1. Human Biology and Medicine
  2. Immunology and Inflammation
Download icon

Yolk-sac-derived macrophages progressively expand in the mouse kidney with age

  1. Shintaro Ide
  2. Yasuhito Yahara
  3. Yoshihiko Kobayashi
  4. Sarah A Strausser
  5. Kana Ide
  6. Anisha Watwe
  7. Shengjie Xu-Vanpala
  8. Jamie R Privratsky
  9. Steven D Crowley
  10. Mari L Shinohara
  11. Benjamin A Alman
  12. Tomokazu Souma  Is a corresponding author
  1. Division of Nephrology, Department of Medicine, Duke University School of Medicine, United States
  2. Regeneration Next, Duke University, United States
  3. Department of Orthopedic Surgery, Duke University School of Medicine, United States
  4. Department of Orthopedic Surgery, Faculty of Medicine, University of Toyama, Japan
  5. Department of Cell Biology, Duke University School of Medicine, United States
  6. Department of Immunology, Duke University School of Medicine, United States
  7. Department of Molecular Genetics and Microbiology, Duke University School of Medicine, United States
Short Report
  • Cited 0
  • Views 1,055
  • Annotations
Cite this article as: eLife 2020;9:e51756 doi: 10.7554/eLife.51756

Abstract

Renal macrophages represent a highly heterogeneous and specialized population of myeloid cells with mixed developmental origins from the yolk-sac and hematopoietic stem cells (HSC). They promote both injury and repair by regulating inflammation, angiogenesis, and tissue remodeling. Recent reports highlight differential roles for ontogenically distinct renal macrophage populations in disease. However, little is known about how these populations change over time in normal, uninjured kidneys. Prior reports demonstrated a high proportion of HSC-derived macrophages in the young adult kidney. Unexpectedly, using genetic fate-mapping and parabiosis studies, we found that yolk-sac-derived macrophages progressively expand in number with age and become a major contributor to the renal macrophage population in older mice. This chronological shift in macrophage composition involves local cellular proliferation and recruitment from circulating progenitors and may contribute to the distinct immune responses, limited reparative capacity, and increased disease susceptibility of kidneys in the elderly population.

eLife digest

Older people are more likely to develop kidney disease, which increases their risk of having other conditions such as a heart attack or stroke and, in some cases, can lead to their death. Older kidneys are less able to repair themselves after an injury, which may help explain why aging contributes to kidney disease. Another possibility is that older kidneys are more susceptible to excessive inflammation. Learning more about the processes that lead to kidney inflammation in older people might lead to better ways to prevent or treat their kidney disease.

Immune cells called macrophages help protect the body from injury and disease. They do this by triggering inflammation, which aides healing. Too much inflammation can be harmful though, making macrophages a prime suspect in age-related kidney harm. Studying these immune cells in the kidney and how they change over the lifespan could help scientists to better understand age-related kidney disease.

Now, Ide, Yahara et al. show that one type of macrophage is better at multiplying in older kidneys. In the experiments, mice were genetically engineered to make a fluorescent red protein in one kind of macrophage. This allowed Ide, Yahara et al. to track these immune cells as the mice aged. The experiments showed that this subgroup of cells is first produced when the mice are embryos. They stay in the mouse kidneys into adulthood, and are so prolific that, over time, they eventually become the most common macrophage in older kidneys.

The fact that one type of embryonically derived macrophage takes over with age may explain the increased inflammation and reduced repair capacity seen in aging kidneys. More studies will help scientists to understand how these particular cells contribute to age-related changes in susceptibility to kidney disease.

Introduction

Tissue-resident macrophages constitute a highly heterogeneous and specialized population of myeloid cells, reflecting the diversity of their developmental origins and tissue microenvironments (Mass, 2018; Ginhoux et al., 2010; Schulz et al., 2012; Yona et al., 2013; Gomez Perdiguero et al., 2015; Mass et al., 2016; Hashimoto et al., 2013; Epelman et al., 2014; Stremmel et al., 2018; Hoeffel et al., 2015). In addition to their critical roles in host defense against pathogens, macrophages are central to sterile inflammation, angiogenesis, and tissue remodeling, making them an attractive target for therapeutic intervention. Tissue-resident macrophages originate from at least three distinct progenitors: (i) macrophage colony-stimulating factor one receptor (CSF1R)-positive yolk-sac macrophages; (ii) CX3C chemokine receptor 1 (CX3CR1)-positive yolk-sac macrophages, also known as pre-macrophages; and (iii) embryonic and neonatal hematopoietic stem cells (HSC) (Mass, 2018; Ginhoux et al., 2010; Schulz et al., 2012; Yona et al., 2013; Gomez Perdiguero et al., 2015; Mass et al., 2016; Hashimoto et al., 2013; Epelman et al., 2014; Stremmel et al., 2018). These populations are maintained in situ by self-renewal, largely independent of adult hematopoiesis (Mass, 2018; Ginhoux et al., 2010; Schulz et al., 2012; Yona et al., 2013; Gomez Perdiguero et al., 2015; Mass et al., 2016; Hashimoto et al., 2013; Epelman et al., 2014; Hoeffel et al., 2015; Sawai et al., 2016; Soucie et al., 2016).

Most tissues have mixed populations of different ontogenically-derived macrophages, and their relative contributions and temporal kinetics are tissue-specific. For example, microglia, Kupffer cells, and Langerhans cells originate from the yolk-sac, with minimal contribution from HSCs (Ginhoux et al., 2010; Schulz et al., 2012; Yona et al., 2013; Gomez Perdiguero et al., 2015; Mass et al., 2016; Sawai et al., 2016). The macrophage composition in the intestinal wall is highly dynamic. Yolk-sac-derived intestinal macrophages are rapidly replaced by HSC-derived macrophages after birth, but a subpopulation of yolk-sac-derived cells persist and self-renew in the specialized intestinal niches in adults (Bain et al., 2014; De Schepper et al., 2018). Importantly, investigators now recognize that macrophage ontogeny contributes to their roles in disease processes such as cancer progression; in pancreatic cancer, for example, yolk-sac-derived macrophages are fibrogenic, while HSC-derived macrophages are immunogenic (Zhu et al., 2017). This raises the possibility that developmental programs influence how macrophages differentially respond to disease insults.

Renal macrophages are found in an intricate network surrounding the renal tubular epithelium (Stamatiades et al., 2016; Berry et al., 2017; Viehmann et al., 2018) and have mixed origins from both yolk-sac and HSC (Schulz et al., 2012; Mass et al., 2016; Epelman et al., 2014; Munro et al., 2019). They exert unique functions depending on their anatomical locations; monitoring and clearing immune complexes is a function of cortical macrophages while bacterial immunity is the responsibility of medullary macrophages (Stamatiades et al., 2016; Berry et al., 2017). Renal macrophages critically control renal inflammation and tissue remodeling after injury with robust phenotypic reprogramming (Lever et al., 2019). Despite the importance of these roles, little is known about how the proportion and distribution of macrophages of different ontogeny change over time in the normal kidney and whether this influences the increased susceptibility and poorer outcomes of older patients to acute and chronic kidney diseases (Chen et al., 2019; Chawla et al., 2014; Sato and Yanagita, 2019). While most preclinical models of kidney diseases have used young animals, recent papers highlight distinct immune responses in aged mouse kidneys. Aged kidneys exhibit more severe inflammation than young kidneys in response to ischemic and toxic insults, leading to maladaptive repair and organ dysfunction (Sato and Yanagita, 2019; Sato et al., 2016). Furthermore, there is a growing interest in aging as a fundamental determinant of macrophage heterogeneity, such as in the heart and serous cavities (Molawi et al., 2014; Bain et al., 2016; Dick et al., 2019). However, data on the effects of aging on the renal macrophage populations are lacking. Here, using complementary in vivo genetic fate-mapping and parabiotic approaches, we identify a previously unappreciated increase in the proportion of yolk-sac-derived macrophages in the mouse kidney with age.

Results and discussion

Two complementary strategies were used to fate-map CSF1R+, and CX3CR1+ yolk-sac-derived macrophages (Figure 1 and Figure 1—figure supplement 1). Erythromyeloid progenitors (EMP) give rise to these populations, and the yolk-sac macrophages appear in the yolk-sac around E8.5. Subsequently, from E9.0 until E14.5, they proliferate, migrate, and colonize the embryo through the vascular system (Stremmel et al., 2018; Munro et al., 2019). To lineage-label these cells, we exposed Csf1r-CreERt; Rosa26tdTomato and Cx3cr1CreERt; Rosa26tdTomato embryos to 4-hydroxytamoxifen (4-OHT) at E8.5 and E9.5, respectively (Figure 1A and Figure 1—figure supplement 1A), (Mass, 2018). This efficiently and irreversibly labels yolk-sac-derived macrophages with the tdTomato reporter. Importantly, the approach does not label fetal monocytes or HSCs (Gomez Perdiguero et al., 2015; Yahara et al., 2020).

Figure 1 with 2 supplements see all
CX3CR1-positive yolk-sac macrophage descendants progressively expand in number in kidneys with age.

(A) Fate-mapping strategies of CX3CR1+ yolk-sac macrophages. 4-hydroxytamoxifen (4-OHT) was injected once into pregnant dams at 9.5 dpc and offspring analyzed at the indicated times (n = 4–6 for P0 to 6-month-old; n = 2 for 12-month-old). Yolk-sac macrophages and their progeny are irreversibly tagged with tdTomato. (B) Distribution of CX3CR1-lineage cells in postnatal kidneys. Arrows: CX3CR1-lineage cells. (C) Percentage of tdTomato+ to F4/80+ cells. Data are represented as means ± S.D. ***, p<0.001; ****, p<0.0001; n.s., not significant. (D) Confocal images of F4/80 and CD64 staining in aged kidneys (six mo) with CX3CR1-lineage tracing (n = 3). Scale bars: 200 μm in B; 20 μm in D.

At postnatal day 0 (P0), we detected a small number of tdTomato+ cells in kidneys from both lines (Figure 1 and Figure 1—figure supplement 1). A previous fate-mapping strategy that labels all HSC-derived cells indicated that 40% to 50% of tissue-resident macrophages in the young adult kidney originate from HSC; the remainder was inferred to derive from yolk-sac hematopoiesis (Schulz et al., 2012). Consistent with this inference, we found CX3CR1-lineage labeled cells in kidneys from birth, with numbers increasing progressively over time (2 weeks, 2 months and 6 months; Figure 1, B and C). Surprisingly, we observed an unexpected large increase in the proportion of tdTomato-positive cells relative to total F4/80-positive cells at 6 months, especially in the cortex and outer medulla, despite no significant change in the number of F4/80-positive cells per section. This increase was maintained up to 1 year (Figure 1B). The tdTomato-labeled cells were positive for mature macrophage markers, F4/80 and CD64 (Figure 1D), (Viehmann et al., 2018; Brähler et al., 2018). By contrast, we observed only a few F4/80-positive CSF1R-lineage cells inside the kidneys (Figure 1—figure supplement 1, B and C), as reported previously (Schulz et al., 2012). As CSF1R+ and CX3CR1+ yolk-sac macrophages represent a developmental sequence of tissue-resident macrophages derived from EMP, the observed low labeling of CSF1R-lineage cells might be attributable to differences in labeling efficacies and migration kinetics of progenitors.

To further delineate the chronological shift of renal macrophages, we examined the HSC contribution to renal macrophages using the Flt3-Cre; Rosa26tdTomato mouse line (Mass, 2018; Yahara et al., 2020; Benz et al., 2008). This mouse line irreversibly labels fetal and adult HSC-derived multipotent hematopoietic progenitors and their progeny with tdTomato expression. Consistent with the increase of yolk-sac-derived renal macrophages with age, we observed a decreased number of F4/80+ tdTomato+ cells in the kidneys from 6-month-old mice compared to those from 2-month-old mice (Figure 2, A and B). These data demonstrate that EMP-derived CX3CR1+ yolk-sac macrophages and their descendants are major contributors to the resident renal macrophage population in aged kidneys.

Age-dependent decrease of HSC-derived F4/80+ cells in the kidneys.

The Flt3-Cre; Rosa26tdTomato mouse line was used to examine the contribution of HSC-derived tissue-resident macrophages. (A) Distribution of F4/80+ Flt3-lineage cells in postnatal kidneys. The kidneys were analyzed at the indicated times (n = 4–5). HSC-derived cells and their progeny are irreversibly tagged with tdTomato. (B) Percentage of tdTomato+ F4/80+ to total F4/80+ cells. Note that the number of tdTomato+ F4/80+ cells decreases with age. Data are represented as means ± S.D. *, p<0.05; **, p<0.01.

Our findings raise the question of the underlying mechanisms responsible for the chronological shift of macrophage composition. We first tested whether the proliferation of yolk-sac-derived renal macrophages can potentially contribute to their population dynamics. CX3CR1-lineage-labeled cells that express the proliferation marker, Ki67, are present in kidneys from birth to 6 months of age, indicating that yolk-sac-derived macrophages retain the potential to expand in numbers through proliferation (Figure 3, A and B). We further found that higher percentages of CX3CR1-lineage-labeled F4/80+ cells express Ki67 in comparison to tdTomato-negative F4/80+ cells at 2 weeks and 2 months of age, suggesting that CX3CR1-lineage cells have a higher proliferating capacity (Figure 3C).

CX3CR1-positive yolk-sac macrophage descendants proliferate locally in the kidneys.

(A) Schematic of fate-mapping strategy. (B and C) CX3CR1-lineage macrophages proliferate in neonatal and aged kidneys. Cx3cr1CreERt; Rosa26tdTomato mice were treated with 4-hydroxytamoxifen (4-OHT) at E9.5 (n = 3–5). Arrowheads: Ki67+ CX3CR1-lineage cells. Percentage of Ki67+ proliferating cells are shown in C. Note that a higher percentage of CX3CR1-lineage F4/80+ cells (tdTomato+) are Ki67-positive compared to tdTomato F4/80+ cells. Data are represented as means ± S.D. *, p<0.05; **, p<0.01; n.s., not significant. Scale bars: 10 μm.

Another possibility is recruitment of yolk-sac-derived macrophages from extra-renal reservoirs through the circulation. To test this hypothesis, we generated a parabiotic union between young Cx3cr1GFP/+ and Cx3cr1CreERt; Rosa26tdTomato mice that had been exposed to 4-OHT in utero at E9.5 (Figure 4A). Effective blood sharing between the pair was confirmed by detecting Cx3cr1-promoter-driven GFP expression in bone marrow cells derived from both mice (data not shown). When analyzed at 5 weeks after parabiosis, we found a few tdTomato+ cells in the extravascular, interstitial area of the cortex and medulla of the Cx3cr1GFP parabiont kidneys (0.105 ± 0.04% of total F4/80+ cells; Figure 4, B and C). We also found that a significantly higher percentage of tdTomato-positive cells expresses Ki67 in the kidneys of Cx3cr1GFP mice compared to the tdTomato-positive cells in the kidneys of Cx3cr1CreERt; Rosa26tdTomato mice (Ki67+tdTomato+ relative to tdTomato+; 28.55 ± 8.23% vs. 3.16 ± 0.74%) (Figure 4, D and E). While further investigation is required, these results suggest that the tdTomato-positive circulating progenitors may have a significant proliferative capacity and can slowly contribute to the adult renal macrophage pool. Currently, the origin of the circulating CX3CR1-lineage cells is not known but we speculate that one site is the spleen, which we recently identified as a reservoir of CX3CR1+ yolk-sac macrophages (Yahara et al., 2020).

Figure 4 with 1 supplement see all
CX3CR1-positive yolk-sac macrophage descendants are recruited into adult kidneys from the circulation.

(A) Schematic of parabiotic experiments. (B and C) Localization of tdTomato+ CX3CR1-lineage cells in Cx3cr1GFP kidneys. The tdTomato-positive cells were lineage-labeled in Cx3cr1CreERt; Rosa26tdTomato mice in utero at E9.5. They migrated into the parabiont Cx3cr1GFP kidneys from circulation. Note that tdTomato+ cells were detected in extravascular interstitium (n = 4 per group). Endomucin; an endothelial cell marker. *, lumen of capillaries. Arrowheads, CX3CR1-lineage cells from circulation. (D and E) Circulation-derived CX3CR1-lineage cells proliferate in adult kidneys. Arrowheads, CX3CR1-lineage cells from circulation. GFP fluorescence was lost during the antigen retrieval process to detect Ki67. Percentages of Ki67+tdTomato+ cells relative to tdTomato+ cells in the kidneys of indicated genotype are shown in E (n = 4 per group). Note that tdTomato+ cells in Cx3cr1GFP kidneys are derived from circulation. Data are represented as means ± S.D. *, p<0.05. Scale bars: 10 μm in B and C; and in 20 μm in D. Legends for the Supplementary Figures.

In conclusion, we have shown here that the proportion of yolk-sac-derived, CX3CR1-positive, macrophages increases significantly in the kidney with age, with recruitment from the circulation and proliferation being two possible mechanisms. Our findings provide a foundation for future studies to investigate the functional heterogeneity of ontogenically distinct renal macrophages in younger versus aged kidneys. These future studies may provide novel insight into age-related susceptibility of the kidney to acute and chronic diseases.

Materials and methods

Key resources table
Reagent type
(species) or resource
DesignationSource or referenceIdentifiersAdditional
information
Genetic Reagent (M. musculus)Cx3cr1CreERtThe Jackson laboratoryRRID:IMSR_JAX:020940
Genetic Reagent (M. musculus)Csf1r-CreERt
(aka, Csf1r-Mer-iCre-Mer)
The Jackson laboratoryRRID:IMSR_JAX:019098
Genetic Reagent (M. musculus)Flt3-CreRRID:IMSR_EM:11790Flt3-Cre mice were bred into the C57BL6/J background for six generations by the Shinohara lab.
Genetic Reagent (M. musculus)Rosa26tdTomatoThe Jackson laboratoryRRID:IMSR_JAX:007914
Genetic Reagent (M. musculus)Cx3cr1GFPThe Jackson laboratoryRRID:IMSR_JAX:005582
AntibodyAnti-F4/80 (Rat monoclonal)Bio-Rad
(MCA497)
RRID:AB_2098196Clone C1:A3-1
IF: 1:100
AntibodyAnti-CD64 (Rat monoclonal)Bio-Rad
(MCA5997)
RRID:AB_2687456Clone AT152-9
IF: 1:200
AntibodyAnti-Endomucin (Rat monoclonal)Abcam
(ab106100)
RRID:AB_10859306Clone V.7C7.1
IF: 1:100
AntibodyAnti-Ki67 (Rat monoclonal)eBioscience
(14-5698-82)
RRID:AB_10854564Clone SolA15
IF: 1:200
AntibodyAnti-Ki67 (Rabbit monoclonal)Thermo (MA5-14520)RRID:AB_10979488Clone SP6
IF: 1:200
AntibodyAnti-dsRed
(Rabbit polyclonal)
Rockland
(600-401-379)
RRID:AB_2209751IF: 1:200
Software, algorithmImageJNIH,
Bethesda, MD
(Version 1.52P)
RRID:SCR_003070https://imagej.nih.gov/ij/
Software, algorithmGraphPad PrismRRID:SCR_002798https://www.graphpad.com/scientific-software/prism/

Study approval

Request a detailed protocol

All experiments were performed according to IACUC-approved protocols (A051-18-02 and A196-16-0).

Animals

The mouse lines were from the Jackson Laboratory (Stock No: 019098; 020940; 007914; and 005582). The Flt3-Cre mouse line was kindly provided from Dr. K Lavine (Washington University, St. Louis, MO). 75 μg/g body weight of 4-hydroxytamoxifen (4-OHT; Sigma Aldrich, St. Louis, MO) dissolved in corn oil (Sigma Aldrich) was intraperitoneally administered into pregnant dams with 37.5 μg/g body weight progesterone (Sigma Aldrich) to avoid fetal abortions. Mice without 4-OHT treatment were used for the specificity of tdTomato signals (Figure 1—figure supplement 2 and Figure 4—figure supplement 1). Animals were allocated randomly into experimental groups and analyses. To determine experimental group sizes, data from our previous study were used to estimate the required numbers.

Parabiosis surgery

Request a detailed protocol

Parabiosis surgery was performed as previously described (Yahara et al., 2020). Briefly, a longitudinal skin incision was performed from elbow to knee joint to each mouse. The two mice were connected by suture. Parabionts were separated 5 weeks after the surgery. The mice were euthanized, and kidneys and spleen were harvested for analyses. Sham surgery was performed in the same way except for joining two animals.

Antibodies and sample processing

Request a detailed protocol

Primary antibodies: F4/80 (Bio-Rad, Hercules, CA; clone Cl:A3-1), CD64 (Bio-Rad; clone AT152-9), Endomucin (Abcam; Cambridge, UK; clone V.7C7.1), Ki67 (eBioscience, San Diego, CA; clone SolA15 and Thermo, Waltham, MA; clone SP6), and dsRed (Rockland, Limerick, PA; cat. #600-401-379). Fluorescent-labeled secondary antibodies were used appropriately. 7 μm cryosections were stained using standard protocols. Heat-induced antigen retrieval was performed using pH6.0 sodium citrate solution (eBioscience). Images were captured using Axio imager and 780 confocal microscopes (Zeiss, Oberkochen, Germany). More than three randomly selected areas from 3 to 5 kidneys were imaged and quantified using ImageJ.

Statistics and reproducibility

Request a detailed protocol

Results are expressed as means ± SD. Unpaired t-test was used for comparing two groups. One-way ANOVA followed by Dunnett's correction was used for multiple group comparison. A P-value less than 0.05 was considered statistically significant.

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30

Decision letter

  1. Florent Ginhoux
    Reviewing Editor; Agency for Science Technology and Research, Singapore
  2. Satyajit Rath
    Senior Editor; Indian Institute of Science Education and Research (IISER), India
  3. Elvira Mass
    Reviewer; University of Bonn

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

Your study provides compelling evidence of an age-dependent expansion of kidney macrophages, and the first evidence of a yolk-sac derived circulating macrophage progenitor. Your findings will be of high interest for the macrophage community.

Decision letter after peer review:

Thank you for submitting your article "Yolk-sac-derived macrophages progressively expand in the mouse kidney with age" for consideration by eLife. Your article has been reviewed by two peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Satyajit Rath as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Elvira Mass (Reviewer #1).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision and identified the major revisions needed to help you prepare a revised submission. The detailed comments of the reviewers are also provided for ease of reference.

Essential revisions:

Below are comments highlighting the important points to consider:

1) The control experiments requested by reviewer #1 should be carefully addressed.

2) Further evidence of local proliferation versus recruitment as requested by reviewer #2 should be provided to strengthen the manuscript.

3) Potential functional implication/s of YS-MF expansion in the kidney should be investigated to highlight the novelty of the current manuscript compared to the parallel submission suggesting recruitment from the spleen.

Reviewer #1:

Ide et al. characterize the origin of renal macrophages during aging and provide evidence of a yolk-sac derived macrophage precursor that persists in adult life. Using 2 different fate-mapping models they show that pre-macrophages (Cx3cr1+ at E9.5) give rise to a small proportion of resident macrophages in the kidney, which will expand throughout aging. The expansion can be explained by proliferation of yolk-sac derived cells and/or a circulating macrophage precursor originating in the spleen, which continuously contributes to the pool of macrophages as shown by parabiosis experiments.

The authors provide a compelling data set of an age-dependent expansion of kidney macrophages, and the first evidence of a yolk-sac derived circulating macrophage progenitor. Especially that latter is of high interest for the macrophage community. However, to ascertain both the expansion and the presence of a circulating progenitor in adult and aging mice, some experiments need to be performed/reanalyzed.

1) Control experiments with Cx3cr1CreER; Rosa26tdTomato adult mice (not only P3) that have not been injected with tamoxifen should be performed (including parabiosis) to make sure that there is no leakiness of the reporter over time (Fonseca et al. 2017, J. Neuroinflammation). Particularly, the tdTomato driver is prone to leakiness, therefore another reporter could also be considered (e.g. dsRed or YFP).

2) Figure 3B: the authors claim that they see tdTomato signal in GFP+ macrophages of the Cx3cr1-GFP parabiont. It is however difficult to see an overlap of the signals. Representation of single colours would be helpful. Further, assessment of the% as done for Figure 1 would give the reader a better understanding of how often macrophage progenitors replace resident macrophages during adulthood.

Reviewer #2:

The current notion is that, with the notable exception of the CNS MF, HSC-derived cells seed peripheral organs with age and replace original YS-derived MF macrophages. The finding of Ide et al. that following this initial dilution of YS-MF there can be a resurgence of YS-derived kidney macrophages is provocative and definitively deserves attention. The source of these cells merits in depth study, be it that they arise from local proliferation, the recruitment from a splenic YS-MF depot, or even from EMP / YS-MF precursors persisting into adulthood that might so far have missed attention. As such this study is very interesting.

Specifically, a key observation and potentially conceptual advance of the present manuscript is that the late YS MF burst might derive from a splenic YS-MF infiltrate. This is supported by the parabiosis experiment, though experimentally not strictly dissected from local proliferation. If this aspect could be further carved out during a revision this would significantly strengthen the paper.

To support the stringency their experimentation but avoid duplication, the authors included a manuscript that is under revision elsewhere, and shows convincing proper controls.

However, this study by Yahara et al. also suggests the existence of splenic YS-macrophages that in this case are mobilized for osteoclast generation. This considerably compromises novelty of the study by Ide et al. To balance this shortcoming, and justify a re-publication of this finding in another organ, the authors would I believe have to include evidence for a specific functional implication of the YS-MF expansion for the kidney. As is the study feels more like an add-on of the Yahara et al. study that hardly stands on its own, not the least because it heavily relies on a yet unpublished paper for proper judgement.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Yolk-sac-derived macrophages progressively expand in the mouse kidney with age" for further consideration by eLife. Your revised article has been evaluated by Satyajit Rath as the Senior Editor, a Reviewing Editor and two peer reviewers.

The manuscript has been improved but there are some remaining issues that need to be addressed before acceptance, as outlined below:

1) The new Figure 4D, E: this staining seems to be unspecific. The authors have to include a macrophage marker such as Iba1 or F4/80 to make sure that the Ki67/tdTomato signal stems from a macrophage.

Add the word 'local' to the last sentence of the Abstract "This chronological shift in macrophage composition involves local cellular proliferation and recruitment from circulating progenitors and may contribute to the distinct immune responses, limited reparative capacity, and increased disease susceptibility of kidneys in the elderly population."

https://doi.org/10.7554/eLife.51756.sa1

Author response

Reviewer #1:

[…] The authors provide a compelling data set of an age-dependent expansion of kidney macrophages, and the first evidence of a yolk-sac derived circulating macrophage progenitor. Especially that latter is of high interest for the macrophage community. However, to ascertain both the expansion and the presence of a circulating progenitor in adult and aging mice, some experiments need to be performed/reanalyzed.

1) Control experiments with Cx3cr1CreER; Rosa26tdTomato adult mice (not only P3) that have not been injected with tamoxifen should be performed (including parabiosis) to make sure that there is no leakiness of the reporter over time (Fonseca et al. 2017, J. Neuroinflammation). Particularly, the tdTomato driver is prone to leakiness, therefore another reporter could also be considered (e.g. dsRed or YFP).

We thank reviewer 1 for raising this important point. We have performed control experiments, as suggested. We also performed multiple experiments, which include two additional protocols to fate map the first wave of pre-macrophages (4-OH tamoxifen injection at E8.5) and hematopoietic stem cell-derived population (Flt3-Cre; Rosa26tdTomato). The new data summarized below strongly support our original notion that yolk-sac derived macrophages expand with age in kidneys.

1) No tamoxifen control. In our recent cohort, we did not observe any basal Cre activity without tamoxifen injection in the spleen, liver, and kidneys of 2-month-old Cx3cr1CreERt; Rosa26tdTomato mice (0 out of 3 biological replicates: See Figure 1—figure supplement 1C). Our results are consistent with the recent report that showed our Cx3cr1CreERt(Jung); Rosa26tdTomato mouse line, which was established by Dr. Steffen Jung, is significantly less leaky compared to another commonly used mouse line Cx3cr1CreERT2 (WganJ); Rosa26tdTomato (Dick SA et al., 2019).

2) No tamoxifen control (Sham surgery group). We further tested whether surgical procedures activate the Cre activity without tamoxifen injection. While we did not observe any Cre basal activity in 2 out of 3 mice tested, we observed a moderate Cre activity in one of the animals (Figure 4—figure supplement 1). In contrast, we observed robust labeling of CX3CR1-lineage cells in all the spleens from 4-OH tamoxifen-treated animals (7 out of 7 spleens, 4-OH tamoxifen was given at E9.5). As shown in Figure 4, we also observed consistent and reproducible results when we analyzed the kidneys of parabiotic animals. Importantly, we observed that the circulation-derived CX3CR1-lineage cells possess a profound proliferative capacity (Figure 4E). Collectively, our data demonstrate that our lineage labeling strategy can effectively identify a yolk-sac derived circulating macrophage progenitor population.

3) Fate mapping of the first wave of pre-macrophages (CX3CR1+ yolk-sac macrophages). We lineage labeled the first wave of pre-macrophages using Cx3cr1CreERt; Rosa26tdTomato mouse line. We injected 4-OH tamoxifen at 8.5 dpc into the pregnant dam and harvested the kidneys at 2 months after birth. We observed significantly less tdTomato-labeled cells in the kidneys of 2-month-old animals compared to those with E9.5 labeling (Author response image 1). These data demonstrate that our lineage labeling strategies effectively fate map different stages of premacrophages.

Author response image 1
Comparison of fate-mapping strategies using different 4-OH tamoxifen treatment protocols.

4-hydroxytamoxifen (4-OHT) was injected once into pregnant dams at 8.5 dpc or 9.5 dpc, and the offspring were analyzed at 2 months of age. The first wave of pre-macrophages can be lineage traced with 4-OH tamoxifen injection at 8.5 dpc.

Note that E9.5 4-OH tamoxifen treatment more efficiently fatemaps the pre-macrophage-derived cells in the 2-month-old kidneys (5-fold increase of tdTomato-positive cells). Image-J was used for quantification. DAPI was used for nuclear staining.

4) Hematopoietic stem cell (HSC)-derived macrophages. We further tested whether the renal macrophages of HSC origin decrease with age using the Flt3-Cre; Rosa26tdTomato mouse line (Mass, 2018; Benz et al., 2008). We observed a decreased number of F4/80+ tdTomato+ cells in the kidneys from 6-month-old animals compared to those from 2-month-old animals (see Author response image 2). This data mirrors our data of YS-derived macrophages (see Figure 1) and supports our original finding of the dynamic temporal changes of renal macrophage composition.

Author response image 2
Age-dependent decline of HSC-derived F4/80-positive cells in kidneys.

The Flt3-Cre; Rosa26tdTomato mouse line was used to delineate the contribution of HSC-derived tissue resident macrophages. Please note that the number of tdTomato+ F4/80+ cells (yellow), which indicates HSC origin, decreases with age. *P<0.05, **P<0.01. Image-J was used for quantification. Flt3-Cre mice were kindly provided by Dr. K. Lavine (Washington University, St. Louis MO). Flt3-Cre mice were bred into the C57BL6/J background for 6 generations by the Shinohala lab.

5) No tdTomato+ cells in the bone marrow CD45+ cells. We have recently published that our lineage labeling strategy did not label CD45+ cells in the bone marrow of adult animals at 2 months old (Yahara et al., 2020), while there are many CX3CR1-positive cells in this population. We extended these analyses to 6 month-old animals, which were exposed to 4-OH tamoxifen in utero at E9.5. We found that there was “no increase” of tdTomato labeled cells in the CD45+ cells over time (biological replicate n=4, Author response image 3).

We obtained kidney tissues from this cohort of animals for Figure 1. These data further support our original contention that our fate-mapping strategy is selective and efficient.

Author response image 3
No increase of CD45+ tdTomato+ cells in bone marrow of 6 month-old Cx3cr1CreERt; Rosa26tdTomato mice.

We analyzed thenumber of tdTomato+ CD45+ cells in bone marrow (N=4). Compared to 2-month-old bone marrow (0.02% of CD45+ cells, Yahara et al., 2020), the number of tdTomato+CD45+ cells did not increase with age (0.025%).

2) Figure 3B: the authors claim that they see tdTomato signal in GFP+ macrophages of the Cx3cr1-GFP parabiont. It is however difficult to see an overlap of the signals. Representation of single colours would be helpful. Further, assessment of the% as done for Figure 1 would give the reader a better understanding of how often macrophage progenitors replace resident macrophages during adulthood.

We agree with the reviewer that the images and descriptions were unclear. In our parabiosis experiments, we created parabiotic unions for 5 weeks between Cx3cr1CreERt; Rosa26tdTomato mice (exposed to 4-OH tamoxifen at E9.5) and Cx3cr1GFP mice. To identify the contribution of circulation derived macrophage progenitors of YS origin in renal macrophage homeostasis, we analyzed the kidneys from Cx3cr1GFP mice. We considered tdTomato-lineage-labeled cells in the kidneys of Cx3cr1GFP mice as circulation-derived cells. As suggested by this reviewer, we provide single-color images of Figure 4B (see Author response image 4).

We also quantified the tdTomato-labeled cells (0.105 ± 0.04% of total F4/80-positive cells). While the number of circulation-derived tdTomato+ cells are low, they showed a significantly high proliferation rate (see Figure 4E).

Author response image 4
Merged and single color panels of Figure 4B.

The tdTomato-positive cells in the Cx3cr1GFP kidneys are derived from the parabiont animals (Cx3cr1CreERt; Rosa26tdTomato) through circulation. GFP-positive cells are resident Cx3cr1-expressing cells.

Reviewer #2:

[…] A key observation and potentially conceptual advance of the present manuscript is that the late YS MF burst might derive from a splenic YS-MF infiltrate. This is supported by the parabiosis experiment, though experimentally not strictly dissected from local proliferation. If this aspect could be further carved out during a revision this would significantly strengthen the paper.

We thank this reviewer for positive and constructive comments to improve our manuscript in clarity and novelty. As suggested by this reviewer, we investigated the proliferative capacity of yolk-sac-derived macrophages. We observed the highest proliferative capacity of circulating progenitor-derived macrophages (28.55 ± 8.23%) compared to total CX3CR1-lineage cells (3.16 ± 0.74%) and other lineage macrophages (1.13 ± 0.85%) in the adult kidneys (see Author response image 5, and Figure 3 and 4).

Author response image 5
Circulation-derived CX3CR1-lineage cells have a profound proliferative capacity.

The percentage of Ki67+ cells is shown. Data are represented as means ± S.D. The mice were exposed to 4-OH tamoxifen in utero at E9.5. The tdTomato+ CX3CR1-lineage cells are descendants of yolk-sac macrophages. Data are derived and summarized from Figures 3 and 4 (see representative images in these figures). Progenitor, circulation-derived CX3CR1-lineage cells identified in Cx3cr1GFP kidneys; Total CX3CR1-lineage cells, macrophages derived from CX3CR1+ yolk-sac macrophages in the kidneys of Cx3cr1CreERt; Rosa26tdTomato mice; Other lineages, F4/80+tdTomato macrophages in the kidneys of Cx3cr1CreERt; Rosa26tdTomato mice. Image-J was used for quantification.

To support the stringency their experimentation but avoid duplication, the authors included a manuscript that is under revision elsewhere, and shows convincing proper controls.

However, this study by Yahara et al. also suggests the existence of splenic YS-macrophages that in this case are mobilized for osteoclast generation. This considerably compromises novelty of the study by Ide et al. To balance this shortcoming, and justify a re-publication of this finding in another organ, the authors would I believe have to include evidence for a specific functional implication of the YS-MF expansion for the kidney. As is the study feels more like an add-on of the Yahara et al. study that hardly stands on its own, not the least because it heavily relies on a yet unpublished paper for proper judgement.

We thank this reviewer for carefully reviewing our manuscripts. The related manuscript by Yahara et al. is now published in Nature Cell Biology (Yahara et al., 2020). As this reviewer pointed out, we reported the potential contribution of YS-derived macrophage progenitors in “bone” homeostasis and repair. However, we did not know whether the identified YS-derived macrophage progenitors contribute to homeostasis of tissue-resident macrophages in visceral organs as the bone is a highly specialized unique organ, which is mineralized.

In the current manuscript, we could identify the progenitor-derived macrophages of circulation origin in adult kidneys. We also found that these progenitors possess impressive proliferative capacity (Figure 4D and E; Author response image 5). Our result supports the novel concept that “yolk-sac-derived macrophage progenitors circulate and contribute to tissue-resident macrophage pool in diverse niches in adults”. Our data may be widely generalized to tissue macrophage homeostasis in other visceral organs.

We also believe our report has significant biological and clinical implications. Clinically, aging is one of the major risk factors of kidney diseases, which afflict around 10% of the population worldwide. A recent elegant report highlighted that renal inflammatory responses after ischemia-reperfusion injury are primarily dependent on the age of mice (Sato et al., 2016). They showed that aged kidneys exhibit significantly more severe inflammation and limited renal repair compared to the young. The aged kidneys produced inflammatory cytokines (ex. osteopontin and Tnf-α), which are known to be produced by macrophages, at higher levels compared to those of young kidneys (Sato et al., 2016).

Our observed age-dependent chronological shift in macrophage composition may explain these distinct immune responses, limited organ reparative ability, and poor outcomes of aged kidneys. We would like to address these exciting questions as future studies. We are aware of the eLife’s excellent mechanism, “Research Advances”, which is intended to publish following-up findings after a publication in eLife. We hope to publish future findings as a Research Advances in eLife.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

The manuscript has been improved but there are some remaining issues that need to be addressed before acceptance, as outlined below:

1) The new Figure 4D, E: this staining seems to be unspecific. The authors have to include a macrophage marker such as Iba1 or F4/80 to make sure that the Ki67/tdTomato signal stems from a macrophage.

We thank the editors for raising this important point. As mentioned above, we rephrased the high potential of circulating CX3CR1-lineage progenitors less conclusive.

Regarding the septicity of the signal, we have the following supporting evidence that shows the tdTomato signal in Figure 4D and E is specific to “CX3CR1-lineage cells”, as we communicated in the previous letter.

– We observed a morphological variation of CX3CR1-lineage cells in the parabiont Cx3cr1GFP kidneys from oval to the elongated shape (see Author response image 6B). This data is consistent with our lineage-tracing analyses in Figures 1 and 3, as we also observed the similar morphological variations in these kidneys (see Figure 3B for high magnification).

– Many of Ki67-positive CX3CR1-lineage cells in parabiont Cx3cr1GFP kidneys are oval-shaped, but Ki67-negative ones have elongated shapes. We speculate that the circulation-derived CX3CR1-lineage cells spread their processes to survey the renal environment after proliferation in situ.

– The oval CX3CR1-lineage cells are also stained with DAPI, excluding a possibility that the signal is an artifact (see Author response image 6A).

– We did not observe any tdTomato-positive cells in negative control kidneys without tamoxifen injection.

Author response image 6
Morphological variation of CX3CR1-lineage cells.

The CX3CR1-lineage tdTomato-positive cells in the Cx3cr1GFP kidneys are derived from the parabiont animals (Cx3cr1CreERt; Rosa26tdTomato) through circulation. The Cx3cr1CreERt; Rosa26tdTomato mice were exposed to 4-OH tamoxifen in utero at E9.5. (A) Ki67+ CX3CR1-lineage cells are positive for DAPI (nuclear staining). (B) The CX3CR1-lineage tdTomato-positive cells exhibit oval and elongated shapes. Scale bars: 2 µm in A and 5 µm in B.

Add the word 'local' to the last sentence of the Abstract "This chronological shift in macrophage composition involves local cellular proliferation and recruitment from circulating progenitors and may contribute to the distinct immune responses, limited reparative capacity, and increased disease susceptibility of kidneys in the elderly population."

We have added the word “local” accordingly.

https://doi.org/10.7554/eLife.51756.sa2

Article and author information

Author details

  1. Shintaro Ide

    1. Division of Nephrology, Department of Medicine, Duke University School of Medicine, Durham, United States
    2. Regeneration Next, Duke University, Durham, United States
    Contribution
    Data curation, Formal analysis, Investigation, Writing - review and editing
    Contributed equally with
    Yasuhito Yahara
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-9301-211X
  2. Yasuhito Yahara

    1. Regeneration Next, Duke University, Durham, United States
    2. Department of Orthopedic Surgery, Duke University School of Medicine, Durham, United States
    3. Department of Orthopedic Surgery, Faculty of Medicine, University of Toyama, Toyama, Japan
    Contribution
    Conceptualization, Resources, Investigation, Methodology, Writing - review and editing
    Contributed equally with
    Shintaro Ide
    Competing interests
    No competing interests declared
  3. Yoshihiko Kobayashi

    Department of Cell Biology, Duke University School of Medicine, Durham, United States
    Contribution
    Resources, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-7031-1478
  4. Sarah A Strausser

    Division of Nephrology, Department of Medicine, Duke University School of Medicine, Durham, United States
    Contribution
    Investigation, Project administration
    Competing interests
    No competing interests declared
  5. Kana Ide

    Division of Nephrology, Department of Medicine, Duke University School of Medicine, Durham, United States
    Contribution
    Formal analysis, Investigation, Project administration
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-2845-8481
  6. Anisha Watwe

    Division of Nephrology, Department of Medicine, Duke University School of Medicine, Durham, United States
    Contribution
    Investigation
    Competing interests
    No competing interests declared
  7. Shengjie Xu-Vanpala

    Department of Immunology, Duke University School of Medicine, Durham, United States
    Contribution
    Resources, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-2716-6230
  8. Jamie R Privratsky

    Division of Nephrology, Department of Medicine, Duke University School of Medicine, Durham, United States
    Contribution
    Resources, Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-3598-4911
  9. Steven D Crowley

    Division of Nephrology, Department of Medicine, Duke University School of Medicine, Durham, United States
    Contribution
    Resources, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-1838-0561
  10. Mari L Shinohara

    1. Department of Immunology, Duke University School of Medicine, Durham, United States
    2. Department of Molecular Genetics and Microbiology, Duke University School of Medicine, Durham, United States
    Contribution
    Resources, Funding acquisition, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-6808-9844
  11. Benjamin A Alman

    1. Regeneration Next, Duke University, Durham, United States
    2. Department of Orthopedic Surgery, Duke University School of Medicine, Durham, United States
    Contribution
    Resources, Supervision, Funding acquisition, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
  12. Tomokazu Souma

    1. Division of Nephrology, Department of Medicine, Duke University School of Medicine, Durham, United States
    2. Regeneration Next, Duke University, Durham, United States
    Contribution
    Conceptualization, Supervision, Funding acquisition, Investigation, Writing - original draft, Project administration, Writing - review and editing
    For correspondence
    tomokazu.souma@duke.edu
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-3285-8613

Funding

American Society of Nephrology (Career Developmental Grant)

  • Tomokazu Souma

Duke University School of Medicine (Start-up Fund)

  • Tomokazu Souma

National Institute on Aging (R01 AG049745)

  • Benjamin A Alman

National Institute of Allergy and Infectious Diseases (AI088100)

  • Mari L Shinohara

American Heart Association (Postdoctoral fellowship)

  • Shintaro Ide

Kanzawa Medical Research Foundation (Fellowship)

  • Yasuhito Yahara

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

This study was supported by grants from the National Institute on Aging R01 AG049745 to BA, and NIH-AI088100 to MLS, and the American Society of Nephrology Career Developmental Grant and Duke Nephrology Start-up Fund to TS. SI and YY are supported in part by fellowship grants from the American Heart Association and the Kanzawa Medical Research Foundation, respectively. We thank Drs. Brigid Hogan, Benjamin Thomson, and Myles Wolf for comments and helpful suggestions on the manuscript. We also thank Mr. Puviindran Nadesan for his technical support. Imaging was performed at the Duke Light Microscopy Core Facility supported by the shared instrumentation grant (1S10RR027528-01).

Ethics

Animal experimentation: This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All animals were used according to the approved protocols (A051-18-02 and A196-16-0) by the Institutional Animal Care and Use Committee of Duke University.

Senior Editor

  1. Satyajit Rath, Indian Institute of Science Education and Research (IISER), India

Reviewing Editor

  1. Florent Ginhoux, Agency for Science Technology and Research, Singapore

Reviewer

  1. Elvira Mass, University of Bonn

Publication history

  1. Received: September 10, 2019
  2. Accepted: April 16, 2020
  3. Accepted Manuscript published: April 17, 2020 (version 1)
  4. Version of Record published: May 7, 2020 (version 2)

Copyright

© 2020, Ide et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,055
    Page views
  • 173
    Downloads
  • 0
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Genetics and Genomics
    2. Human Biology and Medicine
    Eran Mick et al.
    Research Article
    1. Human Biology and Medicine
    AJ Venkatakrishnan et al.
    Research Article