection. miR155 is proinflammatory and targets negative regulators SHIP1 and SOCS1 while miR146 is anti-inflammatory targeting TRAF6 and IRAK1132–134. miR146a knockout mice show symptoms of chronic inflammation and autoimmunity and these mice express higher levels of miR155 adding evidence that these miRNAs act as counterbalances to the homeostatic inflammatory response. Deficiency of miR155 in mice has wide ranging impacts on their immune system. They show reduced responses in endotoxic shock models and their adaptive immune responses are skewed with effects on T and B cell responses during infection or autoimmunity135. For more on miRNAs in the innate immune system we recommend the following comprehensive reviews135,136).

While microRNAs were the first to have an understood regulatory role, the largest group of RNAs produced from the genome are long noncoding RNAs (lncRNAs). Depending on the analysis pipeline, there is predicted to be anywhere between 20,000 to 100,000 lncRNAs, with a small number of these loci being shown to encode small functional peptides137. In 2013 a lncRNA named lincRNA-Cox2 was shown to impact genes of the innate immune system, with downregulation of genes such as IL6 occurring when lincRNA-Cox2 was removed, while interferon genes were upregulated138. XIST represents the best studied lncRNA, first identified in 199121. XIST is encoded on the X chromosome and is required for X chromosome inactivation in females. Interestingly TLR7 is encoded on the X chromosome and is capable of escaping X inactivation139. This has been shown to be particularly important in T cells and links this important noncoding RNA with regulation relating to autoimmunity in particular SLE which disproportionately impacts women as well as individuals with Klinefelter syndrome who carry an extra copy of the X chromosome. Excess TLR7 during this condition could explain some of the signaling defects observed (reviewed in140). A recent study showed that simply overexpressing XIST in male mice resulted in formation of autoantibodies and T and B cells from these mice resembled those of wild type females141. Over the past decade this field has greatly expanded with many publications on lncRNAs and immunity. For in-depth reviews of this field we recommend the following142–144.

Therapeutic possibilities

The uncovering of innate immune processes and their regulation was followed by efforts to exploit these remarkable findings for therapeutic gain. New and exciting prospects are emerging.

The immediate application of the identification of PRRs was in vaccine adjuvancy, as it was highly likely that their discovery would explain how Janeway’s ‘dirty little secret’ might work in molecular terms. Despite much effort, progress on rationally designing adjuvants for vaccines has been slow and is ongoing206,207.

The importance of innate immunity in vaccine adjuvancy was elegantly demonstrated in a study by Bali Pulendran and colleagues,208 involving the vaccine for yellow fever comprising a live attenuated virus termed YF-17D. This is one of the most effective vaccines ever developed, providing protection for decades from a single shot. To elicit its effects, it requires a wide array of innate sensors, comprising TLR2, TLR3, TLR7, TLR9, RIG-I and MDA-5, which is presumably sensing diverse PAMPs in the virus. Separately, detoxified versions of LPS were tested, even before the finding that TLR4 was the receptor for LPS, and monophosphoryl lipid A (MPL) emerged as an adjuvant. It is used in combination with a plant extract termed QS-21 (which is a liposome made from plant saponins from the Chilean soap bark tree) and cholesterol in a vaccine for shingles, as well as in a malaria vaccine. AS04 is an adjuvant comprising aluminum salts with MPL and is used in vaccines for human papilloma virus and hepatitis B virus. A modified form of the TLR9 agonist CpG DNA is used in a vaccine for hepatitis B. Clinical trials are currently running with imiquimod (a small molecule used to treat genital warts and subsequently shown to be a TLR7 ligand) in influenza, flagellin (the TLR5 ligand) in influenza, and a double-stranded RNA polymer (the TLR3 ligand) in influenza and rabies. In addition, there is substantial interest in STING agonists, especially in the context of anti-tumor vaccines but also in influenza, HIV and tuberculosis206,209.

The mRNA vaccines for COVID-19 raised new questions about how innate immune activating ligands relate to adjuvanticity. The vaccines include modified mRNA encoding the Spike protein from SARS-CoV2 in a lipid nanoparticle (LNP) comprising ionizable lipids and cholesterol. The LNP itself was shown to have adjuvant properties when used with a protein antigen, whilst the vaccines were shown to require MDA-5 for their immunogenicity, and intriguingly not a whole range of other PRRs207. Whether MDA-5 is sensing the RNA in the vaccine itself, or endogenous RNA, perhaps of mitochondrial origin, is not known.

While vaccination adjuvants exploit PRR activation to prime the immune system, on the therapeutic flip side of the coin, there are also efforts to block PRRs in autoinflammatory and autoimmune diseases. Antibodies that target TLR2 in such conditions as ischemia reperfusion injury and rheumatoid arthritis showed preclinical promise210 but haven’t advanced, largely because of lack of efficacy in human clinical trials. Attempts were also made to target TLR4 in sepsis211, with the antagonist Eritoran showing only marginal effects in clinical trials, which could have been because of the timing of the intervention clinically, or perhaps because of the need for careful patient stratification.

NLRP3 has proven to be a target of great interest, given its potential role in a wide range of autoinflammatory and autoimmune conditions, most notably in diseases of the CNS such as Alzheimer’s and Parkinson’s disease212. A small molecule inhibitor termed CRID3/MCC950, which was originally shown to block signal 2 for IL1beta production, was also shown to specifically target NLRP3 by binding the NACHT domain essential for ATP-dependent oligomerization during NLRP3 inflammasome activation213,214. Multiple compounds based on this inhibitor as well as other NLRP3 inhibitors are at various stages of clinical development, with trials running in CAPS, osteoarthritis, gout, myelodysplastic syndrome, asthma, Parkinson’s disease and coronary artery disease215. It may well be that inhibiting NLRP3, or even other inflammasomes, will have therapeutic applicability across multiple inflammatory diseases. It might even be possible that inhibiting NLRP3 will be somewhat akin to antibiotics and infectious diseases- one drug bringing benefits in a number of diseases, in this case, diseases where inflammation driven by myeloid cells is pathogenic.

There is also substantial interest in targeting the cGAS-STING pathway, in such conditions as rheumatoid arthritis, stroke, SLE and neurological disorders216,217.

The advent of cytokines led to the development of multiple cytokine-targeting therapeutics, as well as inhibitors of cytokine-driven signals, notably in the JAK family of tyrosine kinases, bringing substantial benefits to patients. Targeting specific cytokines or signals has not been especially problematic in terms of increasing the risk of infection or cancer. The hope is that the targeting of PRRs, or indeed the signals they activate, will bring similar if not superior clinical benefits, especially in diseases where targeting cytokines might not be especially effective or is yet to be proven.

Concluding remarks

50 years ago, our understanding of innate immunity was primitive, as that is how innate immunity was then viewed. A concerted effort, across many hundreds of laboratories and involving thousands of researchers, from research assistants to graduate students to post-doctoral scientists publishing their work, has revealed a whole world of interconnected processes that are far more sophisticated than the primitive assumptions. Exciting findings will continue to be made and will likely reveal even more component parts in innate immunity. These future findings will further increase our understanding of these essential and most fundamental of biological processes, the targeting of which must hold great therapeutic promise for immune-mediated and inflammatory diseases and cancer.

Box topic 2: Evolutionary diversity of innate immunity.

Comparing innate immune processes across animal species has revealed some interesting differences that continue to provide insights into the evolutionary diversity of innate immunity. The sea urchin has a huge repertoire of innate receptors, including 222 TLRs, and 203 NLRs. It’s signaling repertoire is equally expansive, with 58 TIR adapter-like proteins, 36 TRAF proteins and 541 death-domain-containing proteins220. This is likely to be because the sea urchin lacks adaptive immunity and so has an expanded innate repertoire to ensure adequate diversity to deal with infectious microbes that might infect it. The same can be said of plants, with for example Arabidopis having hundreds of TLRs and NLRs221. The cGAMP/STING pathway turns out to be conserved even in bacteria, where it provides defense against bacteriophages222,223. Bats have a different NLRP3 with an altered leucine-rich repeat domain, leading to a less active NLRP3 inflammasome. This might be one of the reasons why bats can tolerate viruses that are otherwise pathogenic in humans224. Finally, horses have TLR4 that can recognize a type of LPS from Rhodobacter sphaeroides, which infects horses. That type of LPS is an antagonist against human TLR4. This allows horses to mount an appropriate immune response to that particular bacterium220.

Acknowledgements

Susan Carpenter is supported by R01 AI148413 from NIAID and R35GM137801 from NIGMS and Luke AJ O’Neill is supported by European Research Council Grant 834370 and Science Foundation Ireland Grant 12/I1/1531.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declaration of interests

Susan Carpenter is a paid consultant for NextRNA Therapeutics. Luke AJ O’Neill is a paid consultant for Sitryx Therapeutics.

References:

• 1.Al B, Suen TK, Placek K, and Netea MG (2023). Innate (learned) memory. J Allergy Clin Immunol 152, 551–566. 10.1016/j.jaci.2023.06.014. [DOI] [PubMed] [Google Scholar]
• 2.Trouw LA, and Daha MR (2011). Role of complement in innate immunity and host defense. Immunol Lett 138, 35–37. 10.1016/j.imlet.2011.02.014. [DOI] [PubMed] [Google Scholar]
• 3.West EE, Afzali B, and Kemper C (2018). Unexpected Roles for Intracellular Complement in the Regulation of Th1 Responses. Adv Immunol 138, 35–70. 10.1016/bs.ai.2018.02.001. [DOI] [PubMed] [Google Scholar]