Senescent cells are irreversibly arrested in the cell cycle and are usually large in size (Biran et al., 2017). Large HSCs are always dysfunctional as they fail to build a blood system in vivo and display proliferative defects. This raises the question whether large HSCs are senescent and whether enlargement facilitates the senescent state (Lanz et al., 2022; Xie et al., 2022)? While little is known about this connection in HSCs, it has been shown that not all large HSCs are also senescent based on SA-β-gal production (Lengefeld et al., 2021). This indicates that HSC enlargement does not strictly induce cellular senescence. Nevertheless, based on the observation that senescent HSCs are always large, we speculate that large HSCs may be more prone to acquire the senescent program when exposed to additional damage. This is supported by the observation that larger human cells in vitro are as well more prone to senescence potentially via expansion of lysosomes and the endoplasmic reticulum (Cheng et al., 2021; Lanz et al., 2022). Overall, it will be important to further investigate how HSC enlargement increases the likelihood of entering the senescent state.

Telomere shortening and the accumulation of DNA damage during aging contribute to cellular dysfunction (Lopez-Otin et al., 2013). Are telomere shortening and DNA damage also contributing to the dysfunction of large HSCs? As outlined above, DNA damage contributes to enlargement as it transiently halts HSC division to repair the DNA damage (Sperka et al., 2012). During this arrest, cell growth continues leading to HSC enlargement (Fingar et al., 2002; Neurohr et al., 2019). This enlargement drives dysfunction of HSCs. When growth is inhibited with rapamycin during this DNA damage, HSCs maintain their functionality (Lengefeld et al., 2021). Thus, if telomere shortening during DNA damage induced enlargement is crucial for their dysfunction, then we expect rapamycin to prevent telomere shortening and the associated accumulation of DNA damage. This is not the case. Rapamycin does not prevent DNA damage in HSCs and, so far, no protective effect of rapamycin on telomere length has been reported (Kawauchi et al., 2005; Qi et al., 2008; Ungar et al., 2011; Gopalakrishnan et al., 2018; Ferrara-Romeo et al., 2020). Additionally, other studies have indicated that it is still unclear whether telomere length plays a crucial role in HSC exhaustion (Allsopp et al., 2003): Telomeres become shorter during divisions of HSCs (Allsopp et al., 2001). However, at least in mice, HSCs become exhausted even when telomere shortening is prevented by telomerase overexpression (Allsopp et al., 2003) arguing that there are telomere-independent barriers to HSC function. Furthermore, we would like to speculate about the possible protective feature of enlargement. DNA-damage leads to enlargement and potentially permanent gene mutations and chromosomal aberrations. However, this enlargement in turn reduces the proliferation potential of these cells and thereby may prevent the propagation of potentially harmful features in the tissue that could facilitate, for example, a malignant transformation. In summary, DNA damage contributes to HSC dysfunction by enlarging their size, while telomere attrition seems to play a less dominant role than enlargement in causing HSC dysfunction.

Mitochondrial volume typically scales with cell size in dividing cells supporting faithful cellular functions (Posakony et al., 1977; Rafelski et al., 2012; Miettinen and Bjorklund, 2016; Cheng et al., 2021; Lanz et al., 2021; Seel et al., 2022). For example, HSCs are able to scale mitochondria upon their activation (Ho et al., 2017). Changes in mitochondrial morphology and volume beyond this scaling are associated with cellular dysfunction (Chistiakov et al., 2014). Mitochondrial number and volume have often been quantified in old cells, but the results have been inconsistent. In vitro, some cell types display decreased mitochondrial volume (Stoll et al., 2011; Yako et al., 2021; Barilani et al., 2022), but most human cell lines increase their mitochondrial volume as they become larger and senescent (Lee et al., 2002; Passos et al., 2007; Koziel et al., 2011; Xie et al., 2011; Correia-Melo et al., 2016; Stab et al., 2016). It is noteworthy that increased mitochondrial volume does not necessarily translate into increased mitochondrial functions in vitro (Passos et al., 2007; Passos et al., 2010; Miettinen and Bjorklund, 2016). Furthermore, in vivo studies often reported decreased mitochondrial volume during aging (Markowska et al., 1994; Navarro and Boveris, 2004; Mathieu-Costello et al., 2005; Martinelli et al., 2006a; Addabbo et al., 2009; Leduc-Gaudet et al., 2015; Del Campo et al., 2018; Brown et al., 2021; Lengefeld et al., 2021). For example, in large HSCs mitochondrial volume decreases per unit volume (Lengefeld et al., 2021). Currently, the underlying reasons for these differences in vitro and in vivo are unclear. One possibility is that they are facilitated by different oxygen levels in vivo (hypoxic) and in vitro (hyperoxic). Nevertheless, these changes in mitochondrial volume during aging are associated with a decline in mitochondrial function, potentially driven by differences in fusion, fission and mitophagy (Mai et al., 2010; Korolchuk et al., 2017; Yako et al., 2021). Whether changes in mitochondrial morphology, volume and number during aging are the cause or consequence of cellular enlargement remains to be determined. In summary, while mitochondria volume and function scale with cellular size during faithful divisions, these parameters become uncoupled when cells enlarge during aging.

The scaling of protein levels with cell size is important for cellular function (Fraser and Nurse, 1978; Padovan-Merhar et al., 2015; Schmoller and Skotheim, 2015; Sun et al., 2020; Cheng et al., 2021; Lanz et al., 2021). Upon substantial enlargement of primary human cells in vitro, protein synthesis rates and ribosome levels decrease (Delarue et al., 2018; Neurohr et al., 2019). However, considering that disruptions of proteostasis contribute to HSC aging (Signer et al., 2014; Kruta et al., 2021), it was surprising that measurements of nucleolar size, cellular density, mTOR activity and protein synthesis in large HSCs indicated no loss of protein synthesis capacity (Lengefeld et al., 2021). While protein synthesis was unaffected by large HSC size, whether protein turn-over, folding state, aggregation and protein-protein interactions were impaired is unknown. Indeed, during aging, misfolded proteins contribute to the loss of cellular function (Lopez-Otin et al., 2013) and it will be important to analyze whether large stem cell size affects the accumulation of these misfolded proteins.

How enlargement of HSCs is related to the remaining hallmarks of aging like chromatin remodeling, genomic instability, nutrient sensing and intercellular communication is at this point unclear. It will be important to understand which aging hallmarks act in the same pathways as cellular enlargement to build a holistic picture of aging. Furthermore, initial experimental evidence suggests that enlargement can be reversed, which improved stem cell function (Lengefeld et al., 2021). This may open the possibility that other aging factors that are connected to cell size can be improved when shrinking stem cells.

Before we dive into the literature evaluating cellular size and age-related dysfunctions, we would like to point out examples of enlargement that are not associated with dysfunction.

Cell cycle - Cells usually double in size during the cell cycle to ensure that their daughter cells are the same size as the original mother cell after symmetric cell division (Lloyd, 2013). In muscle satellite cells, this enlargement can happen already prior to cell cycle entry and is called the G_alert state. This enables an accelerated entry into a proliferating state after repeated muscle injury (Rodgers et al., 2014).

Differentiation - Another example of enlargement as part of faithful cellular processes is differentiation, in particularly during development. For example, 1) keratinocyte enlargement during differentiation is proposed to enable the formation of an insoluble protein envelope at the plasma membrane (Watt and Green, 1981), 2) terminal differentiation drives enlargement of chondrocytes (hypertrophy) as part of skeletal tissue elongation (Cooper et al., 2013), 3) cardiomyocytes increase in size during heart development (Knaapen et al., 1996; Li et al., 1996), and 4) T cell activation requires profound changes in cell size (Kaesler et al., 2012). Furthermore, during cell manipulations that trigger differentiation, cells not only acquire a new cell identity but also a new cell size, which is optimal for this cell type. For example, overexpression of RUNX3, results in an increase in cell size and a block in faithful differentiation of erythrocytes (Menezes et al., 2022).

Ploidy - Changes in ploidy are generally associated with enlargement, but do not necessarily lead to dysfunction (Gillooly et al., 2015; Mu et al., 2020). For example, during adolescence, mouse acinar cells often become tetraploid and increase in size without losing their function (Xanthopoulos et al., 2008). On the other hand, polyploidization of hepatocytes occurs successively during postnatal development and again during advanced aging. The latter being associated with senescence, progressive loss of cell pluripotency and decreased replication capacity (Celton-Morizur and Desdouets, 2010; Wang et al., 2017).

Apoptosis - Cell volume decrease is characteristic of apoptotic cell death (Bortner and Cidlowski, 2002; Núñez et al., 2010). Cancer cells are able to prevent apoptotic shrinking by modulating the activity of ion channels (Lehen’kyi et al., 2011).

In conclusion, changes in cell volume are part of cellular processes and not necessarily associated with dysfunction. In the next part of this section, we will highlight findings, in which cellular enlargement is associated with aging, dysfunction and disease.

Cell size is influenced by matrix stiffness, osmotic pressure and mechanical forces (Wang et al., 2020a). For example, cells substantially shrink when cultivated on substrates of increased stiffness (Guo et al., 2017; Xie et al., 2018; Yang et al., 2020). Interestingly, modulation of the external stiffness, and thereby cellular size, affects cellular function. For example, mesenchymal stem cells (MSCs) enlarge during aging in vitro and in vivo, which is accompanied by a senescent phenotype including high levels of p21, β-galactosidase, and SASP (Block et al., 2017; Yin et al., 2020; Liu et al., 2021). When MSCs are cultivated at an optimal substrate stiffness, they maintain their small size and proliferation capacity and do not become senescent (Kureel et al., 2019). Furthermore, cultivating large aged endothelial cells on substrate with increased stiffness rejuvenates these cells and reduces their size (Mammoto et al., 2019). The underlying mechanism is unknown. Together, this suggests that the environmental stiffness can modulate cellular size and thereby cellular functionality in vitro.

It is important to note that cells in 2D monocultures are exposed to a different microenvironment than in the in vivo 3D tissue context. In vivo, cells face the extracellular matrix (ECM), neighboring cells, biochemical and physical cues (Barthes et al., 2014), which all affect the environmental stiffness. As tissues have different solidity states, it is expected that stiffness varies between tissues and organs (Butcher et al., 2009). During aging in vivo, increased stiffness is associated with malignancy (Lieber et al., 2004; Butcher et al., 2009; Stearns-Reider et al., 2017; Segel et al., 2019; Ghosh et al., 2020), while stiffness decreases in some organs like the skin (Tissot et al., 2018; Runel et al., 2020; Lynch et al., 2022) and muscle tissues (Akagi et al., 2015; Yoshida et al., 2017; Alfuraih et al., 2019). We speculate that reduced stiffness contributes to cellular enlargement during aging. It will be of critical interest to analyze how microenvironmental forces affect cellular size and thereby cellular function during aging in vivo. In summary, the microenvironment modulates cell size and therefore likely cell function as well.

Red blood cell distribution width (RDW) represents variability of the size distribution of red blood cells and has been used as a marker for multiple aging-related pathologies (Salvagno et al., 2015). RDW involves a change in mean corporal volume (MCV). An increase in MCV, meaning increase in red blood cell volume, correlates with aging (Hoffmann et al., 2015; Lippi et al., 2018) and numerous diseases like anemia, autoimmune diseases, cancer types, and cardio- and cerebrovascular diseases (Bessman et al., 1983; Li et al., 2017; Lippi et al., 2018; Katsaros et al., 2020). Interestingly, not only enlargement, but a deviation from optimal red blood cell size in general is associated with diseases, for example, various leukemia forms, Alzheimer’s disease, Parkinson’s disease, autoimmune disease, and macular degeneration (Pilling et al., 2017; Lippi et al., 2018; Katsaros et al., 2020; Kim et al., 2021).

It is unclear why red blood cells of many diseased individuals are large. The fact that red blood cells lack organelles, such as a nucleus and mitochondria, implies that the underlying mechanism driving dysfunction is different from the one in HSCs, in which these organelles are necessary for enlargement and dysfunction (Lengefeld et al., 2021). Furthermore, the correlation of RDW with aging-associated diseases has not been investigated for a causal relationship. Overall, RDW is a robust indicator of numerous diseases, but the origin of this relationship is insufficiently understood.

Large size of adipocytes is associated with several pathophysiological conditions like reduced response to hormones, increased inflammation, impaired body metabolic regulation and insulin resistance, which is associated with type 2 diabetes (Bjorntorp et al., 1972; Bernstein et al., 1975; Weyer et al., 2000; Lundgren et al., 2007; Lonn et al., 2010; Yang et al., 2012; McLaughlin et al., 2014; Laforest et al., 2015; Stenkula and Erlanson-Albertsson, 2018; Liu et al., 2020). However, it is also suggested that it is the enlargement and not the absolute size that predicts insulin resistance (Johannsen et al., 2014). Interestingly, a recent model proposes that enlargement of adipocytes itself affects the interaction of the cell with the ECM. This activates integrin/Erk signaling and modulates gene expression (Farnier et al., 2002; Farnier et al., 2003). It will be exciting to see further studies evaluating the causality of cellular enlargement on adipocyte function.

Mean adipocyte size increases between middle and old age and then decreases with advanced age (Miller et al., 2017). Interestingly, this correlates with the risk for the onset of type 2 diabetes, which peaks at 45–65 years and lowers afterwards (Plummer et al., 2016; Sattar et al., 2019; Jacobs et al., 2020; Hu et al., 2021). Nevertheless, while smaller adipocyte size during advanced age correlates with a lower onset risk of type 2 diabetes, it coincides also with another unfavorable outcome: A result of decreased adipocyte size is the decline in fat depot size, which leads to the accumulation of fat outside adipose tissue like in the bone marrow, muscle and liver. This is associated with lipotoxic stress in the storage organs and lower organ function with age (Rosen and Bouxsein, 2006; Slawik and Vidal-Puig, 2006). Overall, enlargement of adipocytes predicts obesity-associated pathologies like type 2 diabetes and during aging, the shrinkage of adipocytes is likely associated with other pathologies.

At old age, brain volume declines (Haug and Eggers, 1991; Resnick et al., 2003). Studies indicate that this is caused by a shrinking of neuronal cell size (Anderson et al., 1983; Terry et al., 1987; Diaz et al., 1999; Kuwahara et al., 2004; Martinelli et al., 2006b; Fernandez et al., 2007; Stark et al., 2007; Freeman et al., 2008; Verkhratsky et al., 2014; Popov et al., 2021). Nevertheless, in some areas of the brain neurons maintain size or enlarge with age (de Lacalle et al., 1991; Merrill et al., 2000; Martinelli et al., 2006b; Verkhratsky et al., 2014; Thulborn et al., 2016; Delgado-Gonzalez et al., 2020; Yako et al., 2021). Interestingly, during neurodegenerative diseases such as Alzheimer’s, neurons enlarge. Overexpression of the amyloid precursor protein causes enlargement of cortical neuron size (Oh et al., 2009) and larger neuronal cell lines are more Aβ-sensitive (Simakova and Arispe, 2007). This causes cellular toxicity and resembles cellular damage in Alzheimer’s diseased brains. Furthermore, an increase in cell size or neuronal hypertrophy has been reported in elderly that display no cognitive symptoms yet but markers of Alzheimer’s disease (Iacono et al., 2008). At later stages of Alzheimer’s disease progression, oligodendrocyte progenitor cells increase in size (Vanzulli et al., 2020) and neocortical neurons are enlarged in Alzheimer patients (Bundgaard et al., 2001; Pakkenberg et al., 2003). In summary, while most neuronal cell types shrink during aging, areas that contain neurons that enlarge might be especially prone to processes that result in symptoms of Alzheimer’s disease.

As the proliferative capacity of heart cells becomes severely restricted after birth, the postnatal increase of heart volume is achieved by cardiomyocyte hypertrophy (Li et al., 1996; Porrello et al., 2011). Heart hypertrophy coincides at the cellular level with nuclear polyploidization, multinucleation, and cellular growth (Derks and Bergmann, 2020). While cardiomyocyte hypertrophy is associated with development and adjustment to exercise (Ellison et al., 2012), it also predicts numerous cardiovascular diseases like ischemic disease, hypertension, heart failure, and valvular disease (Wang et al., 2013; Akhondzadeh et al., 2020; Peter et al., 2016). This raises an important question: What is the difference between physiological and pathological cellular hypertrophy?

Cellular growth driven by the IGF1/PI3K/Akt1 pathways facilitates heart enlargement after exercise, while signaling via G_q protein-coupled receptors, reduced blood supply, oxidative stress, inflammatory processes and Ca²⁺ abnormalities drive cellular growth and enlargement of the heart associated with pathology (Weeks and McMullen, 2011; Oldfield et al., 2020). Thus, in the case of the heart, enlargement is associated with opposite outcomes and thus, it seems unlikely that they are driven by enlargement itself.

Nevertheless, there are some interesting differences to be noted: For example, enlargement after exercise changes the cellular shape in a different way than during disease (Kuo et al., 2012; Peter et al., 2016). Furthermore, it could be that pathological hypertrophy enlarges cardiomyocytes beyond a size-threshold due to extensive growth. This is supported by the observation that cardiomyocytes after exercise are smaller than cardiomyocytes under diabetic stress (Stolen et al., 2009). While many cardiomyocytes become polyploid during pathological enlargement, whether this enlargement is always accompanied with an increase in DNA copy number is unclear. Interestingly, in yeast, enlargement without increasing DNA copy number results in cytoplasmic dilution and dysfunction (Neurohr et al., 2019). Thus, it will be interesting to investigate whether mechanisms like cytoplasmic dilution contribute to the dysfunction of cardiomyocytes upon pathological enlargement and whether polyploidization is a compensating mechanism to prevent this functional loss. Indeed, polyploidy has been hypothesized as a mechanism to increase metabolic and growth capacity of cells (Frawley and Orr-Weaver, 2015; Mu et al., 2020). Furthermore, if it is true that pathological enlargement facilitates changes in ploidy, this could also include changes like large genomic aberrations and aneuploidy. These genomic defects may also facilitate unwanted downstream effects. Indeed, aneuploidy is a distinct feature of cancer cells (Ben-David and Amon, 2020). Overall, being of the right size is important for cardiomyocytes (Ren and Brown-Borg, 2002) and hypertrophy is a crucial characteristic of adjusting to exercise and pathology.

Heterogeneity of cell size and shape are often observed in tumors and referred to as pleomorphism (Manocha et al., 2003; Terzakis et al., 2005). This raises the question of whether a failure of size regulation is facilitating cancer progression. Some research suggests that size heterogeneity of cancer cells provides the highest malignant potential (Kawada et al., 1994; Manukyan et al., 2021). Furthermore, the population of small cells within tumors raise attention. Small prostate cancer cells build more malignant tumors than large ones (Li et al., 2015). Mutated STAT3 is a crucial cancer facilitator and leukemia patients with STAT3 mutations carry smaller blood cells than patients without these mutations (Tanahashi et al., 2016) suggesting that the smaller cells might facilitate cancer progression. It is also speculated that cancer stem cells are small, just like most stem cells (Berardi et al., 1995; Colter et al., 2001; Young et al., 2004; Virant-Klun and Stimpfel, 2016). The small size could result from the fast proliferation of cancer cells, leaving little time for growth during these short cell cycle durations (Lloyd, 2013; Wright and Schneider, 2014). On the other hand, a small size might support the cancerous potential of the cells. One possibility is that size affects the migration behavior of cancer cells (Leal-Egana et al., 2017; Wang et al., 2020b). Overall, the literature provides interesting clues that a deviation from the optimal cell size is associated with cancer, but whether and how size modulates malignancy is unclear.

Solid tumors are usually an assembly of distinct cell types including infiltrating immune cells, each owning their own cell size (Ginzberg et al., 2015). Thus, the size heterogeneity found in cancer tissues could reflect the tumor’s composition of various cell types (Shembrey et al., 2019) and not a failure in size regulation of the cells that originated the cancer. However, cell size could still serve as a diagnostic marker for certain types of cancers. For example, for acute lymphoblastic leukemia, the significance of cell size as a prognostic indicator is heavily investigated (Mathe et al., 1971; Pantazopoulos and Sinks, 1974; Murphy et al., 1975; Oster et al., 1976; Dubner et al., 1978; Ghani and Krause, 1986; Olcay et al., 1999). Further research is needed to determine whether cell size is a reliable predictor of malignancy. Investigating the potential link between cellular size and cancer potential will be crucial for the development of new therapies for cancer treatments.

As discussed, an arrest in cell cycle causes cellular enlargement (Fingar et al., 2002; Neurohr et al., 2019; Lengefeld et al., 2021). Cell cycle inhibitors like palbociclib, dinaciclib, seliciclib, ribociclib, and abemaciclib therefore enlarge cell size. These kinds of drugs are already successfully used in clinical trials or are even FDA-approved for cancer treatment. As enlargement can cause dysfunction, this could contribute to the inhibitory success of these cell cycle inhibitors in cancer treatments. Recent studies provide important insights into this topic. Uncoupling cell growth and proliferation in cancer cells by arresting them in G₁ phase leads to toxic overgrowth and senescence (Wilson et al., 2021; Crozier et al., 2022; Foy et al., 2022; Manohar et al., 2022). Furthermore, keeping patient-derived glioma stem cells small (rapamycin treatment) during cell cycle arrest (palbociclib) prevents their entry into a senescence state (Morris-Hanon et al., 2019). Thus, growth and enlargement, in addition to cell cycle arrest, effectively inhibit cancer cells from proliferating and cancer treatments using cell cycle inhibitors may therefore not be combined with rapamycin.

Cell cycle inhibitors like palbociclib have also been investigated as treatment for other pathologies: mast cell-mediated allergy (Hou Y. B. et al., 2019), diabetic cardiomyopathy (Wang et al., 2019), pulmonary arterial hypertension (Weiss et al., 2019), replication of herpes virus (Badia et al., 2016), systemic sclerosis (Yamamoto et al., 2022) and spinal muscular atrophy (Hor et al., 2018). It will be important to evaluate how arrest-induced growth and enlargement affect the outcome of these diseases. Overall, this highlights the importance of understanding how growth and enlargement contribute to the multilayered effects of treatment drugs.