Divergent immediate and delayed effects of juvenile exposure to doxorubicin on the thymus in C57BL/6 mice

Abstract

The long-term effects of doxorubicin (DOX) chemotherapy on thymic immune function in childhood cancer survivors remain inadequately understood. This study explores the immediate and delayed impacts of low-dose DOX on thymic immune populations using a juvenile mouse model. Male mice received intraperitoneal DOX injections (4 mg/kg/week) for three weeks, with evaluations performed at one- and five-weeks post treatment. Thymic samples were collected and analyzed using multi-parameter flow cytometry to assess changes in immune cell composition and phenotype. Additionally, real-time polymerase chain reaction (RT-PCR) was employed to measure gene expression of cytokines and senescence markers. One week after DOX administration, significant thymic atrophy was evident. While mature CD3⁺CD4⁺ T-cell frequency remained unchanged, CD3⁺CD8⁺ T-cells significantly increased, suggesting differential effects on T-cell subsets. PD1⁺ expression increased across naïve and memory CD4⁺ T-cell subsets, suggesting activation or exhaustion. Additionally, Ki67⁺ expression was elevated in naïve and memory CD8⁺ T-cells, indicating enhanced proliferation. Gene expression analysis revealed upregulation of Foxn1, Pax1, Ifnγ, and Il7 whereas Il6 and Il17 were downregulated. Furthermore, Cdkn1a (p21) expression was elevated, suggesting immune dysregulation and early immunosenescence. At five weeks, thymic weight rebounded; however, T-cell subsets displayed persistent perturbations. Central memory and effector memory CD4⁺ T-cells were reduced, while naïve CD4⁺ T-cells showed increased Ki67⁺ expression. In contrast, CD8⁺ T-cells subsets remained largely unchanged, except for a decrease in central memory cells. Although expression of thymus-related genes was normalized, p21 expression remained elevated, suggesting lingering immunosenescence. These findings highlight the complex effects of DOX, including acute thymic atrophy due to T-cell depletion, and a delayed recovery with persistent immunosenescence, underscoring the need for strategies to preserve immune function in childhood cancer survivors.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-98617-5.

Introduction

Doxorubicin (DOX) chemotherapy, a widely employed treatment for cancer, has been shown to cause a transient suppression of immune cell function and heightened vulnerability to infections among cancer patients¹. The thymus is a vital organ in generating an effective immune response as the site where T lymphocytes (T-cells) develop and mature. Chemotherapy can cause deleterious effects on the thymus^2,3 as it induces thymic atrophy, leading to a reduction in the generation of naïve T-cells and compromised cellular immunity function⁴. This reduction alters immune cell homeostasis and the immune response to infectious challenge^4,5. However, upon discontinuation of chemotherapy, thymic atrophy may undergo recovery, potentially giving rise to a hyperplastic process known as rebound thymic hyperplasia^5–7. Thymic hyperplasia is associated with robust thymic regeneration. This regeneration is characterized by an increase in both thymic size and density, concomitant with the restoration of thymic T-cell output^7–9. Although this phenomenon is common in children and adolescents and occasionally observed in young adults, it is rare in older patients^10,11. T-cells are adaptive immune cells that play a central role in the immune system. T-cells arise from hematopoietic stem cells differentiating into lymphoid progenitors within the bone marrow⁵. Within the thymus, these lymphoid progenitors are called thymocytes and undergo a maturation process to generate mature naïve T-cells with rearranged, highly specific, T-cell receptors (TCR). Initially, thymocytes are double negative for CD4 and CD8 and do not express the TCR/CD3. They then begin to express TCR/CD3 and become double-positive for CD4 and CD8 where they then undergo positive and negative selection. In the last maturation step within the thymus, mature naïve T-cells are generated as they become single positive for either CD4 or CD8. Once mature, naïve T-cells then circulate in the blood and lymph to encounter their specific antigen⁵. CD8⁺ T-cells are cytotoxic and aid in the immune response by secreting effector cytokines such as tumor necrosis factor alpha (TNF-α) and interferon-gamma (IFN-γ) and directly killing pathogenic cells through cytotoxic mechanisms¹². CD4⁺ T-cells can be divided further into either T-regs or T-helper cells that have specialized roles in supporting a specific immune response¹². Once T-cells encounter their specific antigen they then differentiate into memory T-cells¹³. Central memory T-cells are long lasting, remaining for years after initial antigen encounter, whereas effector memory T-cells are potent responders and are short lived⁵. Alterations in the development and phenotype in these cells can lead to alterations in the immune response to infections and cancer. One such dysfunctional T-cell fate is that of functional exhaustion. Exhausted T-cells are defined by reduced proliferative capacity, reduced cytokine production, and the upregulation of immune checkpoint proteins, such as programmed cell death protein 1 (PD1)¹⁴. Although PD1 is a canonical marker of T- cell exhaustion it is also expressed on newly activated T-cells to prevent overactivation of T-cells¹⁵. Both atrophy and hyperplasia of the thymus can impact T-cell development and function, potentially affecting immune responses to cancer and other diseases¹⁶.

The enduring impacts of cancer treatment extend to prolonged immune suppression and cellular senescence, persisting for over a decade post-treatment¹⁷. Cancer survivors continue to undergo lasting alterations in their immune profiles well after chemotherapy completion¹⁶. These changes involve reduced T-cell counts, impaired T-cell function, and disruptions in immune checkpoint regulation. Additionally, chemotherapy-induced cellular senescence, characterized by permanent T-cell cycle arrest and functional changes, significantly influences aging and age-related diseases, including cancer¹⁸. The induction of senescence in both cancerous and healthy cells, including immune cells, underscores its role in age-related outcomes. Such immune system modifications heighten susceptibility to infections and weaken the body’s ability to detect and combat cancer recurrence. The present study investigates the intricate and multifaceted impact of DOX chemotherapy on immune cell subpopulations and thymic functionality in juvenile male mice, focusing on both immediate and delayed repercussions. While previous studies have largely focused on the immediate cytotoxic impacts of chemotherapy⁴, there remains a significant gap in understanding its specific influence on thymic cell populations. More importantly, the delayed effects of DOX on the thymus have never been reported in an animal model. This research extends into the relatively unexplored domain of chemotherapy-induced immune perturbation by conducting a thorough analysis of both the immediate and long-term immune responses to DOX.

Materials and methods

Animals

All animal-related experimental procedures were ethically approved by the University of Minnesota Institutional Animal Care and Use Committee (IACUC protocol number: 2106–39176 A). All methods were carried out in accordance with relevant guidelines and regulations and are reported in accordance with ARRIVE guidelines (https://arriveguidelines.org). Four-week-old C57BL/6 N male mice were purchased from Charles River Laboratories (Wilmington, Massachusetts). The mice were housed in a controlled environment with specific pathogen-free (SPF) conditions, maintained in a 14-hour light/10-hour dark cycle at a temperature of 21 ± 2 °C, and provided with ad libitum access to food and water. To ensure acclimatization, mice were introduced to the animal facility one week prior to the commencement of experimental procedures. The overall experimental designs are schematically represented in Figs. 1A and 5A. The results were analyzed from two independent experiments. Commencing at five weeks of age, mice received intraperitoneal injections of either DOX at a dosage of 4 mg/kg/week for three consecutive weeks, resulting in a cumulative dose of 12 mg/kg or an equivalent volume of sterile saline. Weekly assessments of animal weights were conducted, and subsets of mice were humanely euthanized one week after the final dose and another subset after a five-week interval. Euthanasia was performed in a humane manner using decapitation under isoflurane anesthesia. Subsequently, thymic tissues were harvested, cleansed under a microscope in ice-cold phosphate-buffered saline solution, and either promptly flash-frozen in liquid nitrogen or placed in cold media for immune cell isolation and analysis. The frozen specimens were stored at − 80 °C for subsequent analytical procedures.

Fig. 1.

DOX-induced changes in thymus weight and T-cell populations one week after treatment. (A) Experimental design to assess the immediate effects of DOX-induced immune dysfunction. Male mice were exposed to a series of low-dose DOX administrations at 4 mg/kg/week for three consecutive weeks. Mice were euthanized one week after the final treatment to observe immediate effects of DOX. Thymus samples were collected for flow cytometry analysis; (B) Immediate effects of DOX on thymus weights (mg); Bar graphs representing the (C) frequency and (D) cell count (per milligram (mg) of thymus) of immune cells (CD45⁺); Bar graphs representing the frequencies of (E) CD3⁻CD4⁺CD8⁺, (F) CD3⁻CD4⁺, (G) CD3⁻CD8⁺ and cell count of (H) CD3⁻CD4⁺CD8⁺, (I) CD3⁻CD4⁺ and (J) CD3⁻CD8⁺ cells per mg of thymus; n = 8 mice/group. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 vs. saline control group (analyzed by unpaired t-test if normal (B–E, G, J) or Mann-Whitney if non-parametric (F, H, I)). (A) is created with BioRender.

Fig. 5.

DOX-induced changes in thymus weight and T-cell populations five weeks after treatment. (A) Experimental design to assess the delayed effects of DOX-induced immune dysfunction. Male mice were exposed to a series of low-dose DOX administrations at 4 mg/kg/week for three consecutive weeks. Eight DOX-treated and seven saline-treated mice were euthanized five weeks after the final treatment to observe the delayed effects of DOX. Thymus samples were collected for flow cytometry analysis; (B) delayed effects of DOX on the thymus weight (mg); Bar graphs representing the (C) frequency and (D) cell count of hematopoietic cells (CD45⁺); Bar graphs representing the frequencies of (E) double-positive CD3⁻CD4⁺CD8⁺, (F) CD3⁻CD4⁺ cells, (G) CD8⁺ cells and cell count of (H) CD3⁻CD4⁺CD8⁺, (I) CD3⁻CD4⁺ and (J) CD3⁻CD8⁺ cells (in cells per mg thymus), five-week DOX group compared with the five-week saline group. n = 7–8 mice/group. ns p > 0.05, *p < 0.05 and ***p < 0.001 vs. the five-week saline control group (analyzed by unpaired t-test if normal (D–E, H, J) or Mann-Whitney if non-parametric (B, C, F–G, I)). (A) is created with BioRender.

Immune cell isolation and flow-cytometry

Immune cells were isolated from the thymus using mechanical disaggregation. Red blood cells were lysed using ACK lysis buffer. For staining, cells were stained with a fixable viability dye for 25 min at 4 °C in the dark. Cells were then incubated in Fc block (anti-CD16 & anti-CD32) and surface antibodies for 45 min at 4 °C protected from light. For intracellular or nuclear staining, cells were fixed and permeabilized using the BD Cytofix/Cytoperm kit (554715) or eBioscience FOXP3/Transcription factor staining buffer set followed by nuclear antibodies staining for 45 min at 4 °C protected from light. All antibodies used were confirmed to have been validated by their manufacturer. Flow cytometry data were acquired on a BD FACSymphony A3 Cell Analyzer R6609 and analyzed using Flowjo software version 10. Gating strategies are provided in Sup. Figures 2 and 3.

RNA extraction and quantitative real-time PCR

Total RNA was isolated from cryopreserved thymus tissues utilizing Invitrogen TRIzol^® reagent (Thermo Fisher Scientific, Waltham, MA, USA), adhering to the manufacturer’s guidelines. RNA concentration was measured at 260 nm employing a NanoDrop Lite Plus spectrophotometer (Thermo Fisher Scientific). Subsequently, first-strand cDNA was synthesized from 1.5 µg of total RNA, employing the Applied Biosystems high-capacity cDNA reverse transcription kit (Thermo Fisher Scientific), following the manufacturer’s instructions. Quantitative real-time polymerase chain reaction (qRT-PCR) was employed for the quantification of specific mRNA expression, involving PCR amplification of the synthesized cDNA within 384-well optical reaction plates using an Applied Biosystems QuantStudio 5 instrument (Thermo Fisher Scientific). The 20 µL reaction mixture consisted of 1 µL of the cDNA sample, 0.025 µL of 30 µM forward primer, and 0.025 µL of 30 µM reverse primer (attaining a final concentration of 40 nM for each primer). Additionally, it included 10 µL of Applied Biosystems SYBR Green Universal Mastermix (Thermo Fisher Scientific) and 8.95 µL of nuclease-free water. The thermocycler conditions comprised an initial denaturation step at 95 °C for 10 min, followed by 40 PCR cycles of denaturation at 95 °C for 15 s and annealing/extension at 60 °C for 1 min. Gene expression analyses were conducted using primers sourced from previously published studies, detailed in Table 1. Normalization of mRNA expression levels was performed relative to Actb, and the relative expressions were determined utilizing the ΔΔCt method. Melting curve analysis was used to confirm the purity of the final PCR product and the specificity of the primers.

Table 1.

Primer sequences used for real-time PCR experiments.

	Forward	Reverse
Foxn1	GACCTTGGGACTGACCTGGAT	GCCTGTTTCTGCCAGACAA
Pax1	AACCAGCACGGAGTATACAGC	TGTAAGCTACCGAGTGCATCC
Il6	CTCTGGGAAATCGTGGAAAT	CCAGTTTGGTAGCATCCATC
Ifnγ	ACTGGCAAAAGGATGGTGAC	TGAGCTCATTGAATGCTTGG
Tnfα	CCAGACCCTCACACTCAGATCA	CACTTGGTGGTTTGCTACGAC
Il2	AAACTAAAGGGCTCTGACAACACA	CACCACAGTTGCTGACTCATCA
Il4	ATCGGCATTTTGAACGAGGTCA	CATCGAAAAGCCCGAAAGAGTCT
Il7	ACTACACCCACCTCCCGCA	TCTCAGTAGTCTCTTTAGG
Il17	TCCAGAAGGCCCTCAGACTA	AGCATCTTCTCGACCCTGAA
p21	GCCTTAGCCCTCACTCTGTG	AGCTGGCCTTAGAGGTGACA
p16	GGGTTTCGCCCAACGCCCCGA	TGCAGCACCACCAGCGTGTCC
p19	GCCGCACCGGAATCCT	TTGAGCAGAAGAGCTGCTACGT
p53	AGCTTTGAGGTTCGTGTTTGTG	TGGGCAGCGCTCTCTTTG
Actb	TATTGGCAACGAGCGGTTCC	GGCATAGAGGTCTTTACGGATGTC

Histopathological analysis

The tissues were fixed in 10% neutral buffered formalin, transferred to 70% ethanol, and processed into paraffin blocks. From each paraffin block, a single 4 μm section was cut and stained with hematoxylin and eosin. All samples were evaluated by a single, board-certified veterinary pathologist (DMS) that was blinded to study group information. Each sample was evaluated for hemorrhage, necrosis, inflammation, edema, and cystic degeneration with each finding semi-quantitatively scored 0–4 according to the following rubric: 0 = no finding identified, 1 = finding is rare and represents < 5% of the total tissue surface area, 2 = mild severity and affects 5–25% of the tissue surface area, 3 = moderate severity and affects 26–50% of the tissue surface area, and 4 = marked severity and affects > 51% of the tissue surface area.

Statistical analysis

Data is presented as mean ± standard error of the mean (SEM). To assess differences between treatment groups, we performed normality testing (D’Agnostin & Pearson if n ≥ 8; Shapiro-Wik if n < 8). If the groups failed normality testing, we performed Mann-Whitney, if they passed, an unpaired Student’s two-tailed t-test was conducted. The precise statistical method ran on each panel can now be found in each respective figure legend. Statistical significance was established at a probability value of < 0.05. Analysis was done in GraphPad Prism 9.5.1 (La Jolla, CA).

Results

The current study aimed to determine the direct effects of DOX on the thymus in young male mice, specifically examining changes in immune cell numbers and proportions. Male mice were subjected to a regimen of repeated low-dose DOX administration at 4 mg/kg/week for three consecutive weeks (Fig. 1A). To maintain clinical relevance, we delivered an intentionally low level of DOX, 4 mg/kg/week, equivalent to approximately ∼40 mg/m² in humans¹⁹. Our prior work has established the absence of severe morbidity or any mortality when administering 4 mg/kg/week for 3–5 weeks²⁰. One week after the last administration of DOX, mice were euthanized to evaluate the immediate effects of DOX exposure. Notably, no mortality incidents were recorded following the third DOX dose. Male mice subjected to DOX showed a significant decline in body weight one week after the final injection compared to saline treated mice (Sup. Figure 1A and 1B).

In DOX-treated male mice, we investigated the immediate impact of DOX on thymus weight, which revealed a significant reduction in weight compared to the saline group (Fig. 1B). The decrease in thymus weight was maintained when normalized to tibial length and or to body weight, indicating that organ specific effects were occurring (Sup. Figure 1C and D). To determine the effects of DOX on thymic immune cell populations one week after DOX treatment, we identified total immune cell (CD45⁺) frequencies and cells per mg of tissue. There were no differences in the frequency of total immune cells, but there was a marked reduction in the overall number of immune cells per mg of thymus (Fig. 1C and D) in the DOX group when compared to the saline group. We used multi-parameter flow cytometry to examine developing thymocytes (Sup. Figures 2 and 3). We observed a reduced frequency of CD3⁻CD4⁺CD8⁺ cells in the DOX mice compared to saline control one week after the last DOX treatment (Fig. 1E). A significant decrease was observed in the single positive, CD3⁻CD4⁺ cell frequency; however, an increase was seen in the single positive, CD3⁻CD8⁺ cell frequency in the DOX group when compared to the saline group (Fig. 1F and G). Furthermore, the cells per mg of thymus of double-positive CD3⁻CD4⁺CD8⁺, CD3⁻CD4⁺, and CD3⁻CD8⁺ cells were significantly reduced in DOX treated mice compared to those given saline (Fig. 1H, I and J).

To address the phenotype of mature CD3⁺ T-cells, we analyzed naïve and memory status, as well as Ki-67 and PD1 expression (Sup. Figure 3 A)²¹. There was no difference in the frequency of CD3⁺CD4⁺ T-cells between the one week saline and DOX-treated groups (Fig. 2A). We next assessed CD44 vs. CD62L expression to distinguish naïve (Tn), central memory (Tcm), effector memory (Tem), and CD44⁺CD62L⁻ subsets and assess alterations between the DOX and saline groups. We found a significant expansion of CD44⁺CD62L⁻ Tem and a reduction in CD44⁻CD62L⁻ CD4⁺ T-cells in the DOX-treated mice compared to the saline controls (Fig. 2B). Although there was an expansion of Tem, CD4⁺ CD44⁻ T-cells maintained the largest proportion of mature CD4⁺ T-cells within the thymus (Fig. 2C). We next assessed Ki67 expression, as a marker of active proliferation or homeostatic proliferation. There were no significant differences in Ki67⁺ frequency between one week DOX- and saline-treated mice within the CD4⁺ Tn, Tcm, Tem, or CD44⁻CD62L⁻ populations (Fig. 2D). Lastly, we assessed PD1 and found a significant expansion of PD1⁺ cells within all four CD4⁺ T-cell subsets in the DOX mice compared to saline control (Fig. 2E). Consistent with the literature, Tn (CD44⁻CD62L⁺) and CD44⁻CD62L⁻ CD4⁺ T-cells largely did not express PD1, with less than 5% of cell expression.

Fig. 2.

DOX-induced changes in CD4+ T-cell populations one week after treatment. (A) Quantification of CD3⁺CD4⁺ cells as a frequency of CD45⁺ cells. (B) Quantification of CD44 vs. CD62L expression as a frequency of CD3⁺CD4⁺ cells. (C) Stack plot of CD44 vs. CD62L expression as a frequency of CD3⁺CD4⁺ cells. (D) Quantification of Ki67⁺ cells as a frequency of CD4⁺ CD44^+/− CD62L^+/− cells. (E) Quantification of PD1⁺ cells as a frequency of CD4⁺ CD44^+/− CD62L^+/− cells. n = 8 mice/group. ns p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 DOX vs. the saline control group (analyzed by unpaired t-test or multiple unpaired t-tests if normal (D–E) or Mann-Whitney or multiple Mann-Whitney’s if non-parametric (A, B)).

We next examined mature CD3⁺CD8⁺ T-cells in the one week saline and DOX-treated groups. We observed a significant expansion of mature CD3⁺CD8⁺ T-cells in the thymus of DOX-treated mice compared to saline controls (Fig. 3A). We then assessed the frequency of CD44 vs. CD62L cells within the mature CD8⁺ T-cells. There was a significant expansion of CD44⁻CD62L⁺ Tn and a reduction in CD44⁺CD62L⁻ in the DOX-treated mice when compared to saline (Fig. 3B). CD44⁻ cells maintained the largest population of mature CD8⁺ T-cells within the thymus making up more than 90% of CD8⁺ cells (Fig. 3C). We next assessed Ki67⁺ and PD1⁺ frequencies of the respective CD8⁺ T-cell subsets. There was a significant increase in Ki67⁺ Tn, Tem, and CD44⁻CD62L⁻ CD8⁺ T-cells in the one week DOX group compared to saline-treated mice (Fig. 3D). Finally, the frequency of PD1⁺ Tn and CD44⁻CD62L⁻ was significantly increased in the DOX-treated mice, although these frequencies remained below 2% of the total and thus are a rare population (Fig. 3E).

Fig. 3.

DOX-induced changes in CD8⁺ T-cell populations one week after treatment. (A) Quantification of CD3⁺CD8⁺ cells as a frequency of CD45⁺ cells. (B) Quantification of CD44 vs. CD62L expression as a frequency of CD3⁺CD8⁺ cells. (C) Stack plot of CD44 vs. CD62L expression as a frequency of CD3⁺CD8⁺ cells. (D) Quantification of Ki67⁺ cells as a frequency of CD8⁺ CD44^+/− CD62L^+/− cells. (E) Quantification of PD1⁺ cells as a frequency of CD8⁺ CD44^+/− CD62L^+/− cells. n = 8 mice/group. ns p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 DOX vs. the saline control group (analyzed by unpaired t-test or multiple unpaired t-tests if normal (E) or Mann-Whitney or multiple Mann-Whitney’s if non-parametric (A, B, D)).

Immediate effects of DOX administration on thymic gene expression

Foxn1 serves as a central orchestrator of the intricate developmental processes governing the lineage of thymic epithelial cells, facilitating the downstream transcriptional activation of genes essential for thymus organogenesis and the comprehensive differentiation of thymic epithelial cells²². Additionally, Pax1 expression within the thymic epithelium is indispensable for establishing a specialized microenvironment crucial for normal T-cell maturation²³. In mice treated with DOX, an immediate increase in Foxn1 and Pax1 expression were observed, indicating potential alterations in thymic development and T-cell maturation (Fig. 4A and B). The cytokine network within the thymus also plays an important role in thymic maturation and involves dynamic interactions between various cytokines such as interleukin-6 (IL-6), interferon (IFN)-γ, transforming growth factor (TGF)-β, and IL-1²⁴. The immediate impact after DOX treatment in male mice resulted in decreased gene expression of Il6 (Fig. 4C). Contrastingly, Ifnγ expression was significantly increased in DOX-treated mice (Fig. 4D). Thymic developmental cytokines showed distinct patterns of regulation, with Il7 expression significantly increased (Fig. 4E) and Il17 expression decreased (Fig. 4F) in DOX-treated mice compared to saline controls. DOX treatment is associated with thymic degeneration and senescence, leading to thymic atrophy and potential immunosuppression⁴. Regarding thymic senescence, we observed an increase in p21 (Fig. 4G) and p53 gene expression (Sup. Figure 5) but no difference in p16 and a decrease in p19 gene expression (Sup. Figure 5) in the thymus one week following DOX administration.

Fig. 4.

Immediate effects of DOX on the cytokines and senescence markers in thymus. Thymus was harvested from male mice one week following the administration of 4 mg/kg/week DOX or saline for three weeks (n = 5–8 per group). Following the extraction of total RNA, the mRNA expression of (A) Foxn1, (B) Pax1, (C) Il6, (D) Ifnγ, (E) Il7, (F) Il17, and (G) p21 was determined by qRT-PCR. Values were normalized to Actb expressed relative to saline-treated male mice. Values are shown as the means ± SEMs. ns p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 vs. saline control group (analyzed by unpaired t-test if normal (A–G).

Delayed effects of DOX on the thymus

Building on our analysis of the immediate effects of DOX on thymic cell populations, we furthered our investigation to examine its long-term impact. Juvenile male mice underwent a regimen of repeated low-dose DOX administrations at 4 mg/kg/week for three consecutive weeks. Subsequently, a five week period without drug administration was introduced to assess the delayed effects of DOX exposure (Fig. 5A). Final body weight was not different between saline and DOX-treated mice (Sup. Figure 6 A and 6B). In contrast to the immediate impact of DOX-induced reduction in thymus weights (Fig. 1B and Sup. Figure 1A), the delayed impact of DOX on thymus weight revealed a notable increase in thymus weight compared to the saline group (Fig. 5B). This increase in thymus weight was maintained when normalized to tibial length or to body weight (Sup. Figure 6C and 6D).

To address the delayed effects of DOX, we quantified the frequency and numbers of developing thymocytes. There was a diminished frequency of immune cells (CD45⁺), but no difference in cells/mg thymus at five weeks after DOX administration (Fig. 5C and D). There was a higher frequency of double-positive CD3⁻CD4⁺CD8⁺ cells (Fig. 5E). We observed an increase in the frequency of CD3⁻CD4⁺ single-positive thymocyte, but no alterations in the frequency of CD3⁻CD8⁺ cells in the DOX group compared to the saline group (Fig. 5F and G). No differences were found in the double-positive CD3⁻CD4⁺CD8⁺ and CD3⁻CD8⁺ cells, but we observed a significant increase in CD3⁻CD4⁺ cells per mg of thymus (Fig. 5H, I, and J).

We then examined the proportion of mature CD4⁺ T-cell subsets within the thymus. Using the same gating strategy in figure S1A., we detected a reduction in the frequency of mature CD3⁺CD4⁺ cells at five weeks following DOX treatment (Fig. 6A). There was a reduction in the frequency of CD4⁺ Tem and Tcm in the 5wk DOX group, although these two populations average less than 2% and 3% of total CD8⁺ T-cells respectively (Fig. 6B-C). When assessing Ki67 expression, the DOX-treated mice exhibited a significant increase in Ki67⁺ frequency in Tn, Tcm, and CD44⁻CD62L⁻ cells within the thymus (Fig. 6D). The frequency of PD1⁺ Tn was significantly up in the DOX-treated mice; however, this proportion makes up only 0.73% (saline) and 0.90% (DOX) of total naïve cells (Fig. 6E). No changes in PD1 frequency were observed in Tem or Tcm.

Fig. 6.

DOX-induced changes in CD4⁺ T-cell populations five weeks after treatment. (A) Quantification of CD3⁺CD4⁺ cells as a frequency of CD45⁺ cells. (B) Quantification of CD44 vs. CD62L expression as a frequency of CD3⁺CD4⁺ cells. (C) Stack plot of CD44 vs. CD62L expression as a frequency of CD3⁺CD4⁺ cells. (D) Quantification of Ki67⁺ cells as a frequency of CD4⁺ CD44^+/− CD62L^+/− cells. (E) Quantification of PD1⁺ cells as a frequency of CD4⁺ CD44^+/− CD62L^+/− cells. n = 8 mice/group. ns p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 DOX vs. the saline control group (analyzed by unpaired t-test or multiple unpaired t-tests if normal (A) or Mann-Whitney or multiple Mann-Whitney’s if non-parametric (B, D–E)).

Next, we examined the mature CD8⁺ T-cell subsets in the thymus from saline and DOX-treated mice. The mature CD3⁺CD8⁺ T-cell population displayed a significant decrease with DOX-treatment at the five-week timepoint (Fig. 7A). No alterations in frequency were observed within the CD8⁺ Tn or Tem populations; however, there was a significant reduction in Tcm with DOX (Fig. 7B). Consistent with the one-week timepoint, CD44⁻ cells make up the majority of total mature CD3⁺CD8⁺ T-cells within the thymus, regardless of treatment group (Fig. 7C). We observed no alterations in the frequency of Ki67⁺ cells within the CD8⁺ subsets (Fig. 7D). Additionally, there were no changes in the frequency of PD1⁺ cells within Tn, Tcm, Tem, and CD44⁻CD62L⁻ CD8⁺ T-cells (Fig. 7E).

Fig. 7.

DOX-induced changes in CD8⁺ T-cell populations five weeks after treatment. (A) Quantification of CD3⁺CD8⁺ cells as a frequency of CD45⁺ cells. (B) Quantification of CD44 vs. CD62L expression as a frequency of CD3⁺CD8⁺ cells. (C) Stack plot of CD44 vs. CD62L expression as a frequency of CD3⁺CD8⁺ cells. (D) Quantification of Ki67⁺ cells as a frequency of CD8⁺ CD44^+/− CD62L^+/− cells. (E) Quantification of PD1⁺ cells as a frequency of CD8⁺ CD44^+/− CD62L^+/− cells. n = 8 mice/group. ns p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 DOX vs. the saline control group (analyzed by unpaired t-test or multiple unpaired t-tests if normal (A) or Mann-Whitney or multiple Mann-Whitney’s if non-parametric (B, D–E)).

To address how T-cell subsets could be altered at this time point, we examined thymus gene expression for markers of thymic development, cytokines and markers associated with cellular senescence. No significant changes were observed in markers of thymic development including Foxn1 (Fig. 8A) and Pax1 (Fig. 8B). Expression of the cytokines Il6 (Fig. 8C), Ifny (Fig. 8D), Il7 (Fig. 8E), Il17 (Fig. 8F), Il2 (Sup. Figure 7 A), Tnfα (Sup. Figure 7B), and Il4 (Sup. Figure 7C) also remained unaltered by DOX treatment. Markers of senescence including p16, p53, and p19 (Sup. Figure 8A-C) were not significantly altered, but there was a persistent elevation of p21 gene expression (Fig. 8G).

Fig. 8.

Delayed effects of DOX on the cytokines and senescence markers in thymus. Thymus was harvested from male mice five weeks following the administration of 4 mg/kg/week DOX or an equivalent volume of sterile saline for three weeks (n = 5–8 per group). Following the extraction of total RNA, the mRNA expression of (A) Fox n1 and (B) Pax1, (C) Il6, (D) Ifnγ, (E) Il7, (F) Il17 and (G) p21¹ at five weeks, was determined by qRT-PCR. Values were normalized to Actb expressed relative to saline-treated male mice. Values are shown as the means ± SEMs. ns p > 0.05, *p < 0.05, **p < 0.01, and ***p < 0.001 vs. the five week saline control group (analyzed by unpaired t-test if normal (A, C–E) or Mann-Whitney if non-parametric (B, F–G)).

Histopathology analysis

To examine possible pathological lesions five weeks after DOX administration, a total of 12 samples (6 control and 6 DOX-exposed animals) were evaluated, all of which demonstrated rare, minimal microscopic pathology. All 12 animals demonstrate rare individual cell necrosis, one animal (8%) demonstrated rare small foci of hemorrhage, and one (8%) demonstrated rare foci of cystic degeneration (consistent with physiologic thymic involution). No inflammation or edema was observed in any evaluated tissue (Fig. 9A–F). Thymocytes appeared to be similarly sized between the two groups, suggesting that the increase in thymus weight is hyperplastic, not hypertrophic (Fig. 9C and F). The observed findings of rare individual cell necrosis, hemorrhage, and cystic degeneration are interpreted as normal background lesions and are not considered histologically significant. There was no statistically significant difference in the cortex to medulla ratio between control and DOX-treated mice (Supplementary Fig. 9 and Supplementary Table 1).

Fig. 9.

Representative thymic histology, H&E-stained sections. (A–C) Saline treated; (D–F) DOX treated. Across all animals, no significant histologic findings or differences were identified between the two study groups. Rare and minimal foci of hemorrhage and rare individually necrotic cells were seen. Note the similarly sized thymocytes (C&F, black dashed outline). A and D = 40X; scale bar = 100 μm, B and E = 200X; scale bar = 50 μm, C and F = 400X; scale bar = 10 μm.

Discussion

Young adult cancer survivors often experience immune dysregulation due to cancer treatment. Therapies such as chemotherapy and radiation weaken the immune system, heightening susceptibility to infections and autoimmune diseases²⁵. While chemotherapy primarily targets cancer cells, it also disrupts rapidly dividing cells, including those in the immune system, leading to widespread immune suppression. This can lead to a temporary suppression or dysregulation of the immune system, leaving survivors vulnerable to infections during treatment and potentially for years after treatment has ended²⁶. T-cells are essential mediators of immune defense against cancer and infections. Cancer treatments, however, can impair T-cell production and function, increasing survivors’ vulnerability to infections and immune-related complications²⁷. Routine blood count tests, including measurements of immune cell counts, can provide valuable information about the immune health of cancer survivors. A decrease in immune cell count, known as leukopenia, can indicate a weakened immune system and increased susceptibility to infections. Monitoring T-cell levels within the immune cell population can offer further insights into the specific components of the immune system affected by cancer and its treatments²⁸.

In male juvenile mice, DOX administration results in pronounced thymic atrophy, characterized by decreased thymus weight, reduced cell count, and altered T-cell populations. This aligns with studies demonstrating the cytotoxic effects of DOX on rapidly dividing thymocytes, resulting in apoptosis and thymic involution²⁹. Thymic involution following DOX treatment arises from reactive oxygen species and DNA damage, culminating in apoptosis. Histopathological analyses confirm decreased cortical cell numbers and increased apoptosis³⁰. Additionally, DOX enhances CD4⁺ T-cell responses by upregulating CD40 ligand and 4-1BB expression, which are crucial for T-cell activation³¹. It also reduces myeloid-derived suppressor cells, thereby restoring T lymphocyte activity³². Additionally, the decrease in thymus weight may also be influenced by alterations in the thymic microenvironment, such as changes in cytokine production and thymic epithelial cell function, induced by DOX exposure³⁰. While the immediate effects of DOX, such as thymic atrophy and immune response alterations, have been reported³³, the long-term consequences on thymic function and immune competence remain underexplored. Understanding these long-term effects is crucial for developing strategies to mitigate the adverse impact of DOX on the immune system.

Interestingly, the delayed effects (five-week) of DOX on the thymus showed a contrasting pattern to the immediate effects (one-week), with an increase in thymus weight observed after a five-week period without drug administration. This phenomenon may reflect a compensatory response to thymic atrophy induced by DOX, characterized by increased thymic cell proliferation and expansion. However, despite the increase in thymus weight, there was a diminished occurrence of hematopoietic cells, suggesting potential alterations in thymic microenvironment and function induced by DOX exposure.

Upon examining the immediate impact of DOX on CD4⁺ T-cells, we observed no significant changes in the frequency of mature CD3⁺CD4⁺ T-cells, but there was an elevated frequency of CD4⁺ Tem. Interestingly, all CD3⁺ CD4⁺ T-cell subsets (Tn, Tem, Tcm, and CD44⁻CD62L⁻) exhibited a significant increase in PD1 expression, with no statistically significant differences in Ki67⁺ frequency. The elevation of PD1 expression on the CD4⁺ T-cells could indicate a function decline (exhaustion) or transient activation. However, the lack of significant Ki67 expression changes suggest that these cells may not be undergoing active proliferation, supporting the possibility of early exhaustion rather than activation. Where, immediately after DOX-treatment cells become activated or damaged and upregulate negative regulators of TCR stimulation, such as PD1. Conversely, the immediate effects of DOX on CD8⁺ T-cells revealed a significant expansion of mature CD3⁺CD8⁺ T-cells and an increase in the frequency of naïve CD8⁺ T-cells (Tn). Contrary to the CD4⁺ results, there was a significant increase in the frequency of Ki67⁺ on CD8⁺ Tn, Tem, and CD44⁻CD62L⁻ which could indicate that the increase in total CD8⁺ T-cells is due to proliferation of these three subsets. There were no differences in PD1 expression with DOX on CD8⁺ Tems and Tcms. Together these data indicate that immediately (one-week) after DOX-treatment CD4⁺ T-cells increase their expression of PD1, whereas CD8⁺ T-cells exhibit a significant increase in Ki67, indicating proliferation, that may explain their total expansion.

Following the cessation of chemotherapy, thymic hyperplasia commonly occurs, often independent of the degree of lymphocyte depletion^8,10. Numerous case reports have described thymic hyperplasia in breast cancer patients following adjuvant chemotherapy^8,10. Building on this clinical observation, we investigated the delayed effects of DOX on the thymus in adult mice exposed to DOX during their juvenile phase, with a focus on immune cell dynamics (five-week). Our findings revealed an increase in thymus weight in DOX-treated male mice, independent of lymphocyte depletion, suggesting a complex interplay between chemotherapy and thymic responses. This preclinical model provides novel insights, as thymic hyperplasia following chemotherapy has not been previously reported in animal studies, despite its prevalence in clinical cancer survivors^34,35. Interestingly, while there was no significant difference in the frequency of immune cells (CD45⁺), there was a reduction of absolute CD45⁺ cells/mg thymus in the delayed phase, suggesting a potential long-term effect on thymic cellularity. Interestingly, we observed a significant increase in the frequency of CD3⁻CD4⁺CD8⁺ double positive thymocytes which could indicate an issue with the maturation of T-cells within the thymus. In this phase, DOX treatment led to a significant increase in the expression of Ki67 on CD4⁺ T-cells. Mature CD3⁺CD4⁺ cells exhibited a significant decrease in frequency with DOX treatment. Despite this reduction in total mature CD4⁺ T-cells, we observed an increase in the frequency of Ki67⁺ Tn, Tcm, and CD44⁻CD62L⁻ cells. Similarly, CD8⁺ T-cell subsets were affected by DOX treatment. A significant decrease in mature CD3⁺CD8⁺ T-cells was observed, accompanied by a notable increase in Tn CD8⁺ T-cells. Although the frequencies of Tcm and Tem CD8⁺ T-cells remained unchanged, there were no significant alterations in Ki-67⁺ and PD1⁺ expression across CD8⁺ T-cell subset, indicating stable proliferative and activation states following DOX treatment. The treatment with DOX significantly increased the proliferation (indicated by higher Ki67⁺ expression) of Tn and Tcm CD4⁺ T-cells subsets. Additionally, PD1⁺ expression was significantly increased in Tn CD4⁺ T-cells, though this subset remained a minor fraction of the total population. These findings suggests that DOX treatment alters CD4⁺ T-cells homeostasis by increasing proliferation in Tn and Tcm subset, alongside an increase in the exhaustion expression in Tn CD4⁺ T-cells. While the functional implications of these changes remain unclear, they may reflect an altered activation state rather than direct immunosuppression.

The roles of Foxn1 and Pax1 in thymic development are essential, with Foxn1 driving the transcription of genes critical for thymus organogenesis and TEC differentiation²², while Pax1 establishes the microenvironment necessary for T-cell maturation²³. In DOX-treated male mice, increased expression of these transcription factors suggests adaptive mechanisms aimed at maintaining thymic integrity and function under stress. This indicates that despite the chemotherapeutic challenge, the thymus attempts to preserve its epithelial microenvironment and support T-cell development. The cytokine network within the thymus also adapts to DOX treatment, evidenced by decreased Il6 expression, which is vital for thymocyte proliferation and differentiation, particularly under inflammatory or stress conditions³⁶. This reduction may impair thymic function, contributing to degeneration and compromised T-cell development³⁷. Conversely, increased Ifnγ suggests a shift towards a Th1-biased immune response³⁸. Additionally, the upregulation of Il7 indicates a compensatory effort to support thymocyte survival and proliferation despite thymic degeneration³⁹. However, the decrease in Il17 expression could further contribute to thymic atrophy and impaired T-cell development, complicating the immune regulatory landscape under DOX treatment^39,40. These findings highlight the delicate balance of cytokine signaling and transcriptional regulation necessary for maintaining thymic function under chemotherapeutic stress.

DOX induces DNA double-strand breaks that activate the ATM/ATR/Chk1/Chk2 pathway, stabilizing and activating p53. Subsequently, p53 upregulates p21 transcription, driving cell cycle arrest and promoting senescence^41,42. In the thymus, DOX-induced DNA damage activates a p53-dependent response, upregulating p21 and other cell cycle arrest proteins^43,44. DOX treatment leads to increased p21 expression in the thymus, which is associated with thymic involution and impaired T-cell development. This can result in a reduced T-cell repertoire and compromised immune function, increasing the risk of infections and diseases in aging individuals⁴. The persistent high expression of p21 following DOX treatment suggests a sustained DNA damage response, which can disrupt thymic architecture and function, further impacting T-cell maturation^4,44. The findings presented in this study (summarized in Fig. 10) have important implications for understanding the impact of DOX on thymus function and immune responses, particularly in childhood cancer survivors. The observed alterations in thymic morphology and T-cell populations may contribute to immunosenescence, potentially affecting the long-term immune health of survivors. Although the increase in p21 expression may suggest immunosenescence, we have not confirmed the immunosenescent phenotype by more specific experiments such as β-galactosidase staining, which would have further validated the findings. Further research is warranted to elucidate the mechanisms underlying DOX-associated thymus changes and their effects on immune function, aiming to refine therapeutic strategies and improve outcomes for childhood cancer survivors. Future studies should also be performed to identify the delayed effects of DOX beyond the five weeks and determine how juvenile exposure to DOX may alter the natural aging trajectory of the thymus.

Fig. 10.

Summary of Immediate Effects and Delayed Effects of Doxorubicin on the Thymus. (A) Immediate (one-week) effects of doxorubicin (DOX) treatment compared to the control saline-treated mice include thymus atrophy, upregulation of thymic developmental genes (Foxn1 and Pax1), varying expression of inflammatory cytokines (upregulation of Ifnγ and Il7, and downregulation of Il6 and Il17), increased senescent marker (p21 expression, reduced frequencies of double-positive (CD3⁻CD4⁺CD8⁺) thymocytes, increased frequencies of PD1⁺ expression memory CD8⁺ cells, PD1⁺ expressing memory CD4⁺ cells, and Ki67⁺ expressing memory CD8⁺ cells, and no significant difference in the frequency of Ki67⁺ expressing memory CD4⁺ cells. (B) Delayed (five-week) effects of DOX treatment compared to the control saline-treated mice consist of thymus hypertrophy, normalization of thymic gene expression and inflammatory cytokines, the continued upregulation of p21, increase in double-positive thymocytes, normalization of PD1⁺ expression memory CD8⁺ cells, PD1⁺ expressing memory CD4⁺ cells, and Ki67⁺ expressing memory CD8⁺ cell frequencies, and an increased frequency of Ki67⁺ expressing memory CD4⁺ cells. Figure created with BioRender.

The current study has several limitations. First, we studied male mice only. Studying the effects of DOX on the thymus in female mice and discovering any potential sexual dimorphism will be planned for future studies. Although DOX is administered IV to cancer patients, it was administered IP in this study due to technical limitation for administering the drug IV to young mice. Another potential limitation of this study is that we were unable to dissect out subsets of CD4⁺ T-cells, such as Tregs and T-helper cells, with the use of markers such as FOXP3, CD25, and CD127. Since we did not determine the effects of DOX on stromal cell populations, the effects of DOX on thymic T-cells could have been caused by a direct effect on thymocytes or indirectly by perturbing the stromal microenvironment.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

B.G.; Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draftK.J.V.D; Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draftM.K.O.G; Data curation, Formal analysis, Methodology, Project AdministrationD.M.S; Data curation, Formal analysis, MethodologyM.R.D; Data curation, Formal analysis, MethodologyI.A.: Formal analysis, Writing – original draftD.S; Data curation, Formal analysis, MethodologyK.T.S; Conceptualization, Writing – reviewing and editing C.D.C; Conceptualization, Funding acquisition, Supervision, Writing – reviewing and editingB.N.Z; Conceptualization, Funding acquisition, Supervision, Writing – reviewing and editing.

Funding

This work was supported by the National Institutes of Health (NIH/NHLBI) R01HL151740 to B.N.Z. and the Children’s Cancer Research Fund to B.N.Z. and C.D.C. K.J.V.D. was supported by T32AG029796. I.A. was supported by the Summer College of Pharmacy Experiences (SCoPE) Program, University of Minnesota College of Pharmacy.

Data availability

Data is provided within the manuscript or supplementary files.

Declarations

Competing interests

The authors declare no competing interests.