key: cord-0837108-gxz41nah
authors: Martignoni, Marcella; Benedetti, Margherita; Davey, Gavin P.; Tipton, Keith F.; McDonald, Andrew
title: Degradation of thymic humoral factor 2 in human, rat and mouse blood: An experimental and theoretical study
date: 2020-06-05
journal: Biochim Biophys Acta Proteins Proteom
DOI: 10.1016/j.bbapap.2020.140467
sha: 9db3b8978a53c39f6c370bb9aa8144d696f353ca
doc_id: 837108
cord_uid: gxz41nah

The degradation of the immunomodulatory octapeptide, thymic humoral factor γ2 (THF-γ2, thymoctonan) has been studied in whole blood samples from human, rat and mouse. The peptide, Leu-Glu-Asp-Gly-Pro-Lys-Phe-Leu, was shown to be rapidly degraded by peptidases. The half-life of the intact peptide was less than 6 min at 37 °C in blood from the three species tested. The main fragments formed from THF-γ2 were found to be Glu-Asp-Gly-Pro-Lys-Phe-Leu (2–8), Asp-Gly-Pro-Lys-Phe-Leu (3–8) and Glu-Asp-Gly-Pro-Lys (2–6) in human and in rat blood and 2–8 and 2–6 in mouse blood. Analysis of the time course of degradation revealed a sequential removal of single amino acids from the N-terminus (aminopeptidase activities) in a process that was apparently unable to cleave the Gly-Pro bond (positions 4–5 in the peptide) together with an independent cleavage of the Lys-Phe bond (positions 6–7 in the peptide) to release the dipeptide Phe-Leu. This behaviour and the effects of inhibitors showed the involvement of metallo-exopeptidases in the N-terminal digestion and a phosphoramidon-sensitive metallo-endopeptidase in the cleavage of the Lys-Phe bond. The degradation patterns in human blood were modelled in terms of the competing pathways involved approximating to first-order kinetics, and an analytical solution obtained via the method of Laplace Transforms. The half-life of THF degradation in whole rat blood sample was found to be significantly lower than in human or mouse.

Thymic humoral factor  2 (THF- 2, thymoctonan; THF) is an immunomodulatory octapeptide that was originally isolated from calf thymus and shown to have the amino-acid sequence:

Leu-Glu-Asp-Gly-Pro-Lys-Phe-Leu [1] . The peptide was shown to account for the observed min [7] and preliminary studies suggested that metabolism was faster in whole blood [8] . This led to the synthesis of analogues with greater stability [9] , a process that could be aided by the identification of sites that are susceptible to degradative enzymes in blood. Since the behaviour in whole blood might be expected to be more complicated because of the co ntributions of cell-associated peptidases, an aim of the present study was to examine the rates of degradation product formation in human blood in vitro for comparison with the behaviour reported in plasma.

As there may be species differences in THF metabo lism the behaviour in mouse and rat blood was also studied for comparison with that determined in the human.

Attempts were made to model the behaviour of the system in terms of the competing pathways involved and the application of the Laplace transform [10] to the data obtained with human blood was shown to provide a more robust model of the system that could be applied to the degradation of other peptides and different deterministic systems.

THF- 2 trifluoroacetate, [ 3 H]-THF- 2 trifluoroacetate labelled in the proline residue (specific activity 1.17 TBq/mmol, and [ 3 H]-THF- 2 trifluoroacetate, labelled in the C-terminal leucine residue (specific activity 1.75 TBq/mmol), synthesized as described by Fontana et al. (1996) [11] were kindly provided by Dr P. Dostert (Pharmacia, Italy). Each had a radiochemical purity > 9 8 %, as assessed by TLC and HPLC. The unlabelled peptides H-Glu-Asp-Gly-Pro-Lys-Phe-Leu-OH , H-Asp-Gly-Pro-Lys-Phe-Leu-OH, H-Gly-Pro-Lys-Phe-Leu-OH, H-Pro-Lys-Phe-Leu-OH, H-Leu-Glu-Asp-Gly-Pro-Lys-OH, H-Glu-Asp-Gly-Pro-Lys-OH were from the same source. The dipeptide H-Phe-Leu-OH and L-leucine were obtained from Sigma-Aldrich Chemical Co.

analyses.

Separations were carried out on Silica gel plates (Merk-Millipore F254) 2 0 2 0  cm, 0.5 mm thickness. The mobile phase was water-methanol-acetic acid-chloroform (6:14:20:40 by volume).

After drying the TLC plates in a current of warm air, the gel was scraped from 1 cm sections into scintillation vials containing 1 ml Soluene-350. 10 ml Hionic-Fluor was then added for liquid scintillation counting.

Under these conditions R f values of THF and the synthetic peptides that could be derived from it were as shown in Table 1 . These were used to identify the breakdown products from the zonal analysis of the blood incubations. where half-life, 1 / 2 = ( ln 2 ) / tk . Curves were fitted using the program Prism 8 (GraphPad Software, LLC.) and values are expressed  SEM. Apparent rate constants for the appearance of metabolite peptides were estimated by curve fitting the concentration-time curves for each substrate. The peptide degradation patterns for THF in human blood were fitted using the analytical functions described in a later (section 3.3).

THF was extensively degraded when incubated in human, mouse or rat blood. The time courses of degradation of the proline-labelled THF in rat and mouse blood are shown in Incubations and TLC analyses were performed as described in the text. Each point is the mean value  SEM of 3 independent determinations with blood samples from 3 separate individuals.

Studies with C-terminal leucine-labelled THF (not shown) showed the release of the dipeptide Phe-Leu (7-8), followed the same curve as that of peptide 2-6. As might be expected from the greater genetic heterogeneity, the behaviour of THF in blood samples from different human donors showed a greater variability than those from rat and mouse, as is evident from the larger errors with the human data (Fig. 2) . However, these differences were largely reflected in the rates of degradation rather than the nature of the products formed.

Comparison of degradation profiles in Figs 1 and 2 shows the metabolic patterns to be J o u r n a l P r e -p r o o f Journal Pre-proof broadly similar for rat and human blood but with some differences when mouse blood was used.

Analysis showed that THF was degraded in three main fragments in human and rat blood: the heptapeptide 2-8 (Glu-Asp-Gly-Pro-Lys-Phe-Leu), the hexapeptide 3-8

(Asp-Gly-Pro-Lys-Phe-Leu) and the pentapeptide 2-6 (Glu-Asp-Gly-Pro-Lys).

It can be seen from Figs 1A and 2 that peptide 2-8 (Glu-Asp-Gly-Pro-Lys-Phe-Leu) appeared rapidly, after only 1 min of incubation in human and rat blood, and that its concentration, expressed as the mean value  SD % of the total peptides present, increased until about 10 min (39.28 %  9.3 in rat blood; 39.69 %  6.9 in human blood). After this time, the concentration of this degradation product started to decrease. (Gly-Pro-Lys-Phe-Leu) were only detectable in small amounts ( < 10 % of total), these included a product that did not correspond to any of the markers used (UK-1). This became detectable after about 20 min and may represent peptide 3-6 (Asp-Gly-Pro-Lys), formed from 2-6, or 4-6 (Gly-Pro-Lys) formed from 4-8. This peptide was also detected in rat (Fig. 1A ), but not human, blood after extended incubation periods with THF. Amastatin is an inhibitor of aminopeptidases from several sources [12, 13] and thus the above results would support such a conclusion.

The degradation at the C-terminal side of the lysine residue in position 6 would be consistent with the involvement of an enzyme with a trypsin-type specificity. However, this activity was not due to the activity of a such an enzyme since the serine peptidase inhibitor AEBSF Phosphoramidon is an inhibitor of some metallo-endopeptidases such as membrane metallo-endopeptidases (EC 3.4.24.11) [14] , and the thermolysin (EC 3.4.24.27) group [15] .

However these enzymes are also inhibited by 1,10-phenanthroline. In the present case, as shown in 

The overall catabolic pattern for THF is summarized in Fig. 4A . 

The Laplace Transform Method is a convenient and robust means for obtaining solutions to networks of competing first-order ordinary differential equations (ODEs) through algebraic manipulation [10, 16] . Assuming that the initial concentrations of all peptide metabolites are zero initially, and denoting the initial concentration of THF by 

A graph of these functions as applied to the metabolism of THF in human blood is shown in Fig. 5 . While analytical solutions to systems of linear ODEs with constant coefficients can be obtained comparatively easily using the method of Laplace transforms, the use of curve-fitting methods to estimate the parameters is hampered by the high degree of sensitivity of higher-order exponentials functions to measurement error [17] . Furthermore, the slow rate of formation of some J o u r n a l P r e -p r o o f of the products and the low levels obtained within the time of the experiments will compound the curve-fitting errors. Nevertheless, the satisfactory agreement between the simulated curve in Fig. 5 and the experimentally observed results suggests that the model shown in Fig 4B provides a good explanation for the degradation process.

In Fig. 1A , the majority of the flux appears to end in 2-6, with 1-6 appearing only briefly as an intermediate. UK-1 starts to increase as 3-8 peaks, thereafter their traces appearing to mirror the other, which might indicate a reaction of 3-8 to 3-6, according to the scheme of Fig. 4B . Similarly, 4-8, which appears as the final degradation product in human blood, in rat and mouse, appears in low amounts, peaking at 20-30 minutes, suggesting 4-6 as a possible candidate for the unknown compound in rat. The model for the degradation of THF in human blood was therefore adapted to include additional rate constants for the formation of 3-6 (Asp-Gly-Pro-Lys), formed from 2-6, and 4-6 (Gly-Pro-Lys), formed from 3-6 or 4-8. The ODE system (1) was modified to include additional equations representing the formation of 3-6 ( G ) and 4-6 ( H ), 7 1 0

with d C /d t modified to include the additional term 7 kC  to account for the formation of 3-6 from 3-8 (Fig. 4B ). The new system of ODEs was integrated numerically in R [18] using the deSolve library. Fig. 6 shows the simulated time courses of the rat model, which reproduce aspects of both the rat and mouse data (Fig. 1) , including the delayed rise of the "unknown" peptide, here taken to be 4-6. A lower rate of decay of THF was chosen, according to the findings in mouse. In the absence of inhibition, the percentage of 3-6 would be expected to rise to a proportion of the initial THF percentage, with the remainder being the terminal product, 4-6. The decline in activity after 30 minutes, which is evident in the mouse data, could be an indication of e nzyme inactivation, or the presence of an inhibitor, neither of which are accommodated by the present linear model. 

As might be expected from the peptidases present in whole blood, the degradation of THF in human blood showed a more diverse pattern of products than that reported for plasma by Bramucci et al [7] . The differences are summarized in Fig. 7 . Only three major products with

Leu-Glu-Asp-Gly-Pro-Lys (1-6) being formed rapidly and Glu-Asp-Gly-Pro-Lys (2-6) and

Phe-Leu (7-8) appearing more slowly in plasma. The half-life of THF in plasma was calculated to be about 12 min. Figure 7 : Comparison of the THF- 2 metabolic fragments determined in human blood with those reported for human blood plasma [7] .

Comparison with the behaviour in whole blood, reported here, shows a more rapid degradation with a half-life of less than 4 min, which is over three times lower than the value obtained in plasma [7] . shown to be present in blood serum. Although this zinc-containing enzyme is sensitive to addition to neprilysin (neutral endopeptidase; EC 3.4.24.11) [20, 21] , peptidyl-dipeptidase A (ACE, angiotensin converting enzyme; EC 3.4.15.1) [22] and the erythrocyte Kell protein [23] .

Both peptidyl-dipeptidase A [7] and neprilysin [21] have been shown to catalyse this cleavage in isolated THF. This activity can be seen to have a proportionally greater role in the degradation in plasma than it does in whole blood. The presence of heparin, which has been shown to inhibit neprilysin [24] , may have affected the results, although that anticoagulant was present in both the plasma and whole-blood samples.

Studies with synthetic analogues of THF have shown the Phe residue at position 7 to be important to its effects on the impaired blastogenic response of phytohemagglutinin(PHA)-stimulated T-lymphocytes from uremic patients with infectious diseases [25, 26] . The degradation patterns in each of the species indicate that the facility in which this is removed by cleavage of the Lys-Phe bond, to yield peptides 1-6 and 2-6, may be an important factor in limiting the effectiveness of THF- 2 in vivo. However, amino-acid substitutions at positions 1-5 have also suggested that alterations to each of these residues decrease the potency of THF- 2 indicating that proteolysis at the N-terminal will also be deleterious. From this perspective, the differences in degradation patterns between mouse, rat and human may be of little importance. However, mouse would appear to be a less satisfactory model for the proteolysis in blood than the rat, although further work would be necessary to determine whether this behaviour was specific to the mouse strain used or to their age or weight. As might be expected the variation between the three human samples was greater than those from the rodents, but this affected the rates of product formation rather than the nature of the products formed. It would be of interest in the future to investigate the effects of diseases, such as metabolic syndrome and type 2 diabetes, where the plasma levels of neprilysin and peptidyl-dipeptidase A have been shown to be elevated [27, 28] . Inhibitors of both these enzymes, which have been used in hypertension and have been suggested to be useful the treatment of type 2 diabetes [29, 30] , may also prolong the effectiveness of THF- 2. The suggestion that ACE inhibitors may be of value in the treatment of coronavirus COVID-19 raises the possibility that the addition of THF- 2 might be beneficial because of its immune-stimulating actions [31] .

The model of THF degradation pathways by numerical solution of the differential equations has been shown to approximate the data from mouse and rat, although further work is values of the enzymes involved that apparent first-order kinetics are followed.

Highlights  The immunomodulatory octapeptide THF-2 is rapidly degraded in human and rodent blood. 

The half-life of the intact peptide was <6 min in all species. 

The nature of the peptidases involved was determined by the use of specific inhibitors.  Laplace transforms were used to model the degradation patterns in human blood.

J o u r n a l P r e -p r o o f Journal Pre-proof 

Thymic humoral factor gamma 2: purification and amino acid sequence of an immunoregulatory peptide from calf thymus

Increased mitogenic reactivity of normal spleen cells to T lectins induced by thymus humoral factor (THF)

Effect of synthetic thymic humoral factor (THF-gamma 2) on T cell activities in immunodefficient ageing mice

THF, a thymic hormone, promotes interleukin 2 production in intact and thymus-deprived mice

Treatment of murine cytomegalovirus salivary-gland infection by combined therapy with ganciclovir and thymic humoral factor gamma 2

Thymic humoral factor-gamma 2 (THF-gamma 2) immunotherapy reduces the metastatic load and restores immunocompetence in 3ll tumor-bearing mice receiving anticancer chemotherapy

Degradation of thymic humoral factor gamma2 by human plasma: involvement of angiotensin converting enzyme

In vitro metabolism of THF-γ2 in human, rat and mouse blood

Synthesis of an immunologically active analog of thymic humoral factor-γ2 with enhanced enzymatic stability

From MathWorld-A Wolfram Web Resource

Synthesis of tritium labelled thymic humoral factor γ2 (THF-γ2, thymoctonan)

Purification by affinity chromatography using amastatin and properties of aminopeptidase A from pig kidney

Characterization of membrane-bound aminopeptidases from rat brain: identification of the enkephalin-degrading aminopeptidase

Substance P and [Leu]enkephalin are hydrolyzed by an enzyme in pig caudate synaptic membranes that is identical with the endopeptidase of kidney microvilli

Binding between thermolysin and its specific inhibitor, phosphoramidon

Compartmental Models and Their Application

Applied Analysis

R: A language and environment for statistical computing

The MEROPS database of proteolytic enzymes, their substrates and inhibitors in 2017 and a comparison with peptidases in the PANTHER database

The neprilysin (NEP) family of zinc metalloendopeptidases: genomics and function

Hydrolysis of thymic humoral factor gamma 2 by neutral endopeptidase

Peptidyl dipeptidase A: angiotensin I-converting enzyme

The Kell protein of the common K2 phenotype is a catalytically active metalloprotease, whereas the rare Kell K1 antigen is inactive. Identification of novel substrates for the Kell protein

Neprilysin inhibits angiogenesis via proteolysis of fibroblast growth factor-2

Functional roles of phenylalanine 7 of thymic humoral factor-γ2 in the impaired blastogenic response of uremic Tlymphocytes

Synthesis and immunological effect of thymic humoral factor-gamma 2 analogues

Angiotensin converting enzyme (ACE) activity levels in insulin-independent diabetes mellitus and effect of ACE levels on diabetic patients with nephropathy

Neprilysin, obesity and the metabolic syndrome

Neprilysin inhibition: a new therapeutic option for type 2 diabetes?

Modulation of metabolic control by angiotensin converting enzyme (ace) inhibition

COVID-19) infection and renin angiotensin system blockers

Validation, Investigation; Margherita Benedetti: Conceptualization, Project administration, Funding acquisition

Writing -Review & Editing

Tipton: Conceptualization, Methodology, Resources, Writing -Original Draft, Supervision, Visualization, Funding acquisition

Methodology, Formal analysis, Writing -Original Draft

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.J o u r n a l P r e -p r o o f