The association of prealbumin, transferrin, and albumin with immunosenescence among elderly males

Abstract

Objective

As people get older, the innate and acquired immunity of the elderly are affected, resulting in immunosenescence. Prealbumin (PAB), transferrin (TRF), and albumin (ALB) are commonly used markers to monitor protein energy malnutrition (PEM). However, their relationship with the immune system has not been fully explored.

Methods

In our study, a total of 93 subjects (≥65 years) were recruited from Tongji Hospital between January 2015 and February 2017. According to the serum levels of these proteins (PAB, TRF, and ALB), we divided the patients into the high serum protein group and the low serum protein group. Then, we compared the percent expression of lymphocyte subsets between two groups.

Results

All the low serum protein groups (PAB, TRF, and ALB) had significant decreases in the percentage of CD4+ cells, CD3⁺CD28⁺ cells, CD4⁺CD28⁺ cells and significant increases in the percentage of CD8⁺ cells, CD8⁺CD28⁻ cells. PAB, TRF, and ALB levels revealed positive correlations with CD4/CD8 ratio, proportions of CD4⁺ cells, CD3⁺CD28⁺ cells, CD4⁺CD28⁺ cells, and negative correlation with proportions of CD8⁺ cells, CD8⁺CD28⁻ cells.

Conclusions

This study suggested PAB, TRF, and ALB could be used as immunosenescence indicators. PEM might accelerate the process of immunosenescence in elderly males.

Keywords:

• Prealbumin
• transferrin
• albumin
• protein energy malnutrition
• immunosenescence

1. Introduction

The average human lifespan is steadily rising around the world, leading to an increase in the elderly population [1,2]. It is predicted that by 2050, the number of people over 65 will reach about 1.6 billion, accounting for almost 16.6% of the world’s population [3]. Affected by the aging process, the immune system undergoes various alterations, resulting in innate and adaptive immunity dysfunction, which is termed immunosenescence [4].

Immunosenescence results in substantial changes in almost all lymphocyte subsets within the immune system, especially T lymphocytes, which represent the cellular immune population [5]. Previous studies have indicated that aging brings an inversion of the CD4/CD8 ratio and loss of CD28 expression [6,7]. In addition, immunosenescence makes older people more prone to infections, cancers, neurodegenerative, and autoimmune diseases [8,9], leading to a high mortality risk and has caused extensive concern.

Functional immune response is associated with lipid and glucose metabolism, protein homeostasis, and inflammation activity [10,11]. High glucose can induce enhanced concentrations of reactive oxygen species (ROS) in human senescent CD8⁺ T cells via enhancing capacity to use glycolysis [12]. Lipid metabolism could change inflammatory gene expression and H₂O₂ production of different immune cells and affect the immune response of the body [13]. Patients with hypoproteinemia had significantly lower numbers and functions of CD4⁺, CD8⁺ T cells, and NK cells than healthy controls [14]. Up to now, many studies have shown that protein energy malnutrition (PEM) is associated with immunosenescence [15,16].

PEM is common in older adults because of multiple acute/chronic illnesses, disabilities, physical, and environmental changes [17–19]. As reported, PEM patients are more likely to have atrophy of the thymus and lymphatic tissue, leading to increased immature CD2⁺CD3⁻ and decreased mature CD3⁺ T lymphocytes [20]. In addition, PEM leads to a relatively insufficient concentration of glutamine in the body, which can affect lymphocyte RNA and protein synthesis through concanavalin A [21]. Many studies have also shown that PEM leads to decreased CD4 subsets and preserved CD8 subsets [22,23].

Serum prealbumin (PAB), transferrin (TRF), and albumin (ALB) can partially reflect PEM and host immune status [24–27]. However, their potential relationship with immunosenescence has not been fully explored. In this study, we comprehensively evaluated the changes of lymphocyte subsets at different serum protein (PAB, TRF, and ALB) levels in elderly males. It is hoped that this study will offer a better understanding of the relationship between these proteins and immunosenescence, and facilitate the early evaluation of immunosenescence.

2. Materials and methods

2.1. Study subjects

Our subjects were recruited from the Department of Geriatrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. This retrospective study made use of the data from January 2015 to February 2017. We enrolled patients over the age of 65 who were admitted to our department for different reasons. The exclusion criteria were as follows: (1) age <65 years, (2) females, (3) HIV infection, (4) severe hepatic or renal dysfunction, (5) immunosuppressive drug therapy, (6) chronic hematological diseases, (7) autoimmune or rheumatic disease, and (8) known or suspected malignancy. Initially, a total of 221 males were selected for further qualification. After applying our exclusion criteria, 128 males were excluded and 93 remained. The study protocol was approved by the ethical committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. All patients gave written informed consent.

Sarcopenia was defined as the presence of low skeletal muscle index (SMI) <8.87 kg/m² and grip strength <27 kg in men [28]. Sarcopenic obesity was defined if sarcopenia was accompanied by >25% of total body fat for men [29]. Frailty status was defined on the basis of at least 3 of the following factors: slowness, exhaustion, myopenia, weakness, and low energy expenditure [30].

2.2. Laboratory tests

Heparin anti-coagulating peripheral blood samples of patients were collected by vacuum blood collection tubes. According to the manufacturer’s instructions, PAB (normal range, 200–400 mg/L), TRF (normal range, 2.00–3.60 g/L), and ALB (normal range, 35–52 g/L) were tested using ROCHE COBAS 8000 (Mannheim, Germany). Serum PAB >200 mg/L, serum TRF >2.00 g/L, and serum ALB >35.0 g/L were defined as high level and all others were defined as low level.

2.3. Flow cytometric analysis

The main characteristics of lymphocytes were determined by flow cytometer. FACSCanto flow cytometer (BD Biosciences, San Jose, CA) was used to acquire data and Diva software (BD Biosciences) to calculate percentages as well as absolute numbers of lymphocyte subsets. Of 50 uL of whole blood was stained with 5 μL of the following antibodies: CD3 APC-H7-A, CD4 Pacific Blue-A, CD8 APC-A, CD25 PE-Cy7-A, CD28 FITC-A, CD45 Pacific Orange-A, CD56 ECD-A, and CD127 PE-A (BD Biosciences). It was then incubated in the dark at room temperature for 15 min and added 450 μL of BD FACS^TM lysing solution. Finally, BD TrueCount^™ tubes were used to accurate lymphocyte numbers.

2.4. Statistical analysis

All statistical analyses were conducted by using SPSS version 24.0 (Chicago, IL). Student’s t-test was used for comparisons between two independent groups. Results were expressed as the mean ± standard deviation (SD) and frequency (percentage). Correlation analysis was performed using the Pearson correlation test. The level of statistically significant was set at p < 0.05 (two-sided).

3. Results

3.1. Percent expression of lymphocyte subsets for individuals with CD4/CD8 ratio less than or greater than 1

At baseline, a total of 93 elderly males (≥65 years) were included in this study. The age ranged from 65 to 99 years with the mean (81.28 ± 8.65) years. The mean age of the individuals with CD4/CD8 ratio greater than 1 was younger than individuals with CD4/CD8 ratio less than 1 (p = 0.004). BMI, CMV infection, sarcopenia, sarcopenic obesity, waist circumference, and frailty status did not differ between the two groups. Subjects with CD4/CD8 ratio less than 1 showed a significant decrease in serum PAB levels (p = 0.034). Subjects with CD4/CD8 ratio less than 1 also showed significant decreases in the percentage of CD4⁺ cells (p < 0.001), CD3⁺CD28⁺ cells (p < 0.001), CD4⁺CD28⁺ cells (p < 0.001), and significant increases in CD8⁺ cells (p < 0.001), CD8⁺CD28⁻ cells (p < 0.001) (Table 1).

Table 1. Differences between CD4/CD8 > 1 group and CD4/CD8 < 1 group.

Parameter	CD4/CD8		t/χ^2	p Value
	>1 (n = 54)	<1 (n = 39)
Age (year)	79.17 ± 8.96	84.21 ± 7.34	−2.975	0.004*
BMI (kg/m^2)	24.12 ± 3.49	23.06 ± 3.22	1.482	0.142
CMV-IgG positive (%)	54 (100.0%)	39 (100.0%)	–	–
CMV-IgM positive (%)	1 (1.9%)	1 (2.6%)	0.000	1.000
Sarcopenia (%)	11 (20.4%)	15 (38.5%)	3.680	0.055
Sarcopenic obesity (%)	6 (11.1%)	7 (17.9%)	0.880	0.348
Waist circumference (cm)	74.80 ± 4.95	74.35 ± 5.34	0.409	0.683
Frailty status (%)	22 (40.7%)	20 (51.3%)	1.016	0.313
Protein
Prealbumin (mg/L)	221.70 ± 71.72	191.10 ± 61.80	2.149	0.034*
Transferrin (g/L)	2.03 ± 0.36	1.96 ± 0.42	0.946	0.347
Albumin (g/L)	39.15 ± 4.31	37.82 ± 5.27	1.344	0.182
CD4/CD8 ratio	1.85 ± 1.16	0.68 ± 0.20	6.244	<0.001*
Percentages of subsets (%)
CD4^+/T cells	58.19 ± 9.50	38.03 ± 10.43	9.692	<0.001*
CD8^+/T cells	36.02 ± 7.58	56.46 ± 10.57	−10.866	<0.001*
CD4^+CD25^str+CD127^dim+ (Tregs)/T cells	7.033 ± 2.49	7.31 ± 2.23	−0.553	0.581
CD3^−CD56^+CD45^+ (NK)/Lymphocytes	24.55 ± 12.75	21.47 ± 12.87	1.146	0.255
T cells (CD3^+CD45^+)/Lymphocytes	60.89 ± 12.45	66.60 ± 15.23	−1.983	0.059
CD3^+CD28^+/T cells	72.12 ± 14.20	53.03 ± 16.24	6.022	<0.001*
CD4^+CD28^+/T cells	52.71 ± 9.79	32.40 ± 11.67	9.109	<0.001*
CD4^+CD28^−/T cells	6.73 ± 8.08	6.31 ± 4.57	0.323	0.748
CD8^+CD28^+/T cells	17.90 ± 7.04	19.86 ± 8.28	−1.231	0.222
CD8^+CD28^−/T cells	17.07 ± 8.46	34.70 ± 13.01	−7.913	<0.001*

* p Value < 0.05.

Tregs: regulatory T cells; NK: natural killer.

3.2. Percent expression of lymphocyte subsets categorized by serum PAB, TRF, and ALB levels

The mean age of the subjects with high serum PAB levels was younger than subjects with low serum PAB levels (p = 0.004). Sarcopenia (p = 0.001), sarcopenic obesity (p = 0.026), and frailty status (p < 0.001) occurred significantly less frequently in the high PAB group. The low PAB group showed significant decreases in CD4/CD8 ratio (p = 0.019), CD4⁺ cells (p = 0.014), CD3⁺CD28⁺ cells (p = 0.004), CD4⁺CD28⁺ cells (p = 0.010) and significant increases in CD8⁺ cells (p = 0.002), CD8⁺CD28^- cells (p = 0.002) (Table 2).

Table 2. Percent expression of lymphocyte subsets in elderly males categorized by their serum prealbumin levels.

Parameter	Prealbumin		t/χ^2	p Value
	>200 (n = 48)	<200 (n = 45)
Age (year)	78.83 ± 8.93	83.89 ± 7.60	−2.932	0.004*
BMI (kg/m^2)	24.13 ± 2.85	23.20 ± 3.88	1.326	0.188
CMV-IgG positive (%)	48 (100.0%)	45 (100.0%)	–	–
CMV-IgM positive (%)	1 (2.1%)	1 (2.2%)	0.000	1.000
Sarcopenia (%)	6 (12.5%)	20 (44.4%)	11.767	0.001*
Sarcopenic obesity (%)	3 (6.3%)	10 (22.2%)	4.928	0.026*
Waist circumference (cm)	75.20 ± 3.99	73.99 ± 6.04	1.145	0.255
Frailty status (%)	13 (27.1%)	29 (64.4%)	13.091	<0.001*
Protein
Prealbumin (mg/L)	261.81 ± 42.84	152.40 ± 41.19	12.539	<0.001*
Transferrin (g/L)	2.12 ± 0.35	1.88 ± 0.39	3.138	0.002*
Albumin (g/L)	41.03 ± 4.02	35.99 ± 4.06	6.016	<0.001*
CD4/CD8 ratio	1.60 ± 1.32	1.09 ± 0.59	2.379	0.019*
Percentages of subsets (%)
CD4^+/T cells	53.18 ± 13.42	46.06 ± 13.88	2.515	0.014*
CD8^+/T cells	40.39 ± 12.63	49.08 ± 13.06	−3.265	0.002*
CD4^+CD25^str+CD127^dim+ (Tregs)/T cells	7.32 ± 2.30	6.96 ± 2.46	0.726	0.469
CD3^−CD56^+CD45^+ (NK)/lymphocytes	23.04 ± 12.93	23.49 ± 12.85	−0.171	0.864
T cells (CD3^+CD45^+)/lymphocytes	62.55 ± 13.67	64.07 ± 14.25	−0.524	0.602
CD3^+CD28^+/T cells	69.19 ± 15.65	58.70 ± 18.40	2.965	0.004*
CD4^+CD28^+/T cells	47.93 ± 13.20	40.20 ± 15.09	2.635	0.010*
CD4^+CD28^−/T cells	6.68 ± 6.86	6.42 ± 6.82	0.181	0.857
CD8^+CD28^+/T cells	19.61 ± 6.86	17.77 ± 8.30	1.168	0.246
CD8^+CD28^−/T cells	20.32 ± 11.96	28.88 ± 14.18	1.168	0.002*

* p Value < 0.05.

Tregs: regulatory T cells; NK: natural killer.

Compared to the low-level TRF group, the high-level TRF group was younger (p < 0.001) and had fewer sarcopenia (p = 0.006), sarcopenic obesity (p = 0.017), and frailty status (p = 0.006). Subjects with low serum TRF showed significantly lower CD4/CD8 ratio (p = 0.006) and proportions of CD4⁺ cells (p = 0.001), CD3⁺CD28⁺ cells (p = 0.001), CD4⁺CD28⁺ cells (p = 0.001) and significantly higher proportions of CD8⁺ cells (p = 0.002), CD8⁺CD28⁻ cells (p = 0.002) (Table 3).

Table 3. Percent expression of lymphocyte subsets in elderly males categorized by their serum transferrin levels.

Parameter	Transferrin		t/χ^2	p Value
	>2 (n = 50)	<2 (n = 43)
Age (year)	77.86 ± 8.42	85.26 ± 7.15	−4.527	<0.001*
BMI (kg/m^2)	24.29 ± 3.08	22.96 ± 3.65	1.919	0.058
CMV-IgG positive (%)	50 (100.0%)	43 (100.0%)	–	–
CMV-IgM positive (%)	0 (0.0%)	2 (4.7%)	0.680	0.409
Sarcopenia (%)	8 (16.0%)	18 (41.9%)	7.676	0.006*
Sarcopenic obesity (%)	3 (6.0%)	10 (23.3%)	5.725	0.017*
Waist circumference (cm)	74.98 ± 4.42	74.18 ± 5.81	0.761	0.448
Frailty status (%)	16 (32.0%)	26 (60.5%)	7.564	0.006*
Protein
Prealbumin (mg/L)	228.52 ± 58.17	186.02 ± 74.23	3.093	0.003*
Transferrin (g/L)	2.29 ± 0.24	1.67 ± 0.21	13.390	<0.001*
Albumin (g/L)	41.22 ± 3.55	35.53 ± 4.13	7.146	<0.001*
CD4/CD8 ratio	1.63 ± 1.32	1.03 ± 0.49	2.818	0.006*
Percentages of subsets (%)
CD4^+/T cells	54.02 ± 13.43	44.75 ± 13.18	3.348	0.001*
CD8^+/T cells	40.71 ± 13.14	49.12 ± 12.59	−3.137	0.002*
CD4^+CD25^str+CD127^dim+ (Tregs)/T cells	7.23 ± 2.40	7.05 ± 2.36	0.360	0.720
CD3^−CD56^+CD45^+ (NK)/lymphocytes	21.64 ± 12.17	25.14 ± 13.44	−1.318	0.191
T cells (CD3^+CD45^+)/lymphocytes	63.76 ± 12.58	62.73 ± 15.43	0.354	0.724
CD3^+CD28^+/T cells	69.46 ± 14.64	57.90 ± 19.13	3.297	0.001*
CD4^+CD28^+/T cells	48.70 ± 13.37	38.95 ± 14.34	3.392	0.001*
CD4^+CD28^−/T cells	5.94 ± 6.67	7.26 ± 6.98	−0.932	0.354
CD8^+CD28^+/T cells	19.05 ± 7.22	18.34 ± 8.10	0.452	0.653
CD8^+CD28^−/T cells	20.36 ± 11.78	29.23 ± 14.36	−3.270	0.002*

* p Value < 0.05.

Tregs: regulatory T cells; NK: natural killer.

Participants with high serum ALB levels were younger (p = 0.002) and had fewer sarcopenia (p < 0.001), sarcopenic obesity (p = 0.001) and frailty status (p = 0.024) than those with low serum ALB levels. The low serum ALB group showed significantly lower proportions of CD4⁺ cells (p = 0.032), CD3⁺CD28⁺ cells (p = 0.008), CD4⁺CD28⁺ cells (p = 0.013) and higher proportions of CD8⁺ cells (p = 0.017), CD8⁺CD28⁻ cells (p = 0.012) (Table 4).

Table 4. Percent expression of lymphocyte subsets in elderly males categorized by their serum albumin levels.

Parameter	Albumin		t/χ^2	p Value
	>35 (n = 72)	<35 (n = 21)
Age (year)	79.79 ± 8.58	86.38 ± 6.87	−3.226	0.002*
BMI (kg/m^2)	23.94 ± 3.29	22.78 ± 3.69	1.383	0.170
CMV-IgG positive (%)	72 (100.0%)	21 (100.0%)	–	–
CMV-IgM positive (%)	1 (1.4%)	1 (4.8%)	0.007	0.934
Sarcopenia (%)	13 (18.1%)	13 (61.9%)	15.521	<0.001*
Sarcopenic obesity (%)	5 (6.9%)	8 (38.1%)	10.657	0.001*
Waist circumference (cm)	75.15 ± 5.01	72.76 ± 5.08	1.919	0.058
Frailty status (%)	28 (38.9%)	14 (66.7%)	5.065	0.024*
Protein
Prealbumin (mg/L)	225.14 ± 64.23	153.10 ± 55.52	4.654	<0.001*
Transferrin (g/L)	2.10 ± 0.34	1.68 ± 0.35	4.891	<0.001*
Albumin (g/L)	40.52 ± 3.36	32.00 ± 2.20	10.914	<0.001*
CD4/CD8 ratio	1.46 ± 1.15	1.01 ± 0.60	1.690	0.095
Percentages of subsets (%)
CD4^+/T cells	51.41 ± 13.40	44.00 ± 14.98	2.173	0.032*
CD8^+/T cells	42.82 ± 12.54	50.72 ± 15.12	−2.423	0.017*
CD4^+CD25^str+CD127^dim+ (Tregs)/T cells	7.17 ± 2.21	7.08 ± 2.92	0.137	0.892
CD3^−CD56^+CD45^+ (NK)/lymphocytes	23.10 ± 12.97	23.80 ± 12.61	−0.217	0.828
T cells (CD3^+CD45^+)/lymphocytes	63.39 ± 13.61	62.91 ± 15.21	0.139	0.890
CD3^+CD28^+/T cells	66.72 ± 16.37	55.19 ± 19.72	2.709	0.008*
CD4^+CD28^+/T cells	46.21 ± 13.22	37.29 ± 17.18	2.536	0.013*
CD4^+CD28^−/T cells	6.29 ± 6.50	7.47 ± 7.88	−0.701	0.485
CD8^+CD28^+/T cells	19.29 ± 7.72	16.79 ± 7.00	1.331	0.187
CD8^+CD28^−/T cells	22.55 ± 13.15	31.00 ± 13.83	−2.560	0.012*

* p Value < 0.05.

Tregs: regulatory T cells; NK: natural killer.

3.3. Correlations between PEM parameters, age, and T-cell subsets

The effect of PEM on the different cell populations is shown in Figure 1. PAB levels were positively correlated with CD4/CD8 ratio (p = 0.003), proportions of CD4⁺ cells (p = 0.004), CD3⁺CD28⁺ cells (p = 0.001), CD4⁺CD28⁺ cells (p = 0.001) and negatively correlated with proportions of CD8⁺ cells (p = 0.004), CD8⁺CD28⁻ cells (p = 0.001). Likewise, TRF, ALB levels revealed positive correlations with CD4⁺ cells (p = 0.030, p = 0.023), CD3⁺CD28⁺ cells (p = 0.004, p = 0.005), CD4⁺CD28⁺ cells (p = 0.005, p = 0.009) and negative correlation with CD8⁺ cells (p = 0.021, p = 0.005), CD8⁺CD28⁻ cells (p = 0.017, p = 0.003).

Figure 1. Correlation analysis between PEM and T-cell subsets. A–F) Correlation between PEM parameters (prealbumin, transferrin, and albumin) and CD4/CD8 ratio, proportions of CD4⁺ cells, CD8⁺ cells, CD3⁺CD28⁺ cells, CD4⁺CD28⁺ cells, and CD8⁺CD28⁻ cells. Red indicates prealbumin, blue indicates transferrin, and brown indicates albumin. PEM: protein energy malnutrition.

Dynamic changes in PEM parameters with age are shown in Figure 2. Participants demonstrated significant decreases in the serum PAB (p = 0.007), TRF (p < 0.001), and ALB (p < 0.001) levels with increasing age. Correlation analysis between lymphocyte subsets and age is shown in Figure 3. A decreasing trend in the proportions of CD4⁺ cells (p = 0.010), CD3⁺CD28⁺ cells (p < 0.001), and CD4⁺CD28⁺ cells (p < 0.001) was found to be associated with aging in subjects. In contrast, proportions of CD8⁺ cells (p = 0.001) and CD8⁺CD28⁻ cells (p < 0.001) were all positively correlated with aging.

Figure 2. Correlation analysis between PEM and age. A–C) Correlation between PEM parameters (prealbumin, transferrin, and albumin) and age. Red indicates prealbumin, blue indicates transferrin, and brown indicates albumin. PEM: protein energy malnutrition.

Figure 3. Correlation analysis between age and proportions of T-cell subsets. A–F) Correlation between age and CD4/CD8 ratio, proportions of CD4⁺ cells, CD8⁺ cells, CD3⁺CD28⁺ cells, CD4⁺CD28⁺ cells, and CD8⁺CD28⁻ cells.

A correlation between different PEM parameters with one another is shown in Figure 4. Serum PAB was significantly positively correlated with TRF (p < 0.001) and ALB (p < 0.001). Similarly, strong positive correlations were also detected between serum TRF and ALB (p < 0.001).

Figure 4. Correlation analysis between different PEM parameters with one another. A) Correlation between prealbumin and transferrin. B) Correlation between prealbumin and albumin. C) Correlation between transferrin and albumin.

4. Discussion

More and more people are becoming recognized that the immune system declines with age, which is known as immunosenescence [4]. The geriatric population with PEM contributes to the increased risk of immune aging [15,16]. This study investigated association of PEM parameters (PAB, TRF, and ALB) with immunosenescence among elderly males. Our results showed that PAB, TRF, and ALB are clinical indicators for evaluating immunosenescence.

Inversion of the CD4/CD8 ratio is a frequent finding in older age, considered as an important indicator of immunosenescence. Our results showed that the inverted CD4/CD8 ratio was associated with advanced age and loss of CD28 expression in T cells. In addition, the accumulation of CD28⁻ T cells in the CD8 subsets is one of the most prominent changes in aging humans and represents impaired T-cell immunity [9,31]. CD28 is a B7 ligand and a co-stimulatory receptor expressed on the T-cell surface, it is vital for the second signal generated from CD28-B7 ligand interactions to activate and sustain the activity of T cells [32,33]. Repeated antigen stimulation could lead to loss of CD28 expression in T cells with age [34].

PEM is a worldwide health problem and is frequently observed among adults aged over 65 years [17,35]. PEM is associated with reduced immune responses and can drastically impair T-cell proliferation and cytokine production. PAB, TRF, and ALB are commonly used in clinic to evaluate protein nutritional status and monitor therapeutic effects. This study showed that low serum PAB and TRF groups had a significant decrease in CD4/CD8 ratio and all the low serum protein groups (PAB, TRF, and ALB) had a CD28 loss in T cells. This finding is consistent with the lymphocyte phenotype of immunosenescence, indicating that PEM might participate in immunosenescence [16,22,36]. Furthermore, immunosenescence markers, such as the proportions of CD8⁺CD28⁻ cells, CD4 cells, and the CD4/CD8 ratio were also associated with cytomegalovirus (CMV) infection [37]. Similar to that reported elsewhere, our subjects appear to constitute a higher risk of CMV infection and sero-prevalence rates approach 100% in old age [38]. Hence, we are not able to evaluate the impact of secondary/tertiary CMV infection on serum protein levels (PAB, TRF, and ALB) in this study.

The prevalence of PEM has previously been reported associated with age, sarcopenia, sarcopenic obesity, and frailty status. In line with the above studies, our results revealed all the low serum protein groups (PAB, TRF, and ALB) were older and more likely to have sarcopenia, sarcopenic obesity, and frailty status. The underlying mechanism of sarcopenia, sarcopenic obesity, and frailty is superimposed pathologically to some extent, including gut dysbiosis, endotoxemia, chronic inflammation, and endocrine disturbance [39]. Immune aging accompanied by changes in metabolism functions of immune cells is responsible for the environment of the skeletal muscle, eventually causing fibrosis, atrophy, and other consequences of muscle [40]. Our findings highlight a fact that sarcopenia, sarcopenic obesity and frailty status may be involved in the reduction of PAB, TRF, and ALB levels favoring the immunosenescence. However, how these different complications alter the immunonutritional status require further investigation.

There are a few limitations in this article. First, this is a single-center study with a small sample size. Further studies including females and larger populations are needed to demonstrate the validity of our results. Second, serum PAB, TRF, and ALB levels are not sufficient to indicate malnutrition. Future research should also apply nutrition scales to assess nutritional status, such as the Subjective Global Assessment, Mini-Nutritional Assessment, and Detailed Nutritional Assessment [41–43]. Third, due to the retrospective design, cytokines and sex hormones cannot be complemented for detection and need further exploration. In addition to these limitations, we believe that our study is valuable in evaluating accelerated immunosenescence. In clinical practice, the monitoring of PAB, TRF, and ALB is cheaper and more convenient compared with flow cytometric analysis, which provides the possibility for future individualized and large-scale clinical applications.

5. Conclusion

This study indicated that low serum PAB, TRF, and ALB levels were associated with T-cell immunosenescence phenotypes, such as decreased CD4/CD8 ratio and loss of CD28 expression in T cells. PAB, TRF, and ALB could be used as markers for evaluating accelerated immunosenescence in elderly males.

Ethical approval

The study was conducted in accordance with the Declaration of Helsinki, and approved by the ethical committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (protocol code TJ-C20141112 and 27 November 2014). All participants provided written informed consent. No experimental interventions were performed.

Author contributions

Conceptualization, XQQ, RDZ, and CTZ; methodology, SQJ, LR, FJY, and RDZ; formal analysis, SQJ and RDZ; data curation, SQJ, LR, FJY, and RDZ; writing-original draft preparation, RDZ, FJY, and SQJ; writing-review and editing, XQQ and CTZ; supervision, XQQ and CTZ; funding acquisition, XQQ and CTZ.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work is supported by Shenzhen Science and Technology Program (JCYJ20230807151309019), Guangdong Basic and Applied Basic Research Foundation (2021A1515220123), and National Key R&D Program of China (2020YFC2008000).