1. Introduction
Salivary glands (SG) hypofunction is one of the most common side effects of irradiation (IR) in head and neck cancer (HNC) patients. Up to 80% of these patients who received radiotherapy (RT) consequently suffer from xerostomia (which defines subjective dry mouth symptoms) and hyposalivation (an objective decrease in salivary flow rate) because of the irreversible damage to acinar cells in SGs caused by IR [
1,
2,
3]. The functional cells in SGs are highly differentiated and relatively quiescent, but they are surprisingly radiosensitive, especially the acinar cells [
4]. Thus, a rapid decrease in salivary secretion occurs in the first-week post-RT, and a further significant decrease is observed during the three months following RT [
2]. In addition to volume, other factors, such as salivary electrolyte levels, buffering properties, and saliva’s antibacterial systems, are also changed post-radiotherapy [
1,
2]. The pH of the saliva is reduced from 7.0 to 5.0 [
2,
5,
6], and the concentration of inorganic salt (such as sodium, chloride, and calcium) and organics (such as immunoproteins and lysozyme) are increased. RT-induced saliva reduction is commonly associated with the increased activity of tooth caries, oral fungal infections, and mucosal inflammation [
7]. Such related symptoms can consequently affect the individual’s nutritional intake, speech, and sleep, thus reducing their quality of life.
The process of IR-induced SG injury has been divided into four stages, based on the time phases in the rat model [
8]. In phase one (0–10 days), water secretion is affected with no apparent cell death, while in phase two (10–60 days) and phase three (60–120 days), the amylase secretion significantly decreases, and acinar cells disappear. Then, a lower salivary flow rate is observed in phase four (120–240 days), with loss of functional cells and their supportive environment (of ductal and stem cells). Recently, another study further summarized the classification of IR-induced SG damage into two stages: the acute (0–3 days post-IR) and chronic phases (>30 days post-IR) in mouse and rat models [
4]. Both classifications explored relative histopathological changes and the related mechanism of IR-induced SG injury. During the acute stage, the secretory function decreases, with no severe apoptosis or cell loss observed in SGs [
8,
9,
10]. Acute SG dysfunction is partially attributed to compromised M3-muscarinic receptors and water channels (e.g., aquaporin-5, AQP5) in the plasma membrane during and after the radiation [
11]. At the delayed stage, the amylase secretion, the saliva flow rate, and the volume of irradiated salivary glands were significantly decreased [
12]. Irradiation can damage the cellular DNA via free radicals in the nucleus and mitochondria and induce the death of the reproductive and functional cells (e.g., acinar progenitor cells and endothelial progenitor cells) in SGs [
13,
14,
15]. From 30 to 300 days post-IR, fibrosis, and fatty degeneration develop and consequently increase the dysfunction of salivary glands [
4,
12]. These mechanisms could also be an interpretation of the later cell loss, cell apoptosis, salivary secretion, and blood flow reduction during the chronic phase [
16]. To summarize, irradiation-induced SG damage can be divided into two mechanisms: (1) the acute phase, i.e., cellular dysfunction due to cell membrane damage, and the (2) delayed phase, i.e., a classical killing of progenitor cells because of DNA damage and disorders of the cellular microenvironment [
4,
8].
Most current treatment options for IR-induced SG hypofunction are palliative [
17] or preventive, resulting in a limited efficacy [
18,
19,
20]. The main objective of palliative therapy, such as salivary substitutes, is to relieve the symptoms and reduce the discomfort in patients with xerostomia but not to stimulate the salivary gland to secrete natural saliva [
21]. In other words, these treatments do not aim to reverse the acute damage to the cells or protect the progenitor and functional cells in SG. Therefore, the effect of these salivary substitutes is transient [
21,
22]. As another palliative therapy, the sialagogue treatment can stimulate the secretory function of SGs; however, it becomes ineffective if SG cells are already damaged and insufficient before the treatment.
Additionally, it has severe side effects and inconsistent outcomes for certain patients, especially those of advanced age [
23]. As for preventive therapies, such as intensity-modulated radiotherapy (IMRT) and radical scavengers, they have uncertain effectiveness in clinical trials [
24,
25,
26,
27]. Furthermore, surgical treatment (e.g., salivary gland transfer) is underutilized due to its invasiveness and complications, such as ipsilateral facial edema and neck numbness [
28,
29,
30]. Therefore, regenerative therapy has garnered tremendous interest in the field of treatment for SG hypofunction. Current knowledge of restoring saliva secretion comes from various experimental strategies, such as cell-based therapies [
31,
32,
33] and cell-free therapies [
34,
35]. Cell-based therapies have been reported as a regenerative option to increase the salivary flow rate and relieve the xerostomia caused by radiotherapy [
36,
37]. Various cells, including stem cells and non-stem cells from different tissues, restored IR-induced SG dysfunction by reducing cell apoptosis and protecting the structure and function of acinar cells in SGs [
37,
38,
39]. Recently, stem cell treatments have been investigated in phase I–II clinical trials [
33,
40,
41]. Evidence suggests that the paracrine effect is the primary mechanism for cell-based therapy in the SG post-IR [
37,
38,
42]. The paracrine factors released from cells allow tissue repair and regeneration via modulating the immune reaction, mitigating inflammation and fibrotic effects, promoting angiogenesis and neurogenesis, and preventing apoptosis [
37,
43,
44,
45]. Based on this theory, various cell-free therapies have been developed in the past few decades, including cell extract therapy, conditioned medium therapy, and others (e.g., extracellular vesicle therapy). This review focuses on cell-free therapy as an alternate strategy to palliative or conventional preventive treatments, with a particular emphasis on using cell extract (CE). First, we summarize a variety of cell extracts and their therapeutic uses in SG. Then, we provide optimization approaches for cell extract treatments from different aspects and compare CE with several other cell-free derivatives for the treatment of IR-induced SG hypofunction.
2. Cell Extract Therapies and IR-Damage SG
The cell extract (CE) is the heterogeneous mixture isolated from soluble components of cell lysates. It contains proteins, nucleic acids, lipids, carbohydrates, and organelles from cells [
43]. The CE is a cell lysate that can be obtained from all types of cells by breaking down their membranes. For many years, cell lysis has been used as a step for cell fractionation, organelle isolation, and protein extraction and purification. With the recent isolation and identification of various proteins, lipids, and genetic materials in CE, their crucial roles in regenerative medicine and in disease treatments are being reported. Currently, CE transplantation has been demonstrated as a cell-free treatment for various diseases, including wound-healing [
46], myocardial infarction and ischemic stroke [
47,
48], acute myeloid leukemia [
49], acute colitis [
50], osteoradionecrosis [
51], Alzheimer’s disease [
52], nerve injury [
53], obesity [
54], liver diseases [
55], Sjogren’s syndrome [
56], and irradiation-induced SG injury [
42] (
Table 1). CE harvested from different cell/tissue sources has been analyzed, including bone marrow cells [
42,
51], bone marrow stem cells [
56], bone marrow mononuclear cells [
47], adipose stem cells [
34,
46], spleen tissues [
34], embryonic stem cells [
57], salivary gland stem cells [
35], white blood cells [
58], and plant stem cells [
59]. Most of these showed the potential to mitigate the hypofunction of IR-injured SG.
2.1. Bone Marrow Stem Cell Extract and Bone Marrow Cell Extract Therapies
Mesenchymal stem cell (MSC) is first isolated from bone marrow, and bone marrow stem cell (BMSCs) has become one of the most well-studied stem cells to researchers. The extract from BMSC (BMSCE) was administered to various diseases. For example, Khubutiya et al. described BMSCE transplantation as a potential treatment for acetaminophen-induced liver failure as it reduced the area of necroses and increased the number of mitotically active cells in the liver [
55]. BMSCE treatment also could preserve the exocrine function of salivary and lacrimal glands by promoting cell proliferation and extracellular matrix formation, preventing fibrosis, and regulating immunomodulation [
56].
The use of whole bone marrow cell extract (BMCE) has grown in popularity due to its convenience and clinical feasibility. BMCE offers a convenient on-shelf source without the need for lengthy cell culturing. Yeghiazarian et al. isolated the BMCE and first compared its efficacy with intact bone marrow cells in a myocardial infarction animal model [
47,
64]. Results showed that BMCE was as effective as alive cells in reducing infarct size and cell apoptosis, enhancing vascularity, and improving cardiac function. In agreement with Yeghiazarin’s study, our previous study demonstrated that BMCE was as effective as whole bone marrow cells in repairing SG hypofunction [
42]. This evidence suggested that paracrine action is the principal mechanism of cell-based therapy, and BMCE might be an alternate treatment to intact lived cell treatment. In the following years, the efficacy of the BMCE was investigated by other research groups. Michel and colleagues demonstrated that BMCE significantly enhanced new bone formation in the irradiated bone of rats [
51]. A recent study revealed pain amelioration and anti-inflammatory effect of BMCE treatment in a peripheral nerve-injured mouse model [
68].
Additionally, Misouno et al. found that BMCE significantly reduced focus scores in the treated NOD mice (Sjogren’s syndrome mouse model) by inhibiting lymphocytic infiltration in SGs [
65]. Furthermore, this study explored the potential target proteins of BMCE by 2D liquid chromatography–mass spectrometry, including the downregulation of inflammation-related proteins (kallikrein 1-related peptidase and Calreticulin), Sjogren’s syndrome biomarker (Sjogren’s syndrome antigen B), apoptosis-related proteins (Caspase-8, CASP8-associated protein 2, and caspase recruitment domain protein 12), and the up-regulation of stem cell and development proteins (Nestin and Vimentin) and salivary gland biology markers (a-amylase, aquaporin 1 (AQP1), AQP5, parotid secretory protein (PSP). In addition to Sjogren’s syndrome, mouse and human BMCE transplantations were beneficial for IR-damaged SGs [
67]. Our recent study identified the human BMCE with three cell subpopulations (mononuclear cell, granulocytes, and red blood cells) from whole bone marrow and their CEs (BMCE, MCE, GCE, and RBCE), respectively. Results showed that BMCE and MCE provided therapeutic efficacy by improving the secretory function of IR-injured SG. Both of these cell extracts did not induce an obvious immune response; GCE was of more limited efficacy but induced an acute inflammatory response. In contrast, RBCE did not restore the salivary flow rate during the observation [
67]. In summary, BMSCE and BMCE, as well as specific cell extracts (MCE and GCE) derived from whole bone marrow sub-fractions, provided a promising treatment effect in SG diseases.
2.2. Embryonic Stem Cell, Adipose Stem Cell, and White Blood Cell Extract Therapies
In addition to bone marrow-derived CE, embryonic stem cells (ESC), white blood cells (WBC), and adipose stem cell (ADSC) are alternate cell sources for CE preparation have been investigated by researchers for disease treatments. The primary function of embryonic stem cell extract (ESCE) is the induction of differentiated cells. For example, ESCE induced to exhibit comparable properties of the ESCs [
57]. Additionally, ESCE therapy on wound healing provided valuable knowledge of ESCE to promote epithelial and granulosa cells to express pluripotency markers and to undergo de-differentiation [
69,
70]. These differentiated cells showed the multi-potential of differentiation after incubating with ESCE. However, both studies demonstrated that ESCs-related characteristic changes were only of short duration and could not be maintained for a longer duration. The death of progenitor and functional cells is the primary mechanism during the late stage of irradiated SGs and results in a substantial loss of SG secretory function. The cell de-differentiation capability of ESCE can hypothetically mitigate SG hypofunction by promoting local SG cells to de-differentiate into progenitor cells. Nevertheless, more research is needed to test this hypothesis.
Crocodile white blood cell (cWBC) extract has been used in treating ultraviolet radiation’s effect on the skin [
73]. One study revealed that cWBC extract significantly promoted cell proliferation and prevented ultraviolet-induced morphological change and skin pigmentation. Interestingly, the crocodile white blood CE induced apoptotic cell death to several cancer cell lines (including Hela, LU-1, LNCaP, PC-3, MCF-7, and CaCo-2 cells) [
58,
74], but no cytotoxicity towards non-cancerous Vero and HaCaT cells [
58]. Although the mechanism remains unknown, this phenomenon may help patients suffering from IR-induced SG hypofunction and reducing their risk of worsening their head and neck cancers. These findings indicated that WBC extract might be a potential source for treating IR-injured SG. However, more studies are needed to investigate the immunogenicity and the optimized dosage of WBC extract before administrating it in the clinics. So far, either ESCE or WBC extracts have yet to be tested in SG diseases.
ADSCs are considered one of the most promising adult stem cells for clinical application because they can be isolated from a plethora of adipose tissues. Adipose stem cell extract (ADSCE) is widely used as a potential treatment in various diseases, such as wound healing [
46], nerve injury and Alzheimer’s disease [
52], obesity [
54], acute inflammation [
50], ischemic stroke [
48] and IR-injured SGs [
34]. One study reported that ADSCE reduced fibrosis and preserved the smooth muscle content in a cavernous nerve injury model. Another study administered ADSCE to Alzheimer’s disease mice and reported the antioxidant and anti-apoptosis effects of ADSCE treatment [
52]. ADSCE also showed an anti-inflammatory effect on macrophage cells and suppressed LPS/IFN, induced NO, COX-2, and PGE2 production via downregulation of iNOS and COX-2 protein expression [
62]. Several other studies further confirmed the anti-inflammatory effect of ADSCE [
48,
50]. Additionally, ADSCE was used in treating SG hypofunction. Our previous study investigated the effect of mouse ADSCE on the IR-damaged SG model. Results showed that ADSCE significantly restored the secretory function of the damaged salivary glands and protected SG functional cells, blood vessels, and parasympathetic nerves [
34]. Therefore, ADSCE is expected to be used in tissue repair and regenerative medicine for SG hypofunction.
2.3. Other Cell Extract Candidates
Extracts derived from tissue-specific stem cells also provided a pronounced effect on the target organ treatment, such as SG stem cells. Human minor SGs can be obtained with minimal invasiveness and provide sufficient labial stem cells (LSC) for CE preparation [
35]. Furthermore, SG stem cells can differentiate into various cells, especially epithelial cells [
78]. Considering LSC’s extraordinary epithelial cell differentiation potential, the LSC extract (LSCE) was prepared and transplanted to rescue the hypofunction of IR-injured SGs [
35]. As expected, our result showed a significant increase in the salivary flow rate of damaged SG post-IR.
In addition to the common CEs described above, animal or plant tissues as well as plant stem cells could be sources for CEs. For example, the function of adipose tissue extract was investigated in vitro with human keratinocytes, fibroblasts, and adipose stem cells. Results showed that the adipose tissue extract promoted keratinocyte proliferation and stimulated fibroblasts and adipose stem cells migration [
75]. Additionally, spleen tissue extract was tested in an IR-injured SGs model, and results suggested that spleen cell extracts could mitigate SG hypofunction [
34]. Plant stem cell extracts were widely used in skin anti-aging and hair loss [
59,
79,
80,
81]. Despite few studies investigating SG diseases, plant stem cell extract could be considered for SG treatment due to its diverse properties. For example, plant stem cells could promote cell regeneration and viability against senescence and apoptosis of human stem cells and delay aging [
59]. Furthermore, the geranium sibiricum extract reduced the number of mast cells in the mouse skin tissue [
80], and birch stem cell extract showed a suppression effect when treated with esophageal squamous carcinoma cells in vitro [
82], which is a comparable inhibitory effect on cancer cells as the one observed with the crocodile white blood CE [
58].
In summary, several cell extracts have demonstrated their effectiveness in treating SG hypofunction, including ADSCE, BMSCE, BMCE, MCE, LSCE, and spleen cell extracts. GCE was effective but also induced an acute inflammatory response. The optimal cell source is still unknown, and further experiments are needed to address this issue. Furthermore, other types of CEs, such as ESCE, plant cell extract, and white blood cell extract, have not yet been tested for SG diseases. They may be effective treatments for IR-injured SGs due to their capacities to repair and regenerate tissues. For example, the de-differentiated function of ESCE might benefit injured SG stem cells to renew and differentiate into functional cells (e.g., acini cells), while plant and cWBC extracts could inhibit cancer cell growth and promote epithelial cell proliferation. These promising findings open new venues for a variety of treatments for IR-damage SGs.
4. Constituents in Cell Extracts and Their Mechanisms
CE is a group of heterogeneous mixtures containing inter-cellular materials, such as DNA, RNA, protein, lipids, and organelles from cells [
43,
88]. Recently, CE has been tested for regenerative medicine in various of disease models in vivo and in vitro. However, the active components in CE have yet to be fully identified. To this end, the first step is to determine which categories of molecules in a CE, for example, nucleic acids or proteins, contain the bioactive factors. The second step is to compare the constituents in CEs from different cell sources, such as adipose tissue or bone marrow. The third step is to profile the ingredient in CEs. The last step is to explore specific candidate factors in the CE responsible for the therapeutic effect and unveil their mechanisms of action.
In a previous study, we answered the first step by uncovering that the effective bioactive factors in the CE were proteins [
43]. We deactivated proteins in the CE by using the proteinase K combined and then heating at 95 °C to inactive proteinase K before injecting this deactivated CE into mice. Then, either normal saline, CE, or deactivated CE were administered in the IR-injury SG mouse model. Results showed that the deactivated protein CE injected was no better than the injection of saline, while the infusion of native CE restored the secretory function of SG. This finding indicated that the native proteins (but not the DNA, RNA, lipids, carbohydrates, or other small organelles) were the effective constituents in CE.
Analyzing the different constituents (proteins) in diverse CEs was tested next. Apart from the difference induced by various isolation methods, the type of cell source is another major factor influencing the components of CEs. A study compared the therapeutic effect and the constituents of the spleen, adipose, and bone marrow CE [
88]. All three CEs restored the hypofunction of IR-injured SGs and protected the functional cells, blood vessels, and parasympathetic nerves during the 8-week observation. Preliminarily, to analyze the protein components in CEs derived from different cell sources, a protein membrane array assay was used to profile angiogenesis-related factors in these three CEs. Results showed that the constituents and concentrations of certain growth factors differed between CEs. For example, a significantly lower SDF-1 was detected in the spleen CE in contrast with bone marrow and adipose CE. The adipose stem cell extract presented a higher number of angiogenesis-related factors than other CEs. One interesting finding is that the ADSCE showed less efficacy than the other two CEs, although it contained the most identified growth factors in this study. Aside from this difference, several overlapping proteins were identified in these CEs, such as MMP-9, CD26, and OPN. Another recent study confirmed these findings and showed that the concentrations of identified proteins in the mononuclear, granulocyte, and red blood cell extract from human bone marrow were different [
67]. However, unlike ADSCE, mononuclear cell extract contained more growth factors and provided the best therapeutic efficacy in treating the IR-injured SG. This study further compared the bone marrow cell extract from different species (mouse and human) and demonstrated that more angiogenic factors were detected in human BMCE. Altogether, CE derived from different cell/tissue sources contained several overlapping proteins and certain different constituents. This difference may influence the therapeutic efficacy of disease treatment.
Admittedly, only some studies systematically profile the stem cell extracts with comprehensive proteomic analysis, but several CEs had been semi-quantified by protein membrane assays. One of the studies presented 171 cytokines identified from ADSCE [
63], including angiogenic factors (FGFs, VEGFs, ANGs), tissue remodeling proteases (MMPs) and its inhibitors (TIMPs), stem cell homing chemokines (SDF-1), anti- and pro-inflammatory cytokines (IL-1β, 6, 8, 11, 17, IL-1ra, and TGF-β1), tissue repair/regeneration-related factors (BMPs, IGFs, PDGFs, and HGF), and many other cytokines. Three categories of proteins in mouse bone marrow cell extract were preliminarily screened, including angiogenesis, cytokines, and chemokines [
35,
43]. Several angiogenic factors were identified, such as CD26, FGF, HGF, MMPs, PF4, and SDF-1, while few cytokines (IL-1ra and IL-16) and chemokines were detected in mouse BMCE. There were 22 angiogenesis-related growth factors detected in human BMCE [
67] and 26 were found in the human labial gland stem cell extract [
35]. One study profiled the human adipose tissue cell extract (adipose liquid extract) through proteomics (
Figure 1) [
76]. A total of 1742 proteins were identified in the adipose tissue cell extract, most of them were from the cytoplasm (62.2%), followed by the nucleus (16.3%), extracellular space (9.3%), and plasma membrane (8.7%). These molecules were mainly involved in the cellular process, biological regulation, and metabolic process. These results demonstrated that the CE contains crucial growth factors and cytokines related to numerous physiological and pathological pathways in our body, and this is perhaps the reason for the broad systemic effect of the CE in treating a variety of diseases.
Different cell extracts might contain several overlapping proteins, such as FGF-1, -2, MMP-8, -9, VEGF, TIMP-1, CD26, PAI-1, and SDF-1 [
35,
43,
63,
67]. Many of these are multifunctional and play a role in treatment. FGF-2 is a highly expressed factor in the labial stem cell extract [
35], bone marrow cell extract [
67], and adipose stem cell extract [
34,
63]. It is a well-known mitogen that plays an essential role in angiogenesis and wound healing. FGF2 promotes fibroblasts and the epithelial and endothelial cell proliferation and induces the regeneration of tissues and blood vessels [
93,
94]. It also encourages the growth of acinar cells [
95], myoepithelial cells, ductal cells [
93], and the development and regeneration of salivary glands [
93,
96]. Studies reported that FGF2 protected IR-injured SGs by inhibiting radiation-induced apoptosis in vivo and in vitro [
95,
97]. Stromal cell-derived factor-1 (SDF-1) is another critical factor in CE that plays a role in the disease treatment [
98], with a potent capacity to repair damaged tissues by regulating immune response, inflammation, cell migration, vascularization, and neurogenesis [
98,
99,
100]. Interestingly, besides the angiogenesis factors, the anti-angiogenesis factors (such as TIMPs and PAI-1) were detected in CEs [
35,
63,
67]. These factors generally inhibit cell proliferation, migration, and angiogenesis [
101,
102,
103,
104,
105,
106]. However, they showed diverse benefits in treating IR-induced SG hypofunction. For example, most of these anti-angiogenic factors are natural inhibitors of the tumorigenesis [
102,
103,
104], which is the main advantage for head and neck cancer patients. Fang et al. reported that CE from bone marrow contained pro- and anti-angiogenesis factors that did not promote tumor cell proliferation [
34,
43]. The modulated interactions of these anti-angiogenic growth factors might be part of the reasons explaining this underexplored phenomenon.
Besides the effect of a single molecule, the interrelation between co-existed factors in a CE is a concern during treatment. VEGF, an angiogenic factor with neurotrophic and neuroprotective effects [
107], was identified in multiple CEs as FGF-2. Studies reported that other angiogenic factors were required to complement VEGF to promote vessel maturation because neo-vessels were unstable with the sole use of VEGF [
108,
109]. This implies that the synergistic effect of diverse proteins played a role during treatment and the therapeutic effect was attributed to the interactions of a variety of growth factors in CE rather than one or two vital factors. Apart from this synergistic effect, antagonist interactions of growth factors were also found in CEs. PAI-1 acts as an anti-angiogenesis factor and an inhibitor of urokinase (uPA), but both of them co-exist in labial stem cell extract [
35]. Studies reported that a fine balance of cell migration, wound-healing, embryogenesis, and angiogenesis happened between the PAI-1 and uPA regulation [
110]. Similarly, our previous study revealed that the therapeutic effect of labial stem cell extract was attributable to the interactions of pro- and anti-angiogenic factors in the CE [
35]. A similar relationship can be found between TIMPs and MMPs in CEs [
67]. Altogether, these results suggest that the therapeutic effects of different CEs are modulated by multiple factors as well as their interactions (synergism and antagonism). These factors are an essential element of the SG repair and regeneration process. More studies need to be carried out in order to decipher these complex interactions.
Indeed, instead of targeting a key specific molecule, CE treatment is a category of treatment given for a broad systemic change with a well-orchestrated cascade. Sam Zhou et al. reported that bone marrow CE partially restored serum proteomic homeostasis and re-established systemic balance to attenuate mechanical hypersensitivity in a nerve injury mouse model [
68]. Interestingly, no strong regulation was detected in the serum of CE-treated mice. These results suggested that CE treatment tends to regulate systemic homeostasis but not that of a single molecule. In addition, the serum from CE-treated nerve-injured mice no longer induced hypersensitivity in naïve mice. This finding verified that CE treatment, as a multifaced systemic approach, alleviated pain by causing a broad modification in the serum of the treated mice. Overall, we propose that the mechanism of CE treatment is complicated and comprehensive and that it tends to establish a new systemic balance by functional molecules and their interactions in CE to improve the pathologic microenvironment and consequently alleviate diseases, such as the IR-injured SG.
It is worth noting that the group of active factors in CEs remains unknown. Further studies are required to verify and purify the effective molecules in CEs (or separate the inactive substances from the CE) and unveil the different mechanism pathways behind them. Our recent study makes an effort to separate the subpopulations of bone marrow cells and compares the effect of three CE fractions (the cell extract from the mononuclear cell, granulocyte, and red blood cell) [
67]. Results showed that the mononuclear cell extract provided the best therapeutic efficacy, while the red blood cell extract did not significantly mitigate salivary hypofunction. This implied that the effect of the bone marrow CE treatment could be improved by removing the red blood cell from bone marrow. On the other hand, this finding suggests that the “purification” of the CE might help to obtain an optimized CE with the best therapeutic effect. Lastly, individual differences, including age, gender, and physical condition, might influence the cell extracts’ constituents and treatment effects. Therefore, it is important to clarify the difference by comparing the CEs from many of donors. Individual differences should be considered by the researchers when preparing the CEs from human cells or tissues and utilizing them in the clinic.