Skip to content
BY 4.0 license Open Access Published by De Gruyter Open Access December 13, 2019

Global properties of virus dynamics with B-cell impairment

  • Ahmed M. Elaiw EMAIL logo , Safiya F. Alshehaiween and Aatef D. Hobiny
From the journal Open Mathematics

Abstract

In this paper we construct a class of virus dynamics models with impairment of B-cell functions. Two forms of the incidence rate have been considered, saturated and general. The well-posedness of the models is justified. The models admit two equilibria which are determined by the basic reproduction number R0. The global stability of each equilibrium is proven by utilizing Lyapunov function and LaSalle’s invariance principle. The theoretical results are illustrated by numerical simulations.

MSC 2010: 34D20; 34D23; 37N25; 92B05

1 Introduction

The study of within-host virus dynamics using mathematical modeling has been an interesting topic to research in the last decades. A proper model could provide insights of a better understanding of the virus dynamics and clinical treatments used to fight against it. In an infection process, the interaction between viruses and cells can be seen as an ecological system within the infected host. A wide of mathematical models focused on exploring the interaction between three basic compartments, uninfected cells (U), infected cells producing viruses (I) and viruses (P). A basic model of virus dynamics was originally developed by Nowak and Bangham [1] which has become highly used by experimentalists and theorists (see e.g., Nowak and May [2]). The model presented in [1] is given by:

U˙=ϱγUωUP, (1)
I˙=ωUPβI, (2)
P˙=ϰIξP, (3)

where U, I and P are the concentrations of uninfected cells, infected cells and viruses, respectively. The parameters ϱ, γ, ω, β, ϰ and ξ are positive. The full description of the model was given in [1]. A huge number of papers have been published as extension of the basic model (see, e.g., [3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22]).

The immune response plays a critical role in controlling the virus spreading. The specificity and memory in adaptive immune responses are the responsibility of lymphocytes. B cells and T cells are the two main types of lymphocytes. The function of T cells is to recognize and kill infected cells, while the function of B cells is to produce antibodies which bind to virus particles and mark it as a foreign structure for elimination by other cells of the immune system. Antibody alone can neutralize, and thus protect against, viruses [23]. The virus dynamics model with B cell immune response was presented by Murase et al. [24] as

U˙=ϱγUωUP, (4)
I˙=ωUPβI, (5)
P˙=ϰIξPρPC, (6)
C˙=εPCμC, (7)

where C is the concentration of B cells. Many extended models are developed with B cell immune response (see, e.g., [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36]).

In certain circumstances, some viruses can suppress immune response or even destroy it especially when the load of viruses is too high. Models with T cell immune impairment were studied several times (see, e.g., [37, 38, 39, 40]). In addition, there are factors affect B cell function and cause the impairment of B cell [41, 42, 43]. These factors include the following; malnutrition, tumors, cytotoxic drugs, irradiation, aging, trauma, some diseases (e.g., diabetes) and immunosuppression by microbes, e.g., malaria, measles virus but especially HIV [23]. In a very recent work, Miao et al. [44] have proposed a virus dynamics model which includes: humoral impairment, time delay, reaction-diffusion, and logistic growth of the target cells. Due to the complexity of the model presented in [44], the global stability analysis of the model’s equilibria did not studied. Studying the global stability of equilibria for virus dynamics models will give us a detailed information and enhances our understanding about the virus dynamics. Therefore, many mathematician have paid great efforts to study global stability of systems in virology (see, e.g., [7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19] and [45, 46, 47, 48, 49, 50, 51, 52, 53, 54]) and epidemiology (see, e.g., [55, 56, 57]).

In [44], the incidence rate of infection is given by bilinear. In reality, the bilinear incidence may not accurate to characterize the virus dynamics during different stages of infection especially when the concentration of the viruses is high [8]. Therefore, in the present paper, we propose viral infection model with B-cell impairment and with two nonlinear forms of the incidence rate, saturation and general. We show that the solutions of the model are nonnegative and bounded. The global stability of the equilibria is established by constructing Lyapunov functions and applying LaSalle’s invariance principle.

2 Model with saturation

In this section we propose a virus dynamics model including B-cell impairment and saturated incidence as:

U˙=ϱγUωUP1+αP, (8)
I˙=ωUP1+αPβI, (9)
P˙=ϰIξPρPC, (10)
C˙=εPμCϑCP, (11)

where, ϑ PC is the B-cell impairment rate and α ≥ 0 is a saturation constant.

2.1 Basic properties

We define the compact set

Ω=U,I,P,CR04:0Us1,0Is1,0Ps2,0Cs3 (12)

where si > 0, i = 1, 2, 3.

Proposition 1

The set Ω is positively invariant for model (8)-(11).

Proof

We have

U˙U=0=ϱ>0,I˙I=0=ωUP1+αP0, when U,P0,P˙P=0=ϰI0, when I0,C˙C=0=εP0, when P0.

Thus R04 is positively invariant for model (8)-(11). Let F(t)=U+I+β2ϰP+βξ4εϰC, then

F˙(t)=ϱγUωUP1+αP+ωUP1+αPβI+β2Iβξ2ϰPβρ2ϰPC+βξ4ϰPβξμ4εϰCβξϑ4εϰPC=ϱγUβ2Iβξ4ϰPβρ2ϰ+βξϑ4εϰPCβξμ4εϰCϱγUβ2Iβξ4ϰPβξμ4εϰCϱσU+I+β2ϰP+βξ4εϰC=ϱσF(t),

where σ=minγ,β2,ξ2,μ. Then,

F(t)ϱσ+F(0)ϱσeσt.

Then, 0 ≤ F(t) ≤ s1, if F(0) ≤ s1 for t ≥ 0 wheres1 = ϱσ . Hence, 0 ≤ U(t), I(t) ≤ s1, 0 ≤ P(t) ≤ s2 and 0 ≤ C(t) ≤ s3 for all t ≥ 0 if U(0)+I(0)+β2ϰP(0)+βξ4εϰC(0)s1, where s2=2ϰs1βands3=4εϰs1βξ. This guarantees that the solutions of the model are bounded.□

The basic infection reproduction number for model (8)-(11) is given by:

R0=ϱωϰβξγ.

Lemma 1

Consider model (8)-(11), we have

  1. if R0 ≤ 1, then the model has only one equilibrium point EP0,

  2. if R0 > 1, then the model has two equilibria EP0 and EP1.

Proof

At any equilibrium EP(U, I, P, C) we have

ϱγUωUP1+αP=0, (13)
ωUP1+αPβI=0, (14)
ϰIξPρPC=0, (15)
εPμCϑCP=0. (16)

From equations (13)-(16) we get an infection-free equilibrium EP0 = (U0, 0, 0, 0), where U0=ϱγ and a unique endemic equilibrium EP1 = (U1, I1, P1, C1), where

U1=ϱ1+αP1γ+αγP1+ωP1,I1=ϱωP1(1+αP1)βγ+αγP1+ωP1,P1=b+b24ac2a,C1=εP1ϑP1+μ,

where

a=βξγαϑ+βξϑω+βγρεα+βρεω,b=βξγϑ+βξγμα+βρεγ+βξμωϱωϑϰ,c=βξγμ1R0.

Then the equilibrium EP1 exists when R0 > 1.□

2.2 Global properties

Define a function G(u) = u – 1 – ln u. Clearly G(u) ≥ 0, for u > 0 and G(1) = 0. The global stability analysis of the two equilibria of model (8)-(11) will be established in the next theorems.

Theorem 1

Let R0 < 1, then the infection-free equilibrium EP0 of model (8-(11) is globally asymptotically stable.

Proof

Construct a Lyapunov function L0(U, I, P, C) as

L0=U0GUU0+I+βϰP+βξεϰ1R0C.

Calculating dL0dt as:

dL0dt=1U0UϱγUωUP1+αP+ωUP1+αPβI+βϰϰIξPρPC+βξεϰ1R0εPμCϑCP=γ1U0UUU0βρϰPCβξϑεϰ1R0PC+ωU01+αPβξϰ+βξϰ(1R0)Pβξμεϰ1R0C=γ(UU0)2Uβρϰ+βξϑεϰ1R0PCβξμεϰ1R0CβξαR0ϰ1+αPP2.

Since R0 < 1, then for all U, P, C > 0 we have dL0dt ≤ 0. Moreover, dL0dt = 0 when U(t) = U0 and P(t) = C(t) = 0. Let D0=(U,I,P,C):dL0dt=0 and M0 be the largest invariant subset of D0. The trajectory of model (8)-(11) tends to M0 [58]. All the elements of M0 satisfyU(t) = U0 and P(t) = C(t) = 0. Then Eq. (10) we get

P˙(t)=0=ϰI(t),I(t)=0.

Hence, M0 = {EP0}. From LaSalle’s invariance principle, we derive that if R0 < 1, then EP0 is globally asymptotically stable.□

Theorem 2

Let R0 > 1, then the endemic equilibrium EP1 of model (8)-(11) is globally asymptotically stable.

Proof

Construct a Lyapunov function L1(U, I, P, C) as

L1=U1GUU1+I1GII1+βϰP1GPP1+βρ2ϰεϑC1CC12.

Note that from the equilibrium condition Eq. (16) that

εϑC1=μC1P1>0.

Then dL1dt is given by:

dL1dt=1U1UϱγUωUP1+αP+1I1IωUP1+αPβI+βϰ1P1PϰIξPρPC+βρϰεϑC1CC1εPμCϑCP=1U1UϱγU+ωU1P1+αPωUP1+αPI1I+βI1βξϰPβρϰPCβP1PI+βξϰP1+βρϰP1C+βρϰεϑC1CC1εPμCϑCP.

From the equilibrium conditions, we have:

ϱ=γU1+ωU1P11+αP1,βI1=ωU1P11+αP1,ϰI1=ξP1+ρP1C1,εP1=μC1+ϑP1C1.

Utilizing the conditions of EP1, we get

dL1dt=1U1UγU1+ωU1P11+αP1γU+ωU1P1+αPωUP1+αPI1I+ωU1P11+αP1βξϰPβρϰPCβP1PI+βξϰP1+βρϰP1C+βρϰεϑC1CC1εPμCϑCPεP1+μC1+ϑC1P1.

Simplifying the result, we obtain

dL1dt=γUU12U+ωU1P11+αP12U1U+ωU1P1+αPωUP1+αPI1IβξϰPP1βρϰPP1CωU1P11+αP1P1IPI1+βρϰεϑC1CC1εPμCϑCPεP1+μC1+ϑC1P1+ϑC1PϑC1P+βρϰPP1C1βρϰPP1C1=γUU12U+ωU1P11+αP12U1UP1IPI1+ωU1P11+αP1P1+αP1P11+αPωU1P11+αP11+αP1UPI11+αPU1P1IβϰPP1ξ+ρC1βρϰPP1CC1+βρεϰεϑC1CC1PP1βρμϰεϑC1CC12βρϑϰεϑC1CC12PβρϑC1ϰεϑC1CC1PP1=γUU12U+ωU1P11+αP13U1UP1IPI11+αP1UPI11+αPU1P1I+ωU1P11+αP1P1+αP1P11+αPωU1P11+αP1PP1βρϰPP1CC1+βρ(εϑC1)ϰ(εϑC1)(CC1)(PP1)βρμ+ϑPϰεϑC1CC12=γUU12U+ωU1P11+αP14U1UP1IPI11+αP1UPI11+αPU1P1I1+αP1+αP1βρμ+ϑPϰεϑC1CC12+ωU1P11+αP1P1+αP1P11+αPPP11+1+αP1+αP1=γUU12U+ωU1P11+αP14U1UP1IPI11+αP1UPI11+αPU1P1I1+αP1+αP1βρμ+ϑPϰεϑC1CC12αωU1PP121+αP1+αP12.

Using geometrical mean (GM) and arithmetical mean (AM) inequality

AMGM, (17)

we get

4U1U+P1IPI1+1+αP1UPI11+αPU1P1I+1+αP1+αP1.

Thus for all U, I, P, C > 0 we have dL1dt ≤ 0. In addition dL1dt = 0 when U = U1, I = I1, P = P1 and C = C1. Let D1=W1(U,I,P,C):dL1dt=0 and M1 be the largest invariant subset of D1. Clearly M1 = {EP1}. Applying LaSalle’s invariance principle we obtain that if R0 > 1, then EP1 is globally asymptotically stable.□

3 Model with general incidence rate

In this section we propose a model with more general incidence rate function Θ(U, P) as:

U˙=ϱγUΘ(U,P), (18)
I˙=Θ(U,P)βI, (19)
P˙=ϰIξPρPC, (20)
C˙=εPμCϑPC, (21)

We need the following Assumptions of the function Θ(U, P):

  1. Θ(U, P) is continuously differentiable, Θ(U, P) > 0, and Θ(0, P) = Θ(U, 0) = 0 for all U > 0 and P > 0,

  2. Θ(U,P)U>0,Θ(U,P)P>0, and Θ(U,0)P>0 for all U > 0 and P > 0,

  3. ddUΘ(U,0)P>0 for all U > 0,

  4. Θ(U,P)P is decreasing with respect to P for all P > 0.

One can show that the set Ω given by Eq. (12) is positively invariant for model (18)-(21).

Lemma 2

Assume that Assumptions (A1)-(A4) are satisfied, then there exists a threshold parameter R0G > 0 such that:

  1. if R0G < 1, then the model has only one equilibrium point EP0; and

  2. if R0G > 1, then the model has two equilibria EP0 and EP1.

Proof

At any equilibrium EP(U, I, P, C) we have

ϱγUΘ(U,P)=0, (22)
Θ(U,P)βI=0, (23)
ϰIξPρPC=0, (24)
εPμCϑCP=0. (25)

From Eq. (25), we have

C=εPμ+ϑP, (26)

and from Eq. (24), we get

I=ξPϰ+ρεϰP2μ+ϑP. (27)

Now from Eqs. (27) and (22)-(23), we obtain

U=ϱγβξγϰP+βρεγϰP2μ+ϑP. (28)

Let

Ψ(P)=ϱγβξγϰPβρεγϰP2μ+ϑP.

Therefore, we can write U as U = Ψ(P). Note that Ψ(0) = ϱγ .

From Eqs. (27) and (22)-(23), we have

Θ(Ψ(P),P)βξPϰ+ρεϰP2μ+ϑP=0. (29)

Observe that, P = 0 is a solution of Eq. (29). Then from Eqs. (26)-(28), we have U = U0, I = 0, and C = 0. Then we get an infection-free equilibrium EP0 = (U0, 0, 0, 0).

Let

H(P)=Θ(Ψ(P),P)βξPϰ+ρεϰP2μ+ϑP,

then H(0) = 0. Let P be such that Ψ(P) = 0, i.e.,

U0βξγϰP¯βρεγϰP¯2μ+ϑP¯=0,

which gives

(βρε+βξϑ)P¯2+(βξμγϰϑU0)P¯γϰμU0=0. (30)

Thus, the positive solution of Eq. (30) is given by

P¯=(γϰϑU0βξμ)+(βξμγϰϑU0)2+4γϰμU0(βρε+βξϑ)2(βρε+βξϑ).

We can see that

H(P¯)=Θ(0,P¯)βξP¯ϰ+ρεϰP¯2μ+ϑP¯=βξP¯ϰ+ρεϰP¯2μ+ϑP¯<0.

Moreover,

H(P)=Θ(U,P)P+Ψ(P)Θ(U,P)Uβξϰβρε2μP+ϑP2ϰ(μ+ϑP)2.

Assumption (A1) implies that Θ(U0,0)U=0, then

H(0)=Θ(U0,0)Pβξϰ=βξϰϰβξΘ(U0,0)P1.

Therefore, if ϰβξΘ(U0,0)P>1, then H(0) > 0 and ∃P1 ∈ (0, P) such that H(P1) = 0. Let us define

R0G=ϰβξΘ(U0,0)P,

which represents the basic reproduction number. Now, let

g(U)=ϱγUΘ(U,P1)=0.

Then we have g(0) = ϱ > 0 and g(U0) = –Θ(U0, P1) < 0. Assumption (A2) implies that g(U) is a strictly decreasing function of U, and then there exists a unique U1 ∈ (0, U0) such that g(U1) = 0. Moreover, from Eqs. (26) and (27), we have

C1=εP1μ+ϑP1>0,I1=ξP1ϰ+ρεϰP12μ+ϑP1>0.

Therefore, an endemic equilibrium EP1 = (U1, I1, P1, C1) exists if R0G > 1.□

3.1 Global stability of equilibria

The global stability analysis of the two equilibria of model (18)-(21) will be investigated in this section.

Theorem 3

Let R0G > 1, then the infection-free equilibrium EP0 of model (18)-(21) is globally asymptotically stable.

Proof

Construct a Lyapunov function Z0(U, I, P, C) as

Z0=UU0U0UlimP0+Θ(U0,P)Θ(η,P)dη+I+βϰP+βξεϰ1R0GC.

Calculating dZ0dt as:

dZ0dt=1limP0+Θ(U0,P)Θ(U,P)ϱγUΘ(U,P)+Θ(U,P)βI+βϰϰIξPρPC+βξεϰ1R0GεPμCϑPC=1limP0+Θ(U0,P)Θ(U,P)ϱγU+Θ(U,P)limP0+Θ(U0,P)Θ(U,P)βξϰPβρϰPC+βξϰ1R0GPβξμεϰ1R0GCβξϑεϰ1R0GPC=ϱγU1limP0+Θ(U0,P)Θ(U,P)+Θ(U,P)limP0+Θ(U0,P)Θ(U,P)βξR0GϰPβξμεϰ1R0GCβρϰ+βξϑεϰ1R0GPC=γU01UU01Θ(U0,0)/PΘ(U,0)/P+βξR0GϰϰΘ(U,P)βξR0GPΘ(U0,0)/PΘ(U,0)/P1Pβξμεϰ1R0GCβρϰ+βξϑεϰ1R0GPC.

From the Assumptions, we have the first term is less than or equal to zero. In addition,

Θ(U,P)PlimP0+Θ(U,P)P=Θ(U,0)P,

for all U > 0. Then

ϰΘ(U,P)βξR0GPΘ(U0,0)/PΘ(U,0)/PϰβξR0GΘ(U0,0)P=1.

It implies that

dZ0dtγU01UU01Θ(U0,0)/PΘ(U,0)/Pβξμεϰ1R0GCβρϰ+βξϑεϰ1R0GPC.

Therefore, if R0G < 1, then dZ0dt0 for all U, P, C > 0. Similar to the proof of Theorem 1, one can show that EP0 is globally asymptotically stable.□

Theorem 4

Let R0G > 1, then the endemic equilibrium EP1 of model (18)-(21) is globally asymptotically stable.

Proof

Define Z1(U, I, P, C) as

Z1=UU1U1UΘ(U1,P1)Θ(η,P1)dη+I1GII1+βϰP1GPP1+βρ2ϰεϑC1CC12.

Then dZ1dt can be calculated as:

dZ1dt=1Θ(U1,P1)Θ(U,P1)ϱγUΘ(U,P)+1I1IΘ(U,P)βI+βϰ1P1PϰIξPρPC+βρϰεϑC1CC1εPμCϑPC.

Using the equilibrium condition, εP1μC1ϑP1C1 = 0, we have

dZ1dt=1Θ(U1,P1)Θ(U,P1)ϱγU+Θ(U,P)Θ(U1,P1)Θ(U,P1)I1IΘ(U,P)+βI1βξϰPβρϰPCβP1PI+βξϰP1+βρϰP1C+βρϰεϑC1CC1εPμCϑPCεP1+μC1+ϑP1C1+ϑPC1ϑPC1+βρϰPP1C1βρϰPP1C1=1Θ(U1,P1)Θ(U,P1)ϱγU+Θ(U,P)Θ(U1,P1)Θ(U,P1)I1IΘ(U,P)+βI1βξϰPβρϰPCβP1PI+βξϰP1+βρϰP1C+βρεϰεϑC1CC1PP1βρμϰεϑC1CC12βρϑPϰεϑC1CC12βρϑC1ϰεϑC1CC1PP1+βρϰPP1C1βρϰPP1C1.

From the equilibrium conditions, we have

βI1=Θ(U1,P1),ϱ=γU1+βI1.

Applying these conditions, we obtain

dZ1dt=1Θ(U1,P1)Θ(U,P1)γU1+βI1γU+βI1Θ(U,P)Θ(U,P1)I1IΘ(U,P)+βI1βξϰPP1βρϰPP1CβP1PI+βρ(εϑC1)ϰεϑC1CC1PP1βρ(μ+ϑP)ϰεϑC1CC12+βρϰPP1C1βρϰPP1C1=1Θ(U1,P1)Θ(U,P1)γU1γU+βI12Θ(U1,P1)Θ(U,P1)+βI1Θ(U,P)Θ(U,P1)βI1I1Θ(U,P)IΘ(U1,P1)βϰPP1ξ+ρC1βρϰPP1CC1βI1P1IPI1+βρϰCC1PP1βρ(μ+ϑP)ϰεϑC1CC12=1Θ(U1,P1)Θ(U,P1)γU1γU+βI13Θ(U1,P1)Θ(U,P1)I1Θ(U,P)IΘ(U1,P1)P1IPI1+βI1Θ(U,P)Θ(U,P1)PP1βρμ+ϑPϰεϑC1CC12=1Θ(U1,P1)Θ(U,P1)γU1γU+βI14Θ(U1,P1)Θ(U,P1)I1Θ(U,P)IΘ(U1,P1)P1IPI1PΘ(U,P1)P1Θ(U,P)+βI1Θ(U,P)Θ(U,P1)PP11+PΘ(U,P1)P1Θ(U,P)βρμ+ϑPϰεϑC1CC12=γU11UU11Θ(U1,P1)Θ(U,P1)+βI14Θ(U1,P1)Θ(U,P1)I1Θ(U,P)IΘ(U1,P1)P1IPI1PΘ(U,P1)P1Θ(U,P)+βI1Θ(U,P)Θ(U,P1)PP11Θ(U,P1)Θ(U,P)βρμ+ϑPϰεϑC1CC12.

From Assumptions (A2) and (A4) we have

1UU11Θ(U1,P1)Θ(U,P1)0,Θ(U,P)Θ(U,P1)PP11Θ(U,P1)Θ(U,P)0.

Therefore, using inequality (17) we get that for all U, I, P, C > 0 we have dZ1dt ≤ 0 and dZ1dt = 0 if and only if U = U1, I = I1, P = P1 and C = C1. Applying LaSalle’s invariance principle, we obtain that if R0G > 1, then EP1 is globally asymptotically stable.□

4 Numerical simulations

We conduct numerical simulations for model (18)-(21) with specific incidence rate function

Θ(U,P)=ωUP1+α1P+α2U.

Then we get following model with Beddington-DeAngelis functional response:

U˙=ϱγUωUP1+α1P+α2U, (31)
I˙=ωUP1+α1P+α2UβI, (32)
P˙=ϰIξPρPC, (33)
C˙=εPμCϑPC, (34)

where ω is a positive parameter, while α1 and α2 are non-negative parameters. We note that if α1 = α2 = 0, then we obtain a model with bilinear incidence, if α1 ≠ 0 and α2 = 0, then we get saturated incidence which given in model (8)-(11), and if α1 = 0 and α2 ≠ 0, then we obtain Holling type-II. We can easily see that Θ(U, P) is continuously differentiable function. Moreover, Θ(U, P) satisfying the following conditions:

We have

Θ(U,P)U=ωP+α1ωP2(1+α1P+α2U)2,Θ(U,P)P=ωU+α2ωU2(1+α1P+α2U)2,

then Θ(U, P) is continuously differentiable. Moreover, Θ(U, P) > 0, and Θ(0, P) = Θ(U, 0) = 0 for all U > 0 and P > 0. Thus (A1) is satisfied.

Since Θ(U,P)U>0,Θ(U,P)P>0, and Θ(U,0)P=ωU1+α2U>0 for all U > 0, then (A2) is satisfied.

We have

ddUΘ(U,0)P=ω(1+α2U)2>0 for all U0,

then (A3) is satisfied.

Finally we have

PΘ(U,P)P=α1ωU(1+α1P+α2U)2<0, for all P0,

then (A4) is also satisfied.

The basic reproduction number of model (31)-(34) is given by

R0G=ϰωU0βξ(1+α2U0).

In the numerical simulations we fix the values of parameters ϱ = 10, ϰ = ξ = 3, ρ = 0.1, μ = γ = 0.01, β = 0.3, ε = 0.2 and vary ω and ϑ

Case(1): Effect of ω on the stability of equilibria.

For this case, we take α1 = α2 = 0 and ϑ = 0.01. We choose three different initial conditions as:

IC1: U(0) = 700, I(0) = 5, P(0) = 5, C(0) = 0.5,

IC2: U(0) = 400, I(0) = 10, P(0) = 10, C(0) = 1,

IC3: U(0) = 300, I(0) = 20, P(0) = 15, C(0) = 1.5.

We consider two values of the parameter ω as:

  1. ω = 0.0001, then we compute R0G = 0.3333 < 1. Figure 1 shows that, for all IC1-IC3, the solution of the model tends to EP0 = (1000, 0, 0, 0). It means that, EP0 is globally asymptotically stable.

    Figure 1 
Solution trajectories of system (31)-(34) in case α1 = α2 = 0.
    Figure 1

    Solution trajectories of system (31)-(34) in case α1 = α2 = 0.

  2. ω = 0.001, then we compute R0G = 3.3333 > 1. Figure 1, shows that the solutions of the model converge to the equilibrium EP1 = (482.9, 17.23, 10.7, 18.29) for all IC1-IC3. Then, EP1 is globally asymptotically stable.

Case(2): Effect of the saturation infection on the virus dynamics.

In this case, we take α2 = 0 and ϑ = 0.01. We choose ω = 0.001, and α1 varied. Moreover we consider the initial condition IC2. Figure 2 shows that as α1 is increased, the concentrations of the uninfected target cells is increased, while the the concentration of infected cells, virus particles and B cells are decreased. We note that the parameter α1 has no effect on the stability of equilibria

Figure 2 
Solution trajectories of system (31)-(34) for different values of α1 when α2 = 0.
Figure 2

Solution trajectories of system (31)-(34) for different values of α1 when α2 = 0.

Case(3): Effect of Holling type-II.

For this case, we take α1 = 0, ω = 0.001, and ϑ = 0.01 then Θ(U, P) represents the Holling type-II. Let us choose the initial condition IC2. We suggest different values of α2 to see its effect on the model as we can see in Figure 3. Moreover, we have the following cases:

Figure 3 
Solution trajectories of system (31)-(34) for different values of α2 when α1 = 0.
Figure 3

Solution trajectories of system (31)-(34) for different values of α2 when α1 = 0.

  1. EP1 is globally asymptotically stable when 0 ≤ α2 < 0.0023,

  2. EP0 is globally asymptotically stable when α2 > 0.0023.

This means that α2 can play the role of controller which can be designed to stabilize the system around the infection-free equilibrium EP0.

Case(4): Effect of the B cell impairment parameter ϑ.

In this case, we take α1 = 0.01 and α2 = 0.002. We choose ω = 0.001, and ϑ varied. Moreover, we consider the following initial condition

IC4: U(0) = 700, I(0) = 10, P(0) = 10, C(0) = 10.

Figure 4 shows that as ϑ is increased, the concentrations of infected cells and virus particles are increased, while the concentration of uninfected cells is decreased. We note that the parameter ϑ has no effect on the stability of equilibria.

Figure 4 
Solution trajectories of system (31)-(34) for different values of ϑ.
Figure 4

Solution trajectories of system (31)-(34) for different values of ϑ.

Acknowledgement

This project was funded by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah, Saudi Arabia under grant no. KEP-MSc-33-130-40. The authors, therefore, acknowledge with thanks DSR technical and financial support.

References

[1] Nowak M., Bangham C., Population dynamics of immune responses to persistent viruses, Science, 1996, 272, 74-79.10.1126/science.272.5258.74Search in Google Scholar PubMed

[2] Nowak M., May R., Virus dynamics: mathematical principles of immunology and virology, Oxford university press, 2000.Search in Google Scholar

[3] Perelson A., Nelson P., Mathematical analysis of HIV-1 dynamics in vivo, SIAM Rev., 1999, 41, 3-44.10.1137/S0036144598335107Search in Google Scholar

[4] Callaway D., Perelson A., HIV-1 infection and low steady state viral loads, B. Math. Biol., 2002, 64, 29-64.10.1006/bulm.2001.0266Search in Google Scholar PubMed

[5] Roy P., Chatterjee A., Greenhalgh D., Khan Q., Long term dynamics in a mathematical model of HIV-1 infection with delay in different variants of the basic drug therapy model, Nonlinear Anal-Real., 2013, 14, 1621-1633.10.1016/j.nonrwa.2012.10.021Search in Google Scholar

[6] Leenheer P., Smith H., Virus dynamics: A global analysis, SIAM J. Appl. Math., 2003, 63, 1313-1327.10.1137/S0036139902406905Search in Google Scholar

[7] Song X., Neumann A., Global stability and periodic solution of the viral dynamics, J. Math. Anal. Appl., 2007, 329, 281-297.10.1016/j.jmaa.2006.06.064Search in Google Scholar

[8] Huang G., Takeuchi Y., Ma W., Lyapunov functionals for delay differential equations model of viral infections, SIAM J. Appl. Math., 2010, 70, 2693-2708.10.1137/090780821Search in Google Scholar

[9] Georgescu P., Hsieh Y., Global stability for a virus dynamics model with nonlinear incidence of infection and removal, SIAM J. Appl. Math., 2006, 67, 337-353.10.1137/060654876Search in Google Scholar

[10] Li M., Shu H., Global Dynamics of an in-host viral model with intracellular delay, B. Math. Biol., 2010, 72, 1492-1505.10.1007/s11538-010-9503-xSearch in Google Scholar PubMed

[11] Li M., Shu H., Impact of intracellular delays and target-cell dynamics on in vivo viral infections, SIAM J. Appl. Math., 2010, 70, 434-2448.10.1137/090779322Search in Google Scholar

[12] Hattaf K., Yousfi N., Tridane A., Mathematical analysis of a virus dynamics model with general incidence rate and cure rate, Nonlinear Anal-Real., 2012, 13, 1866-1872.10.1016/j.nonrwa.2011.12.015Search in Google Scholar

[13] Korobeinikov A., Global properties of basic virus dynamics models, B. Math. Biol., 2004, 66, 879-883.10.1016/j.bulm.2004.02.001Search in Google Scholar PubMed

[14] Elaiw A., Global properties of a class of HIV models, Nonlinear Anal-Real., 2010, 11, 2253-2263.10.1016/j.nonrwa.2009.07.001Search in Google Scholar

[15] Elaiw A., Global properties of a class of virus infection models with multitarget cells, Nonlinear Dynam., 2012, 69, 423-435.10.1007/s11071-011-0275-0Search in Google Scholar

[16] Elaiw A., Almuallem N., Global properties of delayed-HIV dynamics models with differential drug efficacy in cocirculating target cells, Appl. Math. Comput., 2015, 265, 1067-1089.10.1016/j.amc.2015.06.011Search in Google Scholar

[17] Elaiw A., Almuallem N., Global dynamics of delay-distributed HIV infection models with differential drug efficacy in cocirculating target cells, Math. Method. Appl. Sci., 2016, 39, 4-31.10.1002/mma.3453Search in Google Scholar

[18] Elaiw A., Hassanien I., Azoz S., Global stability of HIV infection models with intracellular delays, J. Korean Math. Soc., 2012, 49, 779-794.10.4134/JKMS.2012.49.4.779Search in Google Scholar

[19] Elaiw A., Azoz S., Global properties of a class of HIV infection models with Beddington-DeAngelis functional response, Math. Method. Appl. Sci., 2013, 36, 383-394.10.1002/mma.2596Search in Google Scholar

[20] Elaiw A., Global dynamics of an HIV infection model with two classes of target cells and distributed delays, Discrete Dyn. Nat. Soc., 2012, Article ID 253703.10.1155/2012/253703Search in Google Scholar

[21] Gibelli L., Elaiw A., Alghamdi M., Althiabi A., Heterogeneous population dynamics of active particles: Progression, mutations, and selection dynamics, Math. Method. Appl. Sci., 2017, 27, 617-640.10.1142/S0218202517500117Search in Google Scholar

[22] Bellomo N., Tao Y., Stabilization in a chemotaxis model for virus infection, Discrete Cont. Dyn-S., 2020, 13, 105-117.10.3934/dcdss.2020006Search in Google Scholar

[23] Lydyard P., Whelan A., Fanger M., Instant notes in immunology, Taylor & Francis e-Library, 2005.10.4324/9780203488287Search in Google Scholar

[24] Murase A., Sasaki T., Kajiwara T., Stability analysis of pathogen-immune interaction dynamics, J. Math. Biol., 2005, 51, 247-267.10.1007/s00285-005-0321-ySearch in Google Scholar PubMed

[25] Wang T., Hu Z., Liao F., Stability and Hopf bifurcation for a virus infection model with delayed humoral immunity response, J. Math. Anal. Appl., 2014, 411, 63-74.10.1016/j.jmaa.2013.09.035Search in Google Scholar

[26] Elaiw A., AlShamrani N., Global stability of humoral immunity virus dynamics models with nonlinear infection rate and removal, Nonlinear Anal-Real., 2015, 26, 161-190.10.1016/j.nonrwa.2015.05.007Search in Google Scholar

[27] Wang T., Hu Z., Liao F., Ma W., Global stability analysis for delayed virus infection model with general incidence rate and humoral immunity, Math. Comput. Simulat., 2013, 89, 13-22.10.1016/j.matcom.2013.03.004Search in Google Scholar

[28] Elaiw A., AlShamrani N., Stability of a general delay-distributed virus dynamics model with multi-staged infected progression and immune response, Math. Method. Appl. Sci., 2017, 40, 699-719.10.1002/mma.4002Search in Google Scholar

[29] Elaiw A., Alshaikh M., Stability analysis of a general discrete-time pathogen infection model with humoral immunity, J. Differ. Equ. Appl., 2019, 10.1080/10236198.2019.1662411.Search in Google Scholar

[30] Elaiw A., AlShamrani N., Stability of an adaptive immunity pathogen dynamics model with latency and multiple delays, Math. Method. Appl. Sci., 2018, 36, 125-142.10.1002/mma.5182Search in Google Scholar

[31] Elaiw A., AlShamrani N., Stability of a general adaptive immunity virus dynamics model with multi-stages of infected cells and two routes of infection, Math. Method. Appl. Sci., 2019, 10.1002/mma.5923.Search in Google Scholar

[32] Elaiw A., Elnahary E., Analysis of general humoral immunity HIV dynamics model with HAART and distributed delays, Mathematics, 2019, 7, Article number: 157.10.3390/math7020157Search in Google Scholar

[33] Elaiw A., Almatrafi A., Hobiny A., Hattaf K., Global properties of a general latent pathogen dynamics model with delayed pathogenic and cellular infections, Discrete Dyn. Nat. Soc., 2019, Article ID 9585497.10.1155/2019/9585497Search in Google Scholar

[34] Lin J., Xu R., Tian X, Threshold dynamics of an HIV-1 virus model with both virus-to-cell and cell-to-cell transmissions, intracellular delay, and humoral immunity, Appl. Math. Comput., 2017, 315, 516-530.10.1016/j.amc.2017.08.004Search in Google Scholar

[35] Pan S., Chakrabarty S., Threshold dynamics of HCV model with cell-to-cell transmission and a non-cytolytic cure in the presence of humoral immunity, Commun. Nonlinear Sci., 2018, 61, 180-197.10.1016/j.cnsns.2018.02.010Search in Google Scholar

[36] Miao H., Teng Z., Kang C., Muhammadhaji A., Stability analysis of a virus infection model with humoral immunity response and two time delays, Math. Method. Appl. Sci., 2016, 39, 3434-3449.10.1002/mma.3790Search in Google Scholar

[37] Hu Z., Zhang J., Wang H., Ma W., Liao F., Dynamics analysis of a delayed viral infection model with logistic growth and immune impairment, Appl. Math. Model., 2014, 38, 524-534.10.1016/j.apm.2013.06.041Search in Google Scholar

[38] Regoes R., Wodarz D., Nowak M., Virus dynamics: the effect to target cell limitation and immune responses on virus evolution, J. Theor. Biol., 1998, 191, 451-462.10.1006/jtbi.1997.0617Search in Google Scholar PubMed

[39] Wang S., Song X., Ge Z., Dynamics analysis of a delayed viral infection model with immune impairment, Appl. Math. Model., 2011, 35, 4877-4885.10.1016/j.apm.2011.03.043Search in Google Scholar

[40] Elaiw A., Raezah A., Azoz S., Stability of delayed HIV dynamics models with two latent reservoirs and immune impairment, Adv. Differ. Equ-Ny., 2018, Article number: 414.10.1186/s13662-018-1869-3Search in Google Scholar

[41] Milito A., B Lymphocyte Dysfunctions in HIV Infection, Curr. HIV Res., 2004, 2, 11-21.10.2174/1570162043485068Search in Google Scholar PubMed

[42] Amu S., Ruffin N., Rethi B., Chiodi F., Impairment of B-cell functions during HIV-1 infection, AIDS, 2013, 27, 2323-2334.10.1097/QAD.0b013e328361a427Search in Google Scholar PubMed

[43] Chiodi F., Scarlatti G., Editorial: HIV-Induced damage of B cells and production of HIV neutralizing antibodies, Front. immunol., 2018, 9, Article number: 297.10.3389/fimmu.2018.00297Search in Google Scholar PubMed PubMed Central

[44] Miao H., Abdurahman X., Teng Z., Zhang L., Dynamical analysis of a delayed reaction-diffusion virus infection model with logistic growth and humoral immune impairment, Chaos Soliton. Fract., 2018, 110, 280-291.10.1016/j.chaos.2018.03.006Search in Google Scholar

[45] Wang W., Ma W., Feng Z., Dynamics of reaction-diffusion equations for modeling CD4+ T cells decline with general infection mechanism and distinct dispersal rates, Nonlinear Anal-Real., 2020, 51, Article number: 102976.10.1016/j.nonrwa.2019.102976Search in Google Scholar

[46] Wang W., Ma W., Feng Z., Complex dynamics of a time periodic nonlocal and time-delayed model of reaction-diffusion equations for modeling CD4+ T cells decline, J. Comput. Appl. Math., 2020, 367, Article number: 112430.10.1016/j.cam.2019.112430Search in Google Scholar

[47] Elaiw A., Abukwaik R., Alzahrani E., Global properties of a cell mediated immunity in HIV infection model with two classes of target cells and distributed delays, Int. J. Biomath., 2014, 7, Article ID 1450055.10.1142/S1793524514500557Search in Google Scholar

[48] Elaiw A., AlShamrani N., Global properties of nonlinear humoral immunity viral infection models, Int. J. Biomath., 2015, 8, Article ID 1550058.10.1142/S1793524515500588Search in Google Scholar

[49] Elaiw A., Alshehaiween S., Hobiny A., Global properties of delay-distributed HIV dynamics model including impairment of B-cell functions, Mathematics, 2019, 7, Article number: 837.10.3390/math7090837Search in Google Scholar

[50] Elaiw A., Alshaikh M., Stability of discrete-time HIV dynamics models with three categories of infected CD4+ T-cells, Adv. Differ. Equ-Ny., 2019, Article number: 407.10.1186/s13662-019-2338-3Search in Google Scholar

[51] Elaiw A., Raezah A., Stability of general virus dynamics models with both cellular and viral infections and delays, Math. Method. Appl. Sci., 2017, 40, 5863-5880.10.1002/mma.4436Search in Google Scholar

[52] Hobiny A., Elaiw A., Almatrafi A., Stability of delayed pathogen dynamics models with latency and two routes of infection, Adv. Differ. Equ-Ny., 2018, Article number: 276.10.1186/s13662-018-1720-xSearch in Google Scholar

[53] Elaiw A., Elnahary E., Raezah A., Effect of cellular reservoirs and delays on the global dynamics of HIV, Adv. Differ. Equ-Ny., 2018, Article number: 85.10.1186/s13662-018-1523-0Search in Google Scholar

[54] Elaiw A., AlAgha A., Global dynamics of reaction-diffusion oncolytic M1 virotherapy with immune response, Appl. Math. Comput., 2019 (accepted).10.1016/j.amc.2019.124758Search in Google Scholar

[55] Gao N., Song Y., Wang X., Liu J., Dynamics of a stochastic SIS epidemic model with nonlinear incidence rates, Adv. Differ. Equ-Ny., 2019, article number: 41.10.1186/s13662-019-1980-0Search in Google Scholar

[56] Song Y., Miao A., Zhang T., Wang X., Liu J., Extinction and persistence of a stochastic SIRS epidemic model with saturated incidence rate and transfer from infectious to susceptible, Adv. Differ. Equ-Ny., 2018, Article number: 293.10.1186/s13662-018-1759-8Search in Google Scholar

[57] Wang Y., Cao J., Alofi A., AL-Mazrooei A., Elaiw A., Revisiting node-based SIR models in complex networks with degree correlations, Physica A., 2015, 437, 75-88.10.1016/j.physa.2015.05.103Search in Google Scholar

[58] Hale J., Verduyn Lunel S., Introduction to functional differential equations, Springer Science & Business Media, 2013.10.1007/978-1-4612-4342-7Search in Google Scholar

Received: 2019-04-07
Accepted: 2019-10-04
Published Online: 2019-12-13

© 2019 Elaiw et al., published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 Public License.

Downloaded on 28.4.2024 from https://www.degruyter.com/document/doi/10.1515/math-2019-0113/html
Scroll to top button