Senescent cells secrete a specific mixture of factors, including MMPs, cytokines, chemokines, growth factors, and angiogenic factors, which enable them to affect nearby cells. The composition of the molecules varies according to both the senescence inducer and the cell type. This secretome is known as the SASP [3,34]. The SASP promotes the spread of senescence and inflammation to surrounding cells in a paracrine way [35] and also triggers immune response to expel senescent cells [36]. Interestingly, senescent cells themselves are immune to the apoptotic and damaging influence of the SASP and they generate even more of the secretome in response [6].

The composition of SASP differs depending on the situation, which may be connected with its antagonistic effects. Early research on the SASP revealed the diversity of molecules secreted by senescent cells, among them chemokines and cytokines: C-X-C motif chemokine ligand 1 (CXCL1), C-C motif chemokine ligand 2 (CCL2), IL-6, interleukin 8 (IL-8), prosurvival factors like glial cell line-derived neurotrophic factor (GDNF), insulin-like growth factor-binding proteins (IGFBPs) and growth modulators such as amphiregulin (AREG) [9]. Nowadays the composition of SASP is better known. Research shows that SASP also contains proteases, ceramides, extracellular matrix (ECM) molecules, signaling factors like bradykinins, hemostatic factors and damage-associated molecular patterns (DAMP) [5,37].

Cell type and senescence inducer seem to be major factors influencing the SASP composition [38]. Mitochondrial dysfunction-associated senescence (MiDAS) results in a distinct senescence response where the interleukin 1 (IL-1)-dependent proinflammatory impact does not occur [39]. The latest research shows that some of the SASP components are common in different cells and different types of senescence and may be some kind of core [37]. The composition of SASP depends on the passage of time as well. During OIS, transforming growth factor β (TGF-β) rich early immunosuppressive secretome turns into the SASP due to the fluctuation of neurogenic locus notch homolog protein 1 (NOTCH1) signaling [40]. What is more, expression of type I interferons (IFN) occurs in late OIS and is partly initiated by the reduction of LINE-1 retrotransposable elements [41]. This may partly explain the variability and sometimes contradiction in SASP effects. As can be seen, senescent cells’ secretome is complex and depends on a plethora of different factors. In order to better understand the SASP, further research is needed.

Due to the differences in the quantity and nature of the molecules secreted by senescent cells, SASP may have both positive and negative effects on the organism. Beneficial and detrimental effects of some of the SASP components are presented in Table 1. Considering positives, senescent fibroblasts’ SASP, through platelet-derived growth factor AA (PDGF-AA) and cellular communication network factor 1 (CCN1), boost the skin regeneration process [42,43]. Moreover, the SASP reinforces tissue stemness and plasticity, prevents excessive tissue fibrosis, and restores homeostasis in liver fibrosis [44,45]. It can also promote the reprogramming of nearby cells if the tissue is damaged [46]. In addition, the SASP is responsible for the recruitment of immune cells, such as natural killer (NK) cells, T helper type 1 (Th1) cells and macrophages, which play a key role in initial preneoplastic cells elimination and prevent further carcinoma expansion [47]. Moreover, it can skew the polarity of macrophages to tumor-inhibiting state (M1), creating an antitumor environment [48]. Some elements of SASP, such as IL-6, IL-8, IGFBP7, and plasminogen activator inhibitor (PAI-1), support the senescent cell growth arrest [4].

On the other hand, SASP-induced inflammation may negatively affect cells and even provoke cancer proliferation [4]. IL-1α, which is present in different types of senescence, influences cells in an autocrine way and generates inflammation. It activates NFκB, and enhances IL-6 and IL-8 secretion [80]. Additionally, IL-6 can promote cancer progression in a paracrine way by affecting the IL-6 receptor. There is activation of signal transducer and activator of transcription (STAT3), which leads to transcription of oncogenes and growth regulators, such as c-FOS, cyclin D1, c-MYC, mTOR and mammalian target of rapamycin complex 1 (mTORC1) [81]. It has been proven that as a result of the elimination of senescent cells, the level of proinflammatory cytokines, such as IL-1α, IL-6, and tumor necrosis factor α (TNFα), decreases [82,83], which prevents cancer relapse [84].

In hepatocellular carcinoma, CCL2 was observed to recruit immunosuppressive myeloid cells which restricted NK cells activity leading to further tumor progression [85]. CXCL1 is strongly expressed in both senescent cells and ovarian cancer samples [86]. Secretion of this chemokine can affect stromal cells in a paracrine way and promote tumor invasion and progression [87]. Many of the SASP factors (IL-1α, IL-6, IL-8, C-X-C motif chemokine ligand 12 (CXCL12), CCL2) are also common after anticancer therapy [88] and contribute to physical function decrease, fatigue, loss of appetite and cardiovascular morbidity [89,90,91]. Taking into consideration that senescence is an inherent element of cancer treatment (deliberately or not), it is believed that senescent cells may enhance some of the short-term and long-term side effects of anticancer therapy [87].

As mentioned before, elements of the SASP include growth factors, such as proangiogenic vascular endothelial growth factor (VEGF), which lead to cancer vascularization [92]. Furthermore, MMPs released by senescent cells promote cancer growth [93] and stimulate VEGF-dependent vascularization [94]. Moreover, the SASP is able to induce epithelial-to-mesenchymal transition (EMT), which promotes cancer vascularization and, consequently, tumor growth [34]. Research shows that senescent fibroblasts secretome may support tumor vascularization in xenograft transplants and enhance metastasis and proliferation of premalignant epithelial cells [92,95]. In co-culture composed of senescent lung fibroblasts (line WI38) and preneoplastic epithelial cells, the acceleration of proliferation was observed compared to co-culture with normative fibroblasts [9]. What is more, the enhanced-growth effect was not dependent on senescence inducer and did not occur when co-culturing senescent fibroblasts with normative epithelial cells [9]. Additionally, injection of senescent fibroblasts to breast cancer cells resulted in tumor growth acceleration [96]. What is more, the osteopontin (OPN) secretion increase was detected in senescent stromal cells of murine skin papilloma. Co-injection of OPN-deficient senescent cells was performed and the limitation of cancer growth was observed [97].

Milanovic et al. [98] conducted research to check the influence of chemotherapy-induced senescence on stem-cell-related properties in cancer. In senescent cells, enhancement of an adult tissue stem-cell signature, distinct stem-cell markers and Wnt signaling—a crucial factor in stem-cell restoration—have been observed. Furthermore, treating cells which escaped senescence and never senescent cells with a precise dose of chemotherapy resulted in accelerated and Wnt-dependent growth in previously senescent cells [98]. Wnt signaling is activated in therapy-induced senescence (TIS) and may contribute to higher cancer-initiating potential [98,99,100]. Additionally, in p53-regulatable models of acute leukemia, enhancement of senescence resulted in reprogramming of non-stem bulk leukemia cells to self-renewing stem cells that initiate leukemia [98]. On the other hand, recent studies on skin cancer cells indicated that senescent cells, in fact, promote tumor growth by, among others, P38MAPK and MAPK/extracellular signal-regulated kinase (ERK) up-regulation but do not induce the cancer development [101].

The SASP regulation occurs at several levels, including secretion, mRNA stability, transcription and translation. Paracrine and autocrine feedbacks, which intensify secretion, are also important factors [4]. In control of the SASP regulation are, among others, cyclic GMP-AMP synthase (cGAS)/stimulator of interferon genes (STING) [86,96], DDR [9,21], or P38 mitogen-activated protein kinases (MAPK) [102]. All of them lead to NF-κB and CCAAT/enhancer-binding protein-β (C/EBPβ) activation, which directly controls the transcription of IL-6 and IL-8. These molecules in an autocrine way—through a feed-forward loop—increase the activity of NF-κB and C/EBPβ and intensify SASP [4]. Suppression of C/EBPβ transcriptional activity leads to a decrease in the SASP inflammatory activity [103]. Elimination of NF-κB and C/EBPβ results in IL-8 reduction, which affects OIS growth arrest [104,105]. As can be seen, there are visible connections between SASP, NF-κB activation, DDR and transcription factor GATA4 stabilization.

The SASP can be also affected in an epigenetic way. Repression of transcriptional activator and oncoprotein MLL1 causes reduction of proliferative cell cycle regulators and inhibits SASP expression [106]. What is more, recruitment of bromodomain-containing protein 4 (BRD4) to superenhancers activated by senescence seems to be essential for the SASP induction [107]. Chromatic alterations, such as higher expression of macroH2A1 histone [108], reduced H3K9 histone dimethylation [109] or accumulation of H2AJ histone [110] also contribute to SASP regulation.

SASP induction may be provoked by inflammasomes, in particular inflammasome NLRP3, triggered by DAMP, which activates and processes IL-1β [35]. In OIS, inflammasome activation is caused by priming the toll-like receptor 2 (TLR2) by serum amyloids A1 and A2 [111]. A key role in the SASP induction is played by cytosolic DNA, derived from retrotransposable molecules (for instance LINE-1), cytosolic chromatin fragments (CCF) and mitochondrial DNA [3]. DDR and dysfunctional mitochondria promote formation of CCF and trigger the SASP [112]. The cGAS detects cytosolic DNA and synthesize cyclic GMP-AMP (cGAMP), which leads to STING activation [86,96,113]. STING activation results in TANK-binding kinase 1 (TBK1) secretion and IRF2 and NFκB activation, which stimulate IFN-I response [5]. Activation of P38MAPK stimulates activation of NF-κB as well [102].

MiDAS SASP is heavily influenced by metabolic abnormalities in NAD⁺ (nicotinamide adenine dinucleotide) to NADH ratios, which contribute to activation of P53 via influencing 5′AMP-activated protein kinase (AMPK) and enhance SASP [39]. Additionally, increased NAD⁺ salvage caused by a higher level of nicotinamide phosphoribosyltransferase (NAMPT) accelerates mitochondrial respiration and glycolysis in OIS and enhances SASP [114]. In the long run, all of these pathways lead to the induction of SASP.

A major factor in SASP regulation is the mTOR pathway. It has been proven that phosphorylation of 4EBP—translation repressor protein—mediated by mTOR, influences the translation of mitogen-activated protein kinase-activated protein kinase 2 (MAPKAPK2) and IL-1α [80,115]. The stress-activated serine/threonine-protein kinase MAPKAPK2 activity leads to the inhibition of the mRNA-binding protein ZFP36L1, which directs the SASP components to degrade mRNA. The mTOR pathway can regulate the senescent secretome by influencing the stability of SASP mRNAs [115]. Furthermore, in OIS the TOR-autophagy spatial coupling compartment (TASCC) boosts autophagy and protein synthesis which result in the accumulation of SASP molecules [116]. What is more, the research indicates also that polypyrimidine tract-binding protein 1 (PTB1) can control senescent secretome via influencing an alternative splicing of genes taking part in intracellular trafficking (like EXOC7) [117].

In order to induce cytokine secretion, constant DDR is usually needed [21]. Taking into consideration that deficiency of DDR regulators such as NBS1, CHK2 and ATM restrains cytokine expression caused by genotoxic stress, it can be presumed that these factors have a major impact on the SASP induction [28]. Research shows that ATM contributes to removal of macroH2A1.1 histone variant from SASP genes and so participates in SASP genes expression [108]. Activation of P16 protein may enhance senescence and inhibit cells proliferation, but is not sufficient to initiate SASP. On the other hand, P38MAPK activation is not only sufficient but also necessary to induce senescence and SASP even if DNA damage does not occur [102]. Additionally, P38 enhances the activity of NF-κB and, as a result, induces SASP transcripts expression [28]. Furthermore, P53 inhibition intensify SASP and create a proinflammatory environment, which provoke malignant transformation [21].

As can be observed, the factors related to DDR also influence the SASP, which confirms the thesis, that SASP is a complex phenomenon, dependent on many factors. It seems possible to regulate SASP by manipulating the DDR. It is important to search for more links between DDR and SASP, which may expand SASP regulation possibilities.