Skip to main content

Theory and Modern Applications

Stability of delayed pathogen dynamics models with latency and two routes of infection

Abstract

We consider a pathogen dynamics model with antibodies and both pathogen-to-susceptible and infected-to-susceptible transmissions. We consider two types of infected cells, latently infected cells, and actively infected cells. The model considers three types of discrete or distributed delays to characterize the time between the pathogen or the infected cell contacts a susceptible cell and the creation of mature pathogens. We deduct the basic reproduction number and antibody response activation number which determine the existence and stability of the steady states. The global stability analysis of the steady states is established using Lyapunov method. The theoretical results are confirmed by numerical simulations.

1 Introduction

Modeling and analysis of within-host human pathogen dynamics have been studied in several works (see, e.g., [123]). These works can help researchers to better understand the pathogen dynamical behavior and to provide new suggestions for clinical treatment. A vast amount of the mathematical models presented in the literature focused on modeling the interaction between three main compartments, susceptible cells (s), infected cells (y), and pathogens (p). B cell is one of the central components of the immune system against viral infections. The B cells create antibodies to neutralize the pathogens. Murase et al. [24] considered the effect of antibodies (x) on the pathogen infection model as follows:

$$\begin{aligned} \begin{aligned} &\dot{s}(t) =\omega -ds(t)-\pi_{1}s(t)p(t), \\ &\dot{y}(t) =\pi_{1}s(t)p(t)-\lambda y(t), \\ &\dot{p}(t) =n\lambda y(t)-cp(t)-ap(t)x(t), \\ &\dot{x}(t) =rp(t)x(t)-mx(t). \end{aligned} \end{aligned}$$
(1)

The susceptible cells are produced at rate ω, die at rate ds, and become infected at rate \(\pi_{1}sp\), where \(\pi_{1}\) is the pathogen-susceptible incidence rate constant. λ is the death rate constant of the infected cells, a is the neutralization rate constant of the pathogens, and c is the death rate constant of the pathogens. The infected cell releases a number n of pathogens during its lifespan. The B cells are proliferated and die at rates \(rpx\) and mx, respectively, where r and m are constants. The effect of antibody immune response on the pathogen dynamics has been studied in several works (see, e.g., [2429]). In these works, it was assumed that the susceptible cells become infected by contacting with pathogens (pathogen-to-susceptible transmission). In [3033], it was reported that the pathogens can also spread by infected-to-susceptible transmission. However, in these works the antibody immune response was neglected. In very recent works [34, 35], and [36], both pathogen-to-susceptible and infected-to-susceptible transmissions were incorporated in the pathogen dynamics models with antibody immune response. However, the latently infected cells were neglected in these models.

It is worth stressing that the introduction of delay equations has been widely applied to model complex systems in biology. Indeed, the introduction of delay terms can be viewed as a first step towards modeling multiscale dynamics and heterogeneity features in population dynamics [37].

In the present paper we investigate the global stability of pathogen dynamics models with antibodies and both pathogen-to-susceptible and infected-to-susceptible transmissions. We consider both latently infected cells and actively infected cells. We incorporate three types of discrete or distributed time delays to describe the time between the pathogen or the actively infected cell contacts a susceptible cell and the emission of new mature pathogens. We calculate two bifurcation parameters \(\mathcal{R}_{0}\) (the basic reproduction number) and \(\mathcal{R}_{1}\) (the antibody response activation number) which determine the existence and global stability of all steady states. Numerical simulations are performed to confirm the theoretical results.

2 Model with discrete-time delays

We investigate the following pathogen dynamics model with discrete-time delays:

$$\begin{aligned} \begin{aligned} &\dot{s}(t) =\omega -ds(t)-s(t) \bigl[ \pi_{1}p(t)+\pi_{2}y(t) \bigr] , \\ &\dot{u}(t) =\rho e^{-\varepsilon_{1}\tau_{1}}s(t-\tau_{1})\bigl[ \pi_{1}p(t- \tau_{1})+\pi_{2}y(t- \tau_{1})\bigr]-(\alpha +\lambda_{u})u(t), \\ &\dot{y}(t) =(1-\rho)e^{-\varepsilon_{2}\tau_{2}}s(t-\tau_{2})\bigl[ \pi_{1}p(t- \tau_{2})+\pi_{2}y(t- \tau_{2})\bigr]-\lambda y(t)+\alpha u(t), \\ &\dot{p}(t) =n\lambda e^{-\varepsilon_{3}\tau_{3}}y(t-\tau_{3})-cp(t)-ap(t)x(t), \\ &\dot{x}(t) =rp(t)x(t)-mx(t). \end{aligned} \end{aligned}$$
(2)

The model assumes that the susceptible cells are infected by pathogens at rate \(\pi_{1}s(t)p(t)\) and by infected cells at rate \(\pi_{2}s(t)y(t)\). The fractions ρ and \((1-\rho)\) with \(0<\rho <1\) are the proportions of infection that lead to latency and activation, respectively. \(\lambda_{u}\) is the death rate constant of the latently infected cells. Latently infected cells are activated at rate \(\alpha u(t)\). Here, \(\tau_{1}\) is the time between pathogen entry a susceptible cell to become latent infected, and \(\tau_{2}\) is the time between pathogen entry a susceptible cell and the production of immature pathogens. The immature pathogens need time \(\tau_{3}\) to be mature. The factors \(e^{-\varepsilon_{1}\tau_{1}}\), \(e^{-\varepsilon_{2}\tau_{2}}\), and \(e^{-\varepsilon_{3}\tau_{3}}\) represent the probability of surviving to the age of \(\tau_{1}\), \(\tau_{2}\), and \(\tau_{3}\), respectively, where \(\varepsilon_{1}\), \(\varepsilon_{2}\), and, \(\varepsilon_{3}\) are positive constants.

We consider the initial conditions

$$\begin{aligned} \begin{aligned} & s(\theta) =\phi_{1}(\theta),\qquad u( \theta)=\phi_{2}(\theta),\qquad y(\theta)=\phi_{3}(\theta), \\ &p(\theta) =\phi_{4}(\theta),\qquad x(\theta)=\phi_{5}( \theta), \\ &\phi_{j}(\theta) \geq 0,\quad \theta \in {}[ -\kappa,0], \\ &\phi_{j} \in C \bigl( [-\kappa,0],\mathbb{R}_{\mathbb{\geq }0} \bigr), \quad j=1,\ldots,5, \end{aligned} \end{aligned}$$
(3)

where \(\kappa =\max \{ \tau_{1},\tau_{2},\tau_{3}\}\) and C is the Banach space of continuous functions mapping the interval \([-\kappa,0] \) into \(\mathbb{R}_{\mathbb{\geq }0}\) with norm \(\Vert \phi_{j} \Vert =\sup_{-\kappa \leq \theta \leq 0} \vert \phi _{j}(\theta) \vert \). Then system (2) has a unique solution for \(t>0\) [38].

2.1 Properties of solution

Lemma 1

The solutions of system (2) with initial conditions (3) are nonnegative and ultimately bounded for \(t>0\).

Proof

We have from Eq. (2)1 that \(\dot{s}|_{s=0}= \omega >0\). Therefore \(s(t)>0\) for all \(t\geq 0\). Moreover, for \(t\in {}[ 0,\kappa ]\), we have

$$\begin{aligned} &u(t) =\phi_{2}(0)e^{-(\alpha +\lambda_{u})t}+ \int_{0}^{t}{e^{-( \alpha +\lambda_{u})(t-\theta)}} \bigl\{ {\rho e^{-\varepsilon_{1} \tau_{1}}s(\theta -\tau_{1}) \bigl[ \pi_{1}p(\theta - \tau_{1})+\pi_{2}y( \theta -\tau_{1}) \bigr] } \bigr\} \,{d}\theta \\ &\hphantom{u(t)}\geq 0, \\ &y(t) =\phi_{3}(0)e^{-\lambda t} \\ &\hphantom{y(t)=}{}+ \int_{0}^{t}{e^{-\lambda (t-\theta)}} \bigl\{ {(1- \rho)e^{-\varepsilon_{2}\tau_{2}}s(\theta -\tau_{2}) \bigl[ \pi_{1}p( \theta -\tau_{2})+\pi_{2}y(\theta -\tau_{2}) \bigr] + \alpha u(\theta)} \bigr\} \,{d}\theta \geq 0, \\ &p(t) =\phi_{4}(0)e^{-\int_{0}^{t}(c+ax(\upsilon))\,d\upsilon }+ {n\lambda e^{-\varepsilon_{3}\tau_{3}}} \int_{0}^{t}{{e^{-\int_{{\theta }} ^{t}(c+ax(\upsilon))\,d\upsilon }}y(\theta - \tau_{3})\,d}\theta \geq 0, \\ &x(t) =\phi_{5}(0)e^{-mt+r\int_{0}^{t}p(\upsilon)\,d\upsilon }\geq 0. \end{aligned}$$

By recursive argument we get \(u(t)\geq 0\), \(y(t)\geq 0\) and \(p(t)\geq 0\) \(\forall t\geq 0\). The nonnegativity of the model’s solutions implies that \(\dot{s}(t)\leq \omega -ds(t)\) and then \(\lim_{t\rightarrow \infty }\sup s(t)\leq \frac{\omega }{d}\). Let us define \(X_{1}(t)=\rho e^{-\varepsilon_{1}\tau_{1}}s(t-\tau_{1})+(1-\rho)e ^{-\varepsilon_{2}\tau_{2}}s(t-\tau_{2})+u(t)+y(t)\). Then

$$\begin{aligned} \dot{X_{1}}(t) =&\rho e^{-\varepsilon_{1}\tau_{1}} \bigl\{ \omega -ds(t- \tau_{1})-\pi_{1}s(t-\tau_{1})p(t- \tau_{1})-\pi_{2}s(t-\tau_{1})y(t- \tau_{1}) \bigr\} \\ &{} +(1-\rho)e^{-\varepsilon_{2}\tau_{2}} \bigl\{ \omega -ds(t-\tau _{2})- \pi_{1}s(t-\tau_{2})p(t-\tau_{2})- \pi_{2}s(t-\tau_{2})y(t-\tau _{2}) \bigr\} \\ &{} +\rho e^{-\varepsilon_{1}\tau_{1}}s(t-\tau_{1}) \bigl[ \pi_{1}p(t- \tau_{1})+\pi_{2}y(t-\tau_{1}) \bigr] -(\alpha + \lambda_{u})u(t) \\ &{} +(1-\rho)e^{-\varepsilon_{2}\tau_{2}}s(t-\tau_{2})\bigl[\pi_{1}p(t- \tau _{2})+\pi_{2}y(t-\tau_{2})\bigr]-\lambda y(t)+\alpha u(t) \\ =&\omega \rho e^{-\varepsilon_{1}\tau_{1}}+\omega (1-\rho)e^{- \varepsilon_{2}\tau_{2}} \\ &{}-\rho e^{-\varepsilon_{1}\tau_{1}}\,ds(t-\tau _{1})-(1-\rho)e^{-\varepsilon_{2}\tau_{2}} \,ds(t-\tau_{2})-\lambda_{u}u(t)- \lambda y(t) \\ \leq& \omega -\sigma_{1} \bigl[ \rho e^{-\varepsilon_{1}\tau_{1}}s(t- \tau_{1})+(1-\rho)e^{-\varepsilon_{2}\tau_{2}}s(t-\tau_{2})+u(t)+y(t) \bigr] \\ =&\omega -\sigma_{1}X_{1}(t), \end{aligned}$$

where \(\sigma_{1}=\min { \{ d,\lambda_{u},\lambda \} }\). It follows that \(\lim_{t\rightarrow \infty }\sup X_{1}(t) \leq M_{1}\), where \(M_{1}=\frac{\omega }{\sigma_{1}}\). Since \(s(t)>0\), \(u(t)\geq 0\), and \(y(t)\geq 0\), then \(\lim_{t\rightarrow \infty }\sup u(t)\leq M_{1}\) and \(\lim_{t\rightarrow \infty }\sup y(t)\leq M_{1}\). Moreover, let \(X_{2}(t)=p(t)+\frac{a}{r}x(t)\). Then

$$\begin{aligned} \dot{X_{2}}(t) & =n\lambda e^{-\varepsilon_{3}\tau_{3}}y(t-\tau_{3})-cp(t)- \frac{am}{r}x(t) \\ & \leq n\lambda M_{1}-\sigma_{2}X_{2}(t), \end{aligned}$$

where \(\sigma_{2}=\min { \{ c,m \} }\). Then \(\lim_{t\rightarrow \infty }\sup X_{2}(t)\leq M_{2}\), where \(M_{2}=\frac{n \lambda M_{1}}{\sigma_{2}}\). The nonnegativity of the solution implies that \(\lim_{t\rightarrow \infty }\sup p(t)\leq M_{2}\) and \(\lim_{t\rightarrow \infty }\sup x(t)\leq M_{3}\), where \(M _{3}=\frac{r}{a}M_{2}\). This shows the ultimate boundedness of \(s(t)\), \(u(t)\), \(y(t)\), \(p(t)\), and \(x(t)\). □

2.2 Steady states and threshold parameters

In the following, we derive the basic reproduction number from system (2) by using the next-generation method and calculate the steady states. We first define the matrices \(\mathbb{F}\) and \(\mathbb{V}\) as follows:

F= [ 0 ρ π 2 s 0 e ε 1 τ 1 ρ π 1 s 0 e ε 1 τ 1 0 ( 1 ρ ) π 2 s 0 e ε 2 τ 2 ( 1 ρ ) π 1 s 0 e ε 2 τ 2 0 0 0 ] ,V= [ α + λ u 0 0 α λ 0 0 n λ e ε 3 τ 3 c ] ,

where \(s_{0}=\frac{\omega }{d}\). Then

FV 1 = [ ψ 1 ψ 2 ψ 3 ψ 4 ψ 5 ψ 6 0 0 0 ] ,

where

$$\begin{aligned} &\psi_{1} =\frac{\alpha \rho \pi_{2}s_{0}e^{-\varepsilon_{1}\tau_{1}}}{( \alpha +\lambda_{u})\lambda }+\frac{\alpha \rho \pi_{1}s_{0}ne^{- \varepsilon_{1}\tau_{1}}e^{-\varepsilon_{3}\tau_{3}}}{(\alpha +\lambda _{u})c}, \\ &\psi_{2} =\frac{\rho \pi_{2}s_{0}e^{-\varepsilon_{1}\tau_{1}}}{ \lambda }+\frac{\rho \pi_{1}s_{0}ne^{-\varepsilon_{1}\tau_{1}}e^{- \varepsilon_{3}\tau_{3}}}{c}, \\ &\psi_{3} =\frac{\rho \pi_{1}s_{0}e^{-\varepsilon_{1}\tau_{1}}}{c}, \\ &\psi_{4} =\frac{\alpha (1-\rho)\pi_{2}s_{0}e^{-\varepsilon_{2}\tau _{2}}}{(\alpha +\lambda_{u})\lambda }+\frac{\alpha (1-\rho)\pi_{1}s _{0}ne^{-\varepsilon_{2}\tau_{2}}e^{-\varepsilon_{3}\tau_{3}}}{( \alpha +\lambda_{u})c}, \\ &\psi_{5} =\frac{(1-\rho)\pi_{2}s_{0}e^{-\varepsilon_{2}\tau_{2}}}{ \lambda }+\frac{(1-\rho)\pi_{1}s_{0}ne^{-\varepsilon_{2}\tau_{2}}e ^{-\varepsilon_{3}\tau_{3}}}{c}, \\ &\psi_{6} = \frac{(1-\rho)\pi_{1}s_{0}e^{-\varepsilon_{2}\tau_{2}}}{c}. \end{aligned}$$

The basic reproduction number \(\mathcal{R}_{0}\) can be computed as the spectral radius of \(\mathbb{FV}^{-1}\):

$$ \mathcal{R}_{0}=\frac{n\pi_{1}s_{0}\gamma }{c}+\frac{\pi_{2}s_{0}e ^{\varepsilon_{3}\tau_{3}}\gamma }{\lambda }, $$

where

$$ \gamma = \biggl( \frac{\alpha \rho }{\alpha +\lambda_{u}}e^{-\varepsilon _{1}\tau_{1}-\varepsilon_{3}\tau_{3}}+(1-\rho)e^{-\varepsilon_{2}\tau_{2}- \varepsilon_{3}\tau_{3}} \biggr). $$
(4)

The parameter \(\mathcal{R}_{0}\) can be written as \(\mathcal{R}_{0}= \mathcal{R}_{01}+\mathcal{R}_{02}\), where

$$ \mathcal{R}_{01}=\frac{n\pi_{1}s_{0}\gamma }{c}, \qquad \mathcal{R} _{02}=\frac{\pi_{2}s_{0}e^{\varepsilon_{3}\tau_{3}}\gamma }{\lambda }. $$

The model has three steady states:

  1. (i)

    The pathogen-free steady state \(\Omega_{0}=(s_{0},0,0,0,0)\).

  2. (ii)

    The infected steady state without antibodies \(\Omega_{1}=(s_{1},u _{1},y_{1},p_{1},0)\), where

    $$\begin{aligned} &s_{1} =\frac{s_{0}}{\mathcal{R}_{0}},\qquad y_{1}= \frac{cd}{n \pi_{1}\lambda e^{-\varepsilon_{3}\tau_{3}}+c\pi_{2}}(\mathcal{R}_{0}-1), \\ &u_{1} =\frac{\rho \omega e^{-\varepsilon_{1}\tau_{1}}}{ ( \alpha +\lambda_{u} ) \mathcal{R}_{0}}(\mathcal{R}_{0}-1),\qquad p_{1}=\frac{n\lambda e^{-\varepsilon_{3}\tau_{3}}}{c}y_{1}. \end{aligned}$$
  3. (iii)

    The infected steady state with antibodies \(\Omega_{2}=(s_{2},u _{2},y_{2},p_{2},x_{2})\), where

    $$\begin{aligned} \begin{aligned}&s_{2} =\frac{(\alpha +\lambda_{u})u_{2}}{\rho e^{-\varepsilon_{1} \tau_{1}} ( \pi_{1}p_{2}+\pi_{2}y_{2} ) },\qquad y_{2}= \frac{-B+\sqrt{B^{2}-4AC}}{2A}, \\ &u_{2} =\frac{\omega \rho e^{-\varepsilon_{1}\tau_{1}} ( \pi _{1}m+\pi_{2}ry_{2} ) }{(\alpha +\lambda_{u})[rd+ ( \pi_{1}m+ \pi_{2}ry_{2} ) ]},\qquad p_{2}= \frac{m}{r}, \\ & x_{2}= \frac{c}{a} \biggl( \frac{n\lambda e^{-\varepsilon_{3}\tau_{3}}y_{2}}{cp _{2}}-1 \biggr), \end{aligned} \end{aligned}$$
    (5)

    and

    $$ A=\lambda \pi_{2}r,\qquad B=\lambda (rd+\pi_{1}m)- \frac{\gamma \omega \pi_{2}r}{e^{-\varepsilon_{3}\tau_{3}}},\qquad C=-\frac{ \gamma \omega \pi_{1}m}{e^{-\varepsilon_{3}\tau_{3}}}. $$
    (6)

We note that \(\Omega_{2}\) exists if \(\frac{n\lambda e^{-\varepsilon _{3}\tau_{3}}y_{2}}{cp_{2}}>1\). Now we can define antibody immune response activation number as follows:

$$ \mathcal{R}_{1}=\frac{n\lambda e^{-\varepsilon_{3}\tau_{3}}y_{2}}{cp _{2}}. $$
(7)

It follows that \(x_{2}=\frac{c}{a} ( \mathcal{R}_{1}-1 ) \). Thus, an infected steady state with antibodies \(\Omega_{2}=(s_{2},u _{2},y_{2},p_{2},x_{2})\) exists when \(\mathcal{R}_{1}>1\).

Lemma 2

Let \(\mathcal{R}_{0}>1\), then (i) if \(\mathcal{R} _{1}\leq 1\), then \(p_{1}\leq p_{2}\), and (ii) if \(\mathcal{R}_{1}>1\), then \(p_{1}>p_{2}\).

Proof

(i) Let \(\mathcal{R}_{1}\leq 1\), then \(\frac{n\lambda e^{-\varepsilon_{3}\tau_{3}}y_{2}}{cp_{2}}\leq 1\). Using Eq. (5), we obtain

$$ \frac{n\lambda e^{-\varepsilon_{3}\tau_{3}}}{cp_{2}} \biggl( \frac{-B+\sqrt{B ^{2}-4AC}}{2A} \biggr) \leq 1, $$

which implies that

$$ \biggl( \frac{2Acp_{2}}{n\lambda e^{-\varepsilon_{3}\tau_{3}}}+B \biggr) ^{2}\geq B^{2}-4AC. $$

Using Eq. (6), we get

$$ \frac{4cm^{2}e^{\varepsilon_{3}\tau_{3}}\pi_{2} ( n\lambda \pi _{1}+c\pi_{2}e^{\varepsilon_{3}\tau_{3}} ) }{n^{2}p_{2}} ( p _{2}-p_{1} ) \geq 0. $$

Thus \(p_{1}\leq p_{2}\). The proof of (ii) can be done in a similar way. □

2.3 Global properties

We define a function \(G(\theta)=\theta -1-\ln \theta \) and use the notation \((s,u,y,p,x)=(s(t),u(t),y(t), p(t),x(t))\).

Theorem 1

The pathogen-free steady state \(\Omega_{0}\) of system (2) is globally asymptotically stable when \(\mathcal{R} _{0}\leq 1\).

Proof

Define \(U_{0}(s,u,y,p,x)\) as follows:

$$\begin{aligned} U_{0} =&\gamma s_{0}G \biggl( \frac{s}{s_{0}} \biggr) +\frac{\alpha }{ \alpha +\lambda_{u}}e^{-\varepsilon_{3}\tau_{3}}u+e^{-\varepsilon_{3} \tau_{3}}y \\ &{}+\frac{(1-\mathcal{R}_{02})}{n}p+ \frac{a(1-\mathcal{R}_{02})}{rn}x \\ &{}+e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} \int_{0}^{\tau_{1}}\bigl(\pi_{1}s(t-\theta)p(t- \theta)+\pi_{2}s(t-\theta)y(t-\theta)\bigr)\,d\theta \\ &{}+e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho) \int _{0}^{\tau_{2}} \bigl( \pi_{1}s(t- \theta)p(t-\theta)+\pi_{2}s(t- \theta)y(t-\theta) \bigr)\,d\theta \\ &{}+e^{-\varepsilon_{3}\tau_{3}}(1-\mathcal{R}_{02})\lambda \int_{0} ^{\tau_{3}}y(t-\theta)\,d\theta, \end{aligned}$$

where γ is defined by Eq. (4). We have \(U_{0}(s,u,y,p,x)>0\) for all \(s,u,y,p,x>0\), while \(U_{0}(s_{0},0,0,0,0)=0\). Calculate \(\frac{dU_{0}}{dt}\) along the solution of system (2) as follows:

$$\begin{aligned} \frac{dU_{0}}{dt} =&\gamma \biggl( 1-\frac{s_{0}}{s} \biggr) \{ \omega -ds-\pi_{1}sp-\pi_{2}sy \} \\ &{} +e^{-\varepsilon_{3}\tau_{3}}\frac{\alpha }{\alpha +\lambda_{u}} \bigl\{ \rho e^{-\varepsilon_{1}\tau_{1}}s(t- \tau_{1})\bigl[\pi_{1}p(t- \tau_{1})+ \pi_{2}y(t-\tau_{1})\bigr]-(\alpha +\lambda_{u})u \bigr\} \\ &{} +e^{-\varepsilon_{3}\tau_{3}} \bigl\{ (1-\rho)e^{-\varepsilon_{2} \tau_{2}}s(t-\tau_{2}) \bigl[\pi_{1}p(t-\tau_{2})+\pi_{2}y(t- \tau_{2})\bigr]- \lambda y+\alpha u \bigr\} \\ &{} +\frac{(1-\mathcal{R}_{02})}{n} \bigl\{ n\lambda y(t-\tau_{3})e ^{-\varepsilon_{3}\tau_{3}}-cp-apx \bigr\} +\frac{a(1-\mathcal{R} _{02})}{rn} \{ rpx-mx \} \\ &{}+e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} \bigl\{ \pi_{1}sp+ \pi_{2}sy-\pi_{1}s(t- \tau_{1})p(t- \tau_{1})-\pi_{2}s(t-\tau_{1})y(t- \tau_{1}) \bigr\} \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho) \bigl\{ \pi_{1}sp+ \pi_{2}sy-\pi_{1}s(t-\tau_{2})p(t- \tau_{2})-\pi _{2}s(t-\tau_{2})y(t- \tau_{2}) \bigr\} \\ &{} +\lambda e^{-\varepsilon_{3}\tau_{3}}(1-\mathcal{R}_{02}) \bigl\{ y-y(t- \tau_{3}) \bigr\} . \end{aligned}$$
(8)

Equation (8) can be simplified as follows:

$$\begin{aligned} \frac{dU_{0}}{dt} =&-\gamma \frac{d(s-s_{0})^{2}}{s}+\gamma (\pi_{1}s _{0}p+\pi_{2}s_{0}y)-\lambda e^{-\varepsilon_{3}\tau_{3}}y \\ &{} -\frac{c(1-\mathcal{R}_{02})}{n}p- \frac{am(1-\mathcal{R}_{02})}{rn}x+\lambda (1-\mathcal{R}_{02})e^{- \varepsilon_{3}\tau_{3}}y \\ =&-\gamma \frac{d(s-s_{0})^{2}}{s}+ \biggl( \gamma \pi_{1}s_{0}- \frac{c(1- \mathcal{R}_{02})}{n} \biggr) p \\ &{}+ \bigl( \gamma \pi_{2}s_{0}-\lambda e ^{-\varepsilon_{3}\tau_{3}} \mathcal{R}_{02} \bigr) y-\frac{am(1- \mathcal{R}_{02})}{rn}x. \end{aligned}$$

We have

$$\begin{aligned} & \gamma \pi_{1}s_{0}-\frac{c}{n}(1- \mathcal{R}_{02}) =\frac{c}{n} ( \mathcal{R}_{0}-1 ), \qquad \gamma \pi_{2}s_{0}-\lambda e^{-\varepsilon_{3}\tau_{3}}\mathcal{R} _{02} =0. \end{aligned}$$

Therefore, we obtain

$$ \frac{dU_{0}}{dt}=-\gamma \frac{d(s-s_{0})^{2}}{s}+\frac{c}{n} ( \mathcal{R}_{0}-1 ) p+\frac{am}{rn} ( \mathcal{R}_{02}-1 ) x. $$

Thus, \(\frac{dU_{0}}{dt}\leq 0\) when \(\mathcal{R}_{0}\leq 1\) for all \(s,p,x>0\). Moreover, \(\frac{dU_{0}}{dt}=0\) if and only if \(x(t)=0\), \(p(t)=0\), and \(s(t)=s_{0}\). Let \(D_{0}=\{(s,u,y,p,x): \frac{dU_{0}}{dt}=0\}\) and \(D_{0}^{\prime }\) be the largest invariant subset of \(D_{0}\). The solutions of system (2) tend to \(D_{0}^{\prime }\) [38]. For each element in \(D_{0}^{\prime }\), we have \(p(t)=0\). Thus Eq. (2)4 yields

$$ \dot{p}(t)=0=n\lambda e^{-\varepsilon_{3}\tau_{3}}y(t-\tau_{3}). $$

Then \(y(t)=0\). From Eq. (2)3 we have

$$ 0=\alpha u(t). $$

Then \(u(t)=0\). It follows that \(D_{0}^{\prime }\) contains a single point that is \(\{ \Omega_{0}\}\). From LaSalle’s invariance principle, \(\Omega_{0}\) is globally asymptotically stable when \(\mathcal{R}_{0} \leq 1\). □

Theorem 2

For system (2), assume that \(\mathcal{R} _{1}\leq 1<\mathcal{R}_{0}\), then \(\Omega_{1}\) is globally asymptotically stable.

Proof

Let \(U_{1}(s,u,y,p,x)\) be given as follows:

$$\begin{aligned} U_{1} =&\gamma s_{1}G \biggl( \frac{s}{s_{1}} \biggr) +\frac{\alpha }{ \alpha +\lambda_{u}}e^{-\varepsilon_{3}\tau_{3}}u_{1}G \biggl( \frac{u}{u _{1}} \biggr) +e^{-\varepsilon_{3}\tau_{3}}y_{1}G \biggl( \frac{y}{y_{1}} \biggr) \\ &{} +\frac{\gamma \pi_{1}s_{1}p_{1}}{n\lambda e^{-\varepsilon_{3}\tau _{3}}y_{1}}p_{1}G \biggl( \frac{p}{p_{1}} \biggr) + \frac{\gamma a\pi_{1}s _{1}p_{1}}{rn\lambda e^{-\varepsilon_{3}\tau_{3}}y_{1}}x \\ &{}+e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{1}p_{1} \int_{0}^{\tau_{1}}G \biggl( \frac{s(t-\theta)p(t-\theta)}{s_{1}p_{1}} \biggr)\,d \theta \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{1}s_{1}p_{1} \int_{0}^{\tau_{2}}G \biggl( \frac{s(t-\theta)p(t-\theta)}{s_{1}p_{1}} \biggr)\,d \theta \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{1}y_{1} \int_{0}^{\tau_{1}}G \biggl( \frac{s(t-\theta)y(t-\theta)}{s_{1}y_{1}} \biggr)\,d \theta \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{2}s_{1}y_{1} \int_{0}^{\tau_{2}}G \biggl( \frac{s(t-\theta)y(t-\theta)}{s_{1}y_{1}} \biggr)\,d \theta \\ &{}+\gamma \pi_{1}s_{1}p_{1} \int_{0}^{\tau_{3}}G \biggl( \frac{y(t- \theta)}{y_{1}} \biggr)\,d \theta. \end{aligned}$$

We have \(U_{1}(s,u,y,p,x)>0\) for all \(s,u,y,p,x>0\) and \(U_{1}(s _{1},u_{1},y_{1},p_{1},0)=0\). Calculating \(\frac{dU_{1}}{dt}\), we obtain

$$\begin{aligned} \frac{dU_{1}}{dt} =&\gamma \biggl( 1-\frac{s_{1}}{s} \biggr) ( \omega -ds- \pi_{1}sp-\pi_{2}sy ) \\ &{}+e^{-\varepsilon_{3}\tau_{3}}\frac{\alpha }{\alpha +\lambda_{u}} \biggl( 1-\frac{u_{1}}{u} \biggr) \bigl\{ \rho e^{-\varepsilon_{1}\tau _{1}}s(t-\tau_{1}) \bigl[ \pi_{1}p(t- \tau_{1})+\pi_{2}y(t-\tau_{1}) \bigr] \\ &{} -(\alpha +\lambda _{u})u \bigr\} \\ &{} +e^{-\varepsilon_{3}\tau_{3}} \biggl( 1-\frac{y_{1}}{y} \biggr) \bigl\{ (1- \rho)e^{-\varepsilon_{2}\tau_{2}}s(t-\tau_{2})\bigl[\pi_{1}p(t-\tau _{2})+\pi_{2}y(t-\tau_{2})\bigr]-\lambda y+ \alpha u \bigr\} \\ &{}+\gamma \frac{\pi_{1}s_{1}p_{1}}{n\lambda e^{-\varepsilon_{3}\tau _{3}}y_{1}} \biggl( 1-\frac{p_{1}}{p} \biggr) \bigl( n \lambda e^{- \varepsilon_{3}\tau_{3}}y(t-\tau_{3})-cp-apx \bigr) \\ &{} +\gamma \frac{a\pi _{1}s_{1}p_{1}}{rn\lambda e^{-\varepsilon_{3}\tau_{3}}y_{1}} ( rpx-mx ) \\ &{}+e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{1}p_{1} \biggl[ \frac{sp}{s_{1}p _{1}}-\frac{s(t-\tau_{1})p(t-\tau_{1})}{s_{1}p_{1}}+\ln \biggl( \frac{s(t- \tau_{1})p(t-\tau_{1})}{sp} \biggr) \biggr] \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{1}s_{1}p_{1} \biggl[ \frac{sp}{s_{1}p_{1}}-\frac{s(t-\tau_{2})p(t-\tau _{2})}{s_{1}p_{1}}+\ln \biggl( \frac{s(t-\tau_{2})p(t-\tau_{2})}{sp} \biggr) \biggr] \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{1}y_{1} \biggl[ \frac{sy}{s_{1}y _{1}}-\frac{s(t-\tau_{1})y(t-\tau_{1})}{s_{1}y_{1}}+\ln \biggl( \frac{s(t- \tau_{1})y(t-\tau_{1})}{sy} \biggr) \biggr] \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{2}s_{1}y_{1} \biggl[ \frac{sy}{s_{1}y_{1}}-\frac{s(t-\tau_{2})y(t-\tau _{2})}{s_{1}y_{1}}+\ln \biggl( \frac{s(t-\tau_{2})y(t-\tau_{2})}{sy} \biggr) \biggr] \\ &{} +\gamma \pi_{1}s_{1}p_{1} \biggl[ \frac{y}{y_{1}}-\frac{y(t-\tau_{3})}{y _{1}}+\ln \biggl( \frac{y(t-\tau_{3})}{y} \biggr) \biggr] . \end{aligned}$$
(9)

Simplifying Eq. (9) and applying the steady state conditions for \(\Omega_{1}\)

$$\begin{aligned}& \omega =ds_{1}+\pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1}, \\& e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} [ \pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ] =\alpha e^{-\varepsilon _{3}\tau_{3}}u_{1}, \\& e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} [ \pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ] +e^{-\varepsilon_{2}\tau _{2}-\varepsilon_{3}\tau_{3}}(1-\rho) [ \pi_{1}s_{1}p_{1}+\pi _{2}s_{1}y_{1} ] =\lambda e^{-\varepsilon_{3}\tau_{3}}y_{1}, \\& p_{1}=\frac{n\lambda e^{-\varepsilon_{3}\tau_{3}}y_{1}}{c}, \end{aligned}$$

we get

$$\begin{aligned} \frac{dU_{1}}{dt} =&\gamma \biggl( 1-\frac{s_{1}}{s} \biggr) ( ds _{1}-ds ) +\gamma \biggl( 1-\frac{s_{1}}{s} \biggr) ( \pi_{1}s _{1}p_{1}+\pi_{2}s_{1}y_{1} ) \\ &{} -e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} \biggl( \pi_{1}s_{1}p_{1} \frac{s(t-\tau_{1})p(t- \tau_{1})u_{1}}{s_{1}p_{1}u}+\pi_{2}s_{1}y_{1} \frac{s(t-\tau_{1})y(t- \tau_{1})u_{1}}{s_{1}y_{1}u} \biggr) \\ &{}+e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} ( \pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ) \\ &{}-e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho) \biggl( \pi_{1}s_{1}p_{1} \frac{s(t-\tau_{2})p(t-\tau_{2})y_{1}}{s_{1}p _{1}y}+\pi_{2}s_{1}y_{1} \frac{s(t-\tau_{2})y(t-\tau_{2})}{s_{1}y} \biggr) \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} ( \pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ) +e^{-\varepsilon_{2}\tau _{2}-\varepsilon_{3}\tau_{3}}(1-\rho) ( \pi_{1}s_{1}p_{1}+\pi _{2}s_{1}y_{1} ) \\ &{}-e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} ( \pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ) \frac{uy_{1}}{u_{1}y}+ \gamma \pi_{1}s_{1}p_{1} \biggl( 1-\frac{y(t-\tau_{3})p_{1}}{y_{1}p} \biggr) \\ &{} +\frac{a}{c}\gamma \pi_{1}s_{1} ( p_{1}-p_{2} ) x+e^{- \varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{ \alpha +\lambda_{u}} \pi_{1}s_{1}p_{1}\ln \biggl( { \frac{s(t-\tau_{1})p(t- \tau_{1})}{sp}} \biggr) \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{1}s_{1}p_{1} \ln \biggl( {\frac{s(t-\tau_{2})p(t-\tau_{2})}{sp}} \biggr) \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{1}y_{1} \ln \biggl( {\frac{s(t-\tau _{1})y(t-\tau_{1})}{sy}} \biggr) \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{2}s_{1}y_{1} \ln \biggl( {\frac{s(t-\tau_{2})y(t-\tau_{2})}{sy}} \biggr) \\ &{}+\gamma \pi_{1}s_{1}p_{1}\ln \biggl( { \frac{y(t-\tau_{3})}{y}} \biggr). \end{aligned}$$

Consider the following equalities with (\(i=1 \)):

$$\begin{aligned} \begin{aligned} &\ln \biggl( {\frac{s(t-\tau_{1})p(t-\tau_{1})}{sp}} \biggr) \\ &\quad =\ln \biggl( {\frac{s(t-\tau_{1})p(t-\tau_{1})u_{i}}{s_{i}p_{i}u}} \biggr) + \ln \biggl( { \frac{uy_{i}}{u_{i}y}} \biggr) +\ln \biggl( {\frac{yp_{i}}{y _{i}p}} \biggr) +\ln \biggl( {\frac{s_{i}}{s}} \biggr), \\ &\ln \biggl( {\frac{s(t-\tau_{2})p(t-\tau_{2})}{sp}} \biggr) =\ln \biggl( {\frac{s(t-\tau_{2})p(t-\tau_{2})y_{i}}{s_{i}p_{i}y}} \biggr) + \ln \biggl( {\frac{yp_{i}}{y_{i}p}} \biggr) +\ln \biggl( { \frac{s_{i}}{s}} \biggr), \\ &\ln \biggl( {\frac{s(t-\tau_{1})y(t-\tau_{1})}{sy}} \biggr) =\ln \biggl( {\frac{s(t-\tau_{1})y(t-\tau_{1})u_{i}}{s_{i}y_{i}u}} \biggr) + \ln \biggl( {\frac{uy_{i}}{u_{i}y}} \biggr) +\ln \biggl( { \frac{s_{i}}{s}} \biggr), \\ &\ln \biggl( {\frac{s(t-\tau_{2})y(t-\tau_{2})}{sy}} \biggr) =\ln \biggl( {\frac{s(t-\tau_{2})y(t-\tau_{2})}{s_{i}y}} \biggr) +\ln \biggl( {\frac{s_{i}}{s}} \biggr), \\ &\ln \biggl( {\frac{y(t-\tau_{3})}{y}} \biggr) =\ln \biggl( {\frac{y(t- \tau_{3})p_{i}}{y_{i}p}} \biggr) +\ln \biggl( {\frac{y_{i}p}{yp_{i}}} \biggr), \end{aligned} \end{aligned}$$
(10)

we obtain

$$\begin{aligned} \frac{dU_{1}}{dt} =&-\gamma \frac{d(s-s_{1})^{2}}{s}-\gamma (\pi_{1}s _{1}p_{1}+\pi_{2}s_{1}y_{1})G \biggl( \frac{s_{1}}{s} \biggr) \\ &{} -e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{1}p_{1}G \biggl( \frac{s(t-\tau _{1})p(t-\tau_{1})u_{1}}{s_{1}p_{1}u} \biggr) \\ &{} -e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{1}y_{1}G \biggl( \frac{s(t-\tau _{1})y(t-\tau_{1})u_{1}}{s_{1}y_{1}u} \biggr) \\ &{} -e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{1}s_{1}p_{1}G \biggl( \frac{s(t-\tau_{2})p(t-\tau_{2})y_{1}}{s_{1}p _{1}y} \biggr) \\ &{}-e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{2}s_{1}y_{1}G \biggl( \frac{s(t-\tau_{2})y(t-\tau_{2})}{s_{1}y} \biggr) \\ &{} -e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}(\pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1})G \biggl( \frac{y_{1}u}{yu_{1}} \biggr) \\ &{} -\gamma \pi_{1}s_{1}p_{1}G \biggl( \frac{y(t-\tau_{3})p_{1}}{y_{1}p} \biggr) +\frac{a}{c}\gamma \pi_{1}s _{1} ( p_{1}-p_{2} ) x. \end{aligned}$$

From Lemma 2, we have \(p_{1}\leq p_{2}\) when \(\mathcal{R}_{1}\leq 1\). Thus, \(\frac{dU_{1}}{dt}\leq 0\) and \(\frac{dU_{1}}{dt}=0\) occur at the infected steady state without antibodies \(\Omega_{1}\). Let \(D_{1}^{ \prime }\) be the largest invariant subset of the set \(D_{1}= \{ ( s,u,y,p,x ):\frac{dU_{1}}{dt}=0 \} \). Thus, the solutions of system (2) tend to \(D_{1}^{\prime }\). It is clear that \(D_{1}= \{ \Omega_{1} \} \). Using LaSalle’s invariance principle, we conclude that \(\Omega_{1}\) is globally asymptotically stable when \(\mathcal{R}_{1}\leq 1\) and \(\mathcal{R}_{0}>1\). □

Theorem 3

For system (2), suppose that \(\mathcal{R} _{1}>1\), then \(\Omega_{2}\) is globally asymptotically stable.

Proof

Consider \(U_{2}(s,u,y,p,x)\):

$$\begin{aligned} U_{2} =&\gamma s_{2}G \biggl( \frac{s}{s_{2}} \biggr) +\frac{\alpha }{ \alpha +\lambda_{u}}e^{-\varepsilon_{3}\tau_{3}}u_{2}G \biggl( \frac{u}{u _{2}} \biggr) +e^{-\varepsilon_{3}\tau_{3}}y_{2}G \biggl( \frac{y}{y_{2}} \biggr) \\ &{} +\gamma \frac{\pi_{1}s_{2}p_{2}}{n\lambda e^{-\varepsilon_{3}\tau _{3}}y_{2}}p_{2}G \biggl( \frac{p}{p_{2}} \biggr) +\gamma \frac{a\pi_{1}s _{2}p_{2}}{rn\lambda e^{-\varepsilon_{3}\tau_{3}}y_{2}}x_{2}G \biggl( \frac{x}{x _{2}} \biggr) \\ &{}+e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{2}p_{2} \int_{0}^{\tau_{1}}G \biggl( \frac{s(t-\theta)p(t-\theta)}{s_{2}p_{2}} \biggr) \,d\theta \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{1}s_{2}p_{2} \int_{0}^{\tau_{2}}G \biggl( \frac{s(t-\theta)p(t-\theta)}{s_{2}p_{2}} \biggr) \,d\theta \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{2}y_{2} \int_{0}^{\tau_{1}}G \biggl( \frac{s(t-\theta)y(t-\theta)}{s_{2}y_{2}} \biggr) \,d\theta \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{2}s_{2}y_{2} \int_{0}^{\tau_{2}}G \biggl( \frac{s(t-\theta)y(t-\theta)}{s_{2}y_{2}} \biggr) \,d\theta \\ &{} +\gamma \pi_{1}s_{2}p_{2} \int_{0}^{\tau_{3}}G \biggl( \frac{y(t- \theta)}{y_{2}} \biggr) \,d\theta. \end{aligned}$$

We have \(U_{2}(s,u,y,p,x)>0\) for all \(s,u,y,p,x>0\), while \(U_{2}(s,u,y,p,x)\) reaches its global minimum at \(\Omega_{2}\). Calculate \(\frac{dU_{2}}{dt}\) as follows:

$$\begin{aligned} \frac{dU_{2}}{dt} =&\gamma \biggl( 1-\frac{s_{2}}{s} \biggr) ( \omega -ds-\pi_{1}sp-\pi_{2}sy ) \\ &{} +e^{-\varepsilon_{3}\tau_{3}}\frac{\alpha }{\alpha +\lambda_{u}} \biggl( 1-\frac{u_{2}}{u} \biggr) \\ &{}\times \bigl\{ \rho e^{-\varepsilon_{1}\tau _{1}}s(t-\tau_{1}) \bigl[ \pi_{1}p(t-\tau_{1})+\pi_{2}y(t- \tau_{1}) \bigr] -(\alpha +\lambda _{u})u \bigr\} \\ &{} +e^{-\varepsilon_{3}\tau_{3}} \biggl( 1-\frac{y_{2}}{y} \biggr) \bigl\{ (1- \rho)e^{-\varepsilon_{2}\tau_{2}}s(t-\tau_{2})\bigl[\pi_{1}p(t-\tau _{2})+\pi_{2}y(t-\tau_{2})\bigr]-\lambda y+ \alpha u \bigr\} \\ &{} +\gamma \frac{\pi_{1}s_{2}p_{2}}{n\lambda e^{-\varepsilon_{3}\tau _{3}}y_{2}} \biggl( 1-\frac{p_{2}}{p} \biggr) \bigl( n \lambda y(t-\tau _{3})e^{-\varepsilon_{3}\tau_{3}}-cp-apx \bigr) \\ &{} +\gamma \frac{a\pi_{1}s_{2}p_{2}}{rn\lambda e^{-\varepsilon_{3}\tau _{3}}y_{2}} \biggl( 1-\frac{x_{2}}{x} \biggr) ( rpx-mx ) \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{2}p_{2} \biggl[ \frac{sp}{s_{2}p _{2}}-\frac{s(t-\tau_{1})p(t-\tau_{1})}{s_{2}p_{2}}+\ln \biggl( \frac{s(t- \tau_{1})p(t-\tau_{1})}{sp} \biggr) \biggr] \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{1}s_{2}p_{2} \biggl[ \frac{sp}{s_{2}p_{2}}-\frac{s(t-\tau_{2})p(t-\tau _{2})}{s_{2}p_{2}}+\ln \biggl( \frac{s(t-\tau_{2})p(t-\tau_{2})}{sp} \biggr) \biggr] \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{2}y_{2} \biggl[ \frac{sy}{s_{2}y _{2}}-\frac{s(t-\tau_{1})y(t-\tau_{1})}{s_{2}y_{2}}+\ln \biggl( \frac{s(t- \tau_{1})y(t-\tau_{1})}{sy} \biggr) \biggr] \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{2}s_{2}y_{2} \biggl[ \frac{sy}{s_{2}y_{2}}-\frac{s(t-\tau_{2})y(t-\tau _{2})}{s_{2}y_{2}}+\ln \biggl( \frac{s(t-\tau_{2})y(t-\tau_{2})}{sy} \biggr) \biggr] \\ &{} +\gamma \pi_{1}s_{2}p_{2} \biggl[ \frac{y}{y_{2}}-\frac{y(t-\tau_{3})}{y _{2}}+\ln \biggl( \frac{y(t-\tau_{3})}{y} \biggr) \biggr] . \end{aligned}$$
(11)

Simplifying Eq. (11) and applying the steady state conditions for \(\Omega_{2}\):

$$\begin{aligned}& \omega =ds_{2}+\pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2}, \\& e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} [ \pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ] =\alpha e^{-\varepsilon _{3}\tau_{3}}u_{2}, \\& e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} [ \pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ] +e^{-\varepsilon_{2}\tau _{2}-\varepsilon_{3}\tau_{3}}(1-\rho) [ \pi_{1}s_{2}p_{2}+\pi _{2}s_{2}y_{2} ] =\lambda e^{-\varepsilon_{3}\tau_{3}}y_{2}, \\& p_{2}=\frac{m}{r}, \qquad n\lambda e^{-\varepsilon_{3}\tau_{3}}y _{2}=cp_{2}+ap_{2}x_{2}, \end{aligned}$$

we get

$$\begin{aligned} \frac{dU_{2}}{dt} =&\gamma \biggl( 1-\frac{s_{2}}{s} \biggr) ( ds _{2}-ds ) +\gamma \biggl( 1-\frac{s_{2}}{s} \biggr) ( \pi_{1}s _{2}p_{2}+\pi_{2}s_{2}y_{2} ) \\ &{} -e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} \biggl( \pi_{1}s_{2}p_{2} \frac{s(t-\tau_{1})p(t- \tau_{1})u_{2}}{s_{2}p_{2}u}+\pi_{2}s_{2}y_{2} \frac{s(t-\tau_{1})y(t- \tau_{1})u_{2}}{s_{2}y_{2}u} \biggr) \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} ( \pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) \\ &{} -e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho) \biggl( \pi_{1}s_{2}p_{2} \frac{s(t-\tau_{2})p(t-\tau_{2})y_{2}}{s_{2}p _{2}y}+\pi_{2}s_{2}y_{2} \frac{s(t-\tau_{2})y(t-\tau_{2})}{s_{2}y} \biggr) \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} ( \pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) +e^{-\varepsilon_{2}\tau _{2}-\varepsilon_{3}\tau_{3}}(1-\rho) ( \pi_{1}s_{2}p_{2}+\pi _{2}s_{2}y_{2} ) \\ &{} -e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}} ( \pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) \frac{uy_{2}}{u_{2}y}+ \gamma \pi_{1}s_{2}p_{2} \biggl( 1-\frac{y(t-\tau_{3})p_{2}}{y_{2}p} \biggr) \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{2}p_{2} \ln \biggl( {\frac{s(t-\tau _{1})p(t-\tau_{1})}{sp}} \biggr) \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{1}s_{2}p_{2} \ln \biggl( {\frac{s(t-\tau_{2})p(t-\tau_{2})}{sp}} \biggr) \\ &{} +e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{2}y_{2} \ln \biggl( {\frac{s(t-\tau _{1})y(t-\tau_{1})}{sy}} \biggr) \\ &{} +e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{2}s_{2}y_{2} \ln \biggl( {\frac{s(t-\tau_{2})y(t-\tau_{2})}{sy}} \biggr) \\ &{} +\gamma \pi_{1}s_{2}p_{2}\ln \biggl( { \frac{y(t-\tau_{3})}{y}} \biggr). \end{aligned}$$

Applying equalities (10) when (\(i=2 \)), we obtain

$$\begin{aligned} \frac{dU_{2}}{dt} =&-\gamma \frac{d(s-s_{2})^{2}}{s}-\gamma (\pi_{1}s _{2}p_{2}+\pi_{2}s_{2}y_{2})G \biggl( \frac{s_{2}}{s} \biggr) \\ &{}-e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{2}p_{2}G \biggl( \frac{s(t-\tau _{1})p(t-\tau_{1})u_{2}}{s_{2}p_{2}u} \biggr) \\ &{} -e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{2}y_{2}G \biggl( \frac{s(t-\tau _{1})y(t-\tau_{1})u_{2}}{s_{2}y_{2}u} \biggr) \\ &{} -e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{1}s_{2}p_{2}G \biggl( \frac{s(t-\tau_{2})p(t-\tau_{2})y_{2}}{s_{2}p _{2}y} \biggr) \\ &{} -e^{-\varepsilon_{2}\tau_{2}-\varepsilon_{3}\tau_{3}}(1-\rho)\pi _{2}s_{2}y_{2}G \biggl( \frac{s(t-\tau_{2})y(t-\tau_{2})}{s_{2}y} \biggr) \\ &{} -e^{-\varepsilon_{1}\tau_{1}-\varepsilon_{3}\tau_{3}}\frac{\alpha \rho }{\alpha +\lambda_{u}}(\pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2})G \biggl( \frac{y_{2}u}{yu_{2}} \biggr) \\ &{} -\gamma \pi_{1}s_{2}p_{2}G \biggl( \frac{y(t-\tau_{3})p_{2}}{y_{2}p} \biggr). \end{aligned}$$

Since \(\mathcal{R}_{1}>1\), then \(s_{2}\), \(u_{2}\), \(y_{2}\), \(p_{2}\), and \(x_{2}>0\). We obtain \(\frac{dU_{2}}{dt}\leq 0\), and then the solutions of system (2) tend to \(D_{2}^{\prime }\), the largest invariant subset of \(D_{2}= \{ ( s,u,y,p,x ):\frac{dU_{2}}{dt}=0 \} \). Clearly, \(\frac{dU_{2}}{dt}=0\) when \(s=s_{2}\), \(u=u_{2}\), \(y=y_{2}\), and \(p=p_{2}\). Since \(p=p_{2} \) in \(D_{2}^{\prime }\), then

$$ \dot{p}=0=n\lambda e^{-\varepsilon_{3}\tau_{3}}y_{2}-cp_{2}-ap_{2}x, $$

which gives \(x=x_{2}\). Therefore, \(\frac{dU_{2}}{dt}=0\) when \(s=s_{2}\), \(u=u_{2}\), \(y=y_{2}\), \(p=p_{2}\), and \(x=x_{2}\). The global asymptotic stability of \(\Omega_{2}\) is conducted from LaSalle’s invariance principle. □

3 Model with distributed delays

We consider a pathogen dynamics model with distributed delays:

$$\begin{aligned} \begin{aligned} & \dot{s}(t) =\omega -ds(t)-s(t) \bigl[ \pi_{1}p(t)+\pi_{2}y(t) \bigr] , \\ &\dot{u}(t) =\rho \int_{0}^{h_{1}}f_{1}(\tau)e^{-\mu_{1}\tau }s(t- \tau)\bigl[\pi_{1}p(t-\tau)+\pi_{2}y(t-\tau)\bigr]\,d\tau -(\alpha +\lambda _{u})u(t), \\ &\dot{y}(t) =(1-\rho) \int_{0}^{h_{2}}f_{2}(\tau)e^{-\mu_{2}\tau }s(t- \tau)\bigl[\pi_{1}p(t-\tau)+\pi_{2}y(t-\tau)\bigr]\,d\tau - \lambda y(t)+\alpha u(t), \\ &\dot{p}(t) =n\lambda \int_{0}^{h_{3}}f_{3}(\tau)e^{-\mu_{3}\tau }y(t- \tau)\,d\tau -cp(t)-ap(t)x(t), \\ &\dot{x}(t) =rp(t)x(t)-mx(t), \end{aligned} \end{aligned}$$
(12)

where \(f_{1}(\tau)e^{-\mu_{1}\tau }\) is probability that susceptible host cells contacted by the pathogens at time \(t-\tau \) survived τ time units and became latently infected at time t, \(f_{2}(\tau)e^{-\mu_{2}\tau }\) is probability that susceptible host cells contacted by the pathogens at time \(t-\tau \) survived τ time units and became actively infected at time t, and \(f_{3}(\tau)e^{-\mu_{3}\tau }\) is the probability that an immature pathogen at time \(t-\tau \) survived τ time units to become a mature pathogen at time t. The probability distribution functions \(f_{j}(\tau)\), \(j=1,\ldots,3\), satisfy the following conditions:

(i) \(f_{j}(\tau)>0\), (ii) \(\int_{0}^{h_{j}}f_{j}(\tau)\,d\tau =1\), (iii) \(\int_{0}^{h_{j}}f_{j}(\tau)e^{\ell \tau }\,d\tau<\infty \), where \(\ell >0\).

Let \(\Theta_{j}(\tau)=f_{j}(\tau)e^{-\mu_{j}\tau }\) and \(\eta_{j}= \int_{0}^{h_{j}}\Theta_{j}(\tau)\,d\tau\), \(j=1,2,3\), thus \(0<\eta_{j} \leq 1\).

The initial conditions for system (12) are the same as given by (3) where \(\kappa = \max \{h_{1},h_{2},h_{3}\}\).

3.1 Properties of solution

Lemma 3

The solutions \((s(t),u(t),y(t),p(t),x(t))\) of system (12) with initial conditions (3) are nonnegative and ultimately bounded for \(t>0\).

Proof

From Lemma 1 we have \(s(t)>0\) for all \(t\geq 0\). For \(t\in {}[ 0,\kappa ]\), we have

$$\begin{aligned} &u(t) =\phi_{2}(0)e^{-(\alpha +\lambda_{u})t} \\ &\hphantom{u(t) =}{}+ \int_{0}^{t} \biggl\{ \rho \int_{0}^{h_{1}}\Theta_{1}(\tau)s(\theta - \tau)\bigl[\pi_{1}p( \theta -\tau)+\pi_{2}y(\theta -\tau) \bigr] \biggr\} {e^{-(\alpha +\lambda _{u})(t-\theta)}}\,d\tau \,d\theta, \\ &y(t) =\phi_{3}(0)e^{-\lambda t}+ \int_{0}^{t} \biggl\{ (1-\rho) \int_{0} ^{h_{2}}\Theta_{2}(\tau)s(\theta - \tau)\bigl[\pi_{1}p(\theta -\tau)+\pi_{2}y( \theta -\tau) \bigr]\,d\tau \\ &\hphantom{y(t) =}{} +\alpha u(\theta) \biggr\} e^{-\lambda (t- \theta)}\,d\theta, \\ &p(t) =\phi_{4}(0)e^{-\int_{0}^{t}(c+ax(\upsilon))\,d\upsilon }+n \lambda \int_{0}^{t}{e^{-\int_{\theta }^{t}(c+ax(\upsilon))\,d\upsilon }} \int_{0}^{h_{3}}\Theta_{3}(\tau)y(\theta - \tau)\,d\tau \,d\theta, \\ &x(t) =\phi_{5}(0)e^{-mt+r\int_{0}^{t}p(\theta)\,d\theta }. \end{aligned}$$

We obtain by recursive argument that \(u(t)\geq 0\), \(y(t)\geq 0\), \(p(t) \geq 0\), and \(x(t)\geq 0\ \forall t\geq 0\).

Clearly \(\lim_{t\rightarrow \infty }\sup s(t)\leq \frac{ \omega }{d}\). Let us define \(Y_{1}(t)=\rho \int_{0}^{h_{1}}\Theta_{1}( \tau)s(t-\tau)\,d\tau +(1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau)s(t- \tau)\,d\tau +u(t)+y(t)\). Then

$$\begin{aligned} \dot{Y_{1}}(t) =&\rho \int_{0}^{h_{1}}\Theta_{1}(\tau) \bigl\{ \omega -ds(t-\tau)-s(t-\tau) \bigl[ \pi_{1}p(t-\tau)+ \pi_{2}y(t- \tau) \bigr] \bigr\} \,d\tau \\ &{} +(1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau) \bigl\{ \omega -ds(t- \tau)-s(t-\tau) \bigl[ \pi_{1}p(t-\tau)+ \pi_{2}y(t-\tau) \bigr] \bigr\} \,d\tau \\ &{} +\rho \int_{0}^{h_{1}}\Theta_{1}(\tau)s(t-\tau) \bigl[ \pi_{1}p(t- \tau)+\pi_{2}y(t-\tau) \bigr]\,d\tau -( \alpha +\lambda_{u})u(t) \\ &{} +(1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau)s(t-\tau) \bigl[\pi_{1}p(t- \tau)+\pi_{2}y(t-\tau)\bigr]\,d\tau - \lambda y(t)+\alpha u(t) \\ =&\omega \rho \int_{0}^{h_{1}}\Theta_{1}(\tau)\,d\tau + \omega (1- \rho) \int_{0}^{h_{2}}\Theta_{2}(\tau)\,d\tau - \rho \,d \int_{0}^{h_{1}} \Theta_{1}(\tau)s(t-\tau) \,d\tau \\ &{} -(1-\rho)\,d \int_{0}^{h_{2}}\Theta_{2}(\tau)s(t-\tau)\,d \tau - \lambda_{u}u(t)-\lambda y(t) \\ \leq& \omega -\sigma_{1} \biggl( \rho \int_{0}^{h_{1}}\Theta_{1}( \tau)s(t-\tau) \,d\tau +(1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau)s(t- \tau) \,d\tau +u(t)+y(t) \biggr) \\ =&\omega -\sigma_{1}Y_{1}(t), \end{aligned}$$

where \(\sigma_{1}=\min { \{ d,\lambda_{u},\lambda \} }\). It follows that \(\lim_{t\rightarrow \infty }\sup Y_{1}(t) \leq \tilde{M}_{1}\), where \(\tilde{M}_{1}=\frac{\omega }{\sigma_{1}}\). Since \(s(t)>0\), \(u(t)\geq 0\), and \(y(t)\geq 0\), then \(\lim_{t\rightarrow \infty }\sup u(t)\leq \tilde{M}_{1}\) and \(\lim_{t\rightarrow \infty }\sup y(t)\leq \tilde{M}_{1}\). Further, let us consider \(Y_{2}(t)=p(t)+\frac{a}{r}x(t)\). Then

$$\begin{aligned} \dot{Y_{2}}(t) & =n\lambda \int_{0}^{h_{3}}\Theta_{3}(\tau)y(t- \tau) \,d\tau -cp(t)-\frac{am}{r}x(t) \\ & \leq n\lambda \tilde{M}_{1}-\sigma_{2}Y_{2}(t), \end{aligned}$$

where \(\sigma_{2}=\min { \{ c,m \} }\). Then \(\lim_{t\rightarrow \infty }\sup Y_{2}(t)\leq \tilde{M}_{2}\), where \(\tilde{M}_{2}=\frac{n\lambda \tilde{M}_{1}}{\sigma_{2}}\). The nonnegativity of the solution implies that \(\lim_{t\rightarrow \infty }\sup p(t)\leq \tilde{M}_{2}\) and \(\lim_{t\rightarrow \infty }\sup x(t)\leq \tilde{M}_{3}\), where \(\tilde{M}_{3}= \frac{r}{a}\tilde{M}_{2}\). This shows the ultimate boundedness of \(s(t)\), \(u(t)\), \(y(t)\), \(p(t)\), and \(x(t)\). □

3.2 Steady states and threshold parameters

For system (12) the matrices \(\mathbb{F}\) and \(\mathbb{V}\) are given by

F= [ 0 ρ π 2 s 0 η 1 ρ π 1 s 0 η 1 0 ( 1 ρ ) π 2 s 0 η 2 ( 1 ρ ) π 1 s 0 η 2 0 0 0 ] ,V= [ α + λ u 0 0 α λ 0 0 n λ η 3 c ] ,

and then

FV 1 = [ ψ ˜ 1 ψ ˜ 2 ψ ˜ 3 ψ ˜ 4 ψ ˜ 5 ψ ˜ 6 0 0 0 ] ,

where

ψ ˜ 1 = α ρ ( α + λ u ) λ π 2 s 0 η 1 + α ρ ( α + λ u ) c n π 1 s 0 η 1 η 3 , ψ ˜ 2 = ρ λ π 2 s 0 η 1 + ρ c n π 1 s 0 η 1 η 3 , ψ ˜ 3 = ρ c π 1 s 0 η 1 , ψ ˜ 4 = α ( 1 ρ ) ( α + λ u ) λ π 2 s 0 η 2 + α ( 1 ρ ) ( α + λ u ) c n π 1 s 0 η 2 η 3 , ψ ˜ 5 = ( 1 ρ ) λ π 2 s 0 η 2 + ( 1 ρ ) c n π 1 s 0 η 2 η 3 , ψ ˜ 6 = ( 1 ρ ) c π 1 s 0 η 2 .

Thus, \(\tilde{\mathcal{R}}_{0}\) is given by

$$ \tilde{\mathcal{R}}_{0}=\tilde{\mathcal{R}}_{01}+\tilde{\mathcal{R}} _{02}, $$

where

$$ \tilde{\mathcal{R}}_{01}=\frac{n\pi_{1}s_{0}\tilde{\gamma }}{c},\qquad \tilde{\mathcal{R}}_{02}=\frac{\pi_{2}s_{0}\tilde{\gamma }}{ \lambda \eta_{3}}\quad \text{and}\quad \tilde{\gamma }= \biggl( \frac{ \alpha \rho }{\alpha +\lambda_{u}}\eta_{1}\eta_{3}+(1-\rho) \eta_{2} \eta_{3} \biggr). $$
(13)

The model has three steady states:

  1. (i)

    The pathogen-free steady state \(\Omega_{0}=(s_{0},0,0,0,0)\).

  2. (ii)

    The infected steady state without antibodies \(\Omega_{1}=(s_{1},u _{1},y_{1},p_{1},0)\), where

    $$\begin{aligned} &s_{1} =\frac{s_{0}}{\tilde{\mathcal{R}}_{0}},\qquad y _{1}= \frac{cd}{n\pi_{1}\lambda \eta_{3}+c\pi_{2}}(\tilde{\mathcal{R}} _{0}-1), \\ &u_{1} =\frac{\omega \rho \eta_{1}}{ ( \alpha +\lambda_{u} ) \tilde{\mathcal{R}}_{0}}(\tilde{\mathcal{R}}_{0}-1), \qquad p_{1}=\frac{n\lambda \eta_{3}}{c}y_{1}. \end{aligned}$$
  3. (iii)

    The infected steady state with antibodies \(\Omega_{2}=(s_{2},u _{2},y_{2},p_{2},x_{2})\), where

    $$\begin{aligned} \begin{aligned}&s_{2} =\frac{(\alpha +\lambda_{u})u_{2}}{\rho \eta_{1} ( \pi _{1}p_{2}+\pi_{2}y_{2} ) },\qquad y_{2}= \frac{-\tilde{B}+\sqrt{ \tilde{B}^{2}-4\tilde{A}\tilde{C}}}{2\tilde{A}}, \\ &u_{2} =\frac{\omega \rho \eta_{1} ( \pi_{1}m+\pi_{2}ry_{2} ) }{(\alpha +\lambda_{u})[rd+ ( \pi_{1}m+\pi_{2}ry_{2} ) ]},\qquad p_{2}= \frac{m}{r}, \\ & x_{2}=\frac{c}{a} \biggl( \frac{n\lambda \eta_{3}y _{2}}{cp_{2}}-1 \biggr), \end{aligned} \end{aligned}$$
    (14)

    where

    $$ \begin{aligned}&\tilde{A}=\lambda \pi_{2}r, \qquad \tilde{B}=\lambda (rd+ \pi_{1}m)-\frac{ \tilde{\gamma }\omega \pi_{2}r}{\eta_{3}}, \\ & \tilde{C}=-\frac{ \tilde{\gamma }\omega \pi_{1}m}{\eta_{3}}. \end{aligned}$$
    (15)

We note that \(\Omega_{2}\) exists when \(\frac{n\lambda \eta_{3}y_{2}}{cp _{2}}>1\). Now we define

$$ \tilde{\mathcal{R}}_{1}=\frac{n\lambda \eta_{3}y_{2}}{cp_{2}}. $$
(16)

Hence, \(x_{2}=\frac{c}{a} ( \tilde{\mathcal{R}}_{1}-1 ) \). Thus, an infected steady state with antibodies \(\Omega_{2}=(s_{2},u _{2},y_{2}, p_{2},x_{2})\) exists when \(\tilde{\mathcal{R}}_{1}>1\).

Lemma 4

Let \(\tilde{\mathcal{R}}_{0}>1\), then (i) if \(\tilde{\mathcal{R}}_{1}\leq 1\), then \(p_{1}\leq p_{2}\), and (ii) if \(\tilde{\mathcal{R}}_{1}>1\), then \(p_{1}>p_{2}\).

Proof

(i) Let \(\tilde{\mathcal{R}}_{1}\leq 1\), then \(\frac{n\lambda \eta_{3}y_{2}}{cp_{2}}\leq 1\). Using Eq. (14), we obtain

$$ \frac{n\lambda \eta_{3}}{cp_{2}} \biggl( \frac{-\tilde{B}+\sqrt{ \tilde{B}^{2}-4\tilde{A}\tilde{C}}}{2\tilde{A}} \biggr) \leq 1, $$

which gives

$$ \biggl( \frac{2\tilde{A}cp_{2}}{n\lambda \eta_{3}}+\tilde{B} \biggr) ^{2}\geq \tilde{B}^{2}-4\tilde{A}\tilde{C}. $$

Using Eq. (15), we get

$$ \frac{4m^{2}c\pi_{2} ( n\lambda \eta_{3}\pi_{1}+c\pi_{2} ) }{n ^{2}(\eta_{3})^{2}p_{2}} ( p_{2}-p_{1} ) \geq 0. $$

It follows that \(p_{1}\leq p_{2}\). In a similar way, we can prove (ii). □

3.3 Global properties

Theorem 4

The pathogen-free steady state \(\Omega_{0}\) of system (12) is globally asymptotically stable when \(\tilde{\mathcal{R}}_{0}\leq 1\).

Proof

Let \(V_{0}(s,u,y,p,x)\) be given as follows:

$$\begin{aligned} V_{0} =&\tilde{\gamma }s_{0}G \biggl( \frac{s}{s_{0}} \biggr) +\frac{ \eta_{3}\alpha }{\alpha +\lambda_{u}}u+\eta_{3}y+\frac{(1-\tilde{\mathcal{R}}_{02})}{n}p+ \frac{a(1-\tilde{\mathcal{R}}_{02})}{rn}x \\ &{} +\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} \int_{0}^{h_{1}} \Theta_{1}(\tau) \int_{0}^{\tau }\bigl(\pi_{1}s(t-\theta)p(t- \theta)+\pi _{2}s(t-\theta)y(t-\theta)\bigr)\,d\theta \,d\tau \\ &{} +\eta_{3}(1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau) \int_{0}^{ \tau } \bigl( \pi_{1}s(t- \theta)p(t-\theta)+\pi_{2}s(t-\theta)y(t- \theta) \bigr)\,d\theta \,d \tau \\ &{}+\lambda (1-\tilde{\mathcal{R}}_{02}) \int_{0}^{h_{3}}\Theta_{3}( \tau) \int_{0}^{\tau }y(t-\theta)\,d\theta \,d\tau, \end{aligned}$$

where γ̃ is defined by Eq. (13). We get \(V_{0}(s,u,y,p,x)>0\) for all \(s,u,y,p,x>0\), \(V_{0}(s_{0},0, 0,0,0)=0\) and

$$\begin{aligned} \frac{dV_{0}}{dt} =&\tilde{\gamma } \biggl( 1-\frac{s_{0}}{s} \biggr) ( \omega -ds-\pi_{1}sp-\pi_{2}sy) \\ &{} +\frac{\eta_{3}\alpha }{(\alpha +\lambda_{u})} \biggl\{ \rho \int _{0}^{h_{1}}\Theta_{1}(\tau)s(t-\tau) \bigl[\pi_{1}p(t-\tau)+\pi_{2}y(t- \tau)\bigr]\,d\tau -( \alpha +\lambda_{u})u \biggr\} \\ &{} +\eta_{3} \biggl\{ (1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau)s(t- \tau) \bigl[\pi_{1}p(t-\tau)+\pi_{2}y(t-\tau)\bigr]\,d\tau - \lambda y+\alpha u \biggr\} \\ &{} +\frac{(1-\tilde{\mathcal{R}}_{02})}{n} \biggl( n\lambda \int_{0} ^{h_{3}}\Theta_{3}(\tau)y(t-\tau) \,d\tau -cp-apx \biggr) +\frac{a(1-\tilde{\mathcal{R}}_{02})}{rn} ( rpx-mx ) \\ &{} +\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} \int_{0}^{h_{1}} \Theta_{1}(\tau) \bigl\{ s [ \pi_{1}p+\pi_{2}y ] -s(t- \tau) \bigl[ \pi_{1}p(t-\tau)+\pi_{2}y(t-\tau) \bigr] \bigr\} \,d\tau \\ &{} +\eta_{3}(1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau) \bigl\{ s [ \pi_{1}p+\pi_{2}y ] -s(t-\tau) \bigl[ \pi_{1}p(t- \tau)+\pi_{2}y(t- \tau) \bigr] \bigr\} \,d\tau \\ &{} +\lambda (1-\tilde{\mathcal{R}}_{02}) \int_{0}^{h_{3}}\Theta_{3}( \tau) \bigl( y-y(t-\tau) \bigr)\,d\tau \\ =&-\tilde{\gamma }\frac{d(s-s_{0})^{2}}{s}+\tilde{\gamma } ( \pi _{1}s_{0}p+\pi_{2}s_{0}y ) - \frac{c(1-\tilde{\mathcal{R}}_{02})}{n}p-\frac{am(1-\tilde{\mathcal{R}}_{02})}{rn}x-\lambda \eta_{3}\tilde{\mathcal{R}}_{02}y \\ =&-\tilde{\gamma }\frac{d(s-s_{0})^{2}}{s}+\frac{c}{n} ( \tilde{\mathcal{R}}_{0}-1 ) p+\frac{am}{rn} ( \tilde{\mathcal{R}}_{02}-1 ) x. \end{aligned}$$

Thus, \(\frac{dV_{0}}{dt}\leq 0\) when \(\tilde{\mathcal{R}}_{0}\leq 1\) for all \(s,p,x>0\). Similar to Theorem 1, we get \(\frac{dV_{0}}{dt}=0\) at \(\Omega_{0}\). Therefore, \(\Omega_{0}\) is globally asymptotically stable when \(\tilde{\mathcal{R}}_{0}\leq 1\). □

Theorem 5

For system (12), suppose that \(\tilde{\mathcal{R}}_{1}\leq 1<\tilde{\mathcal{R}}_{0}\), then \(\Omega_{1}\) is globally asymptotically stable.

Proof

Let us consider \(V_{1}(s,u,y,p,x)\):

$$\begin{aligned} V_{1} =&\tilde{\gamma }s_{1}G \biggl( \frac{s}{s_{1}} \biggr) +\frac{ \alpha \eta_{3}}{\alpha +\lambda_{u}}u_{1}G \biggl( \frac{u}{u_{1}} \biggr) +\eta_{3}y_{1}G \biggl( \frac{y}{y_{1}} \biggr) \\ &{} +\tilde{\gamma }\frac{\pi_{1}s_{1}p_{1}}{n\lambda \eta_{3}y_{1}}p _{1}G \biggl( \frac{p}{p_{1}} \biggr) +\tilde{\gamma }\frac{a\pi_{1}s _{1}p_{1}}{rn\lambda \eta_{3}y_{1}}x \\ &{}+\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{1}p_{1} \int_{0}^{h_{1}}\Theta_{1}(\tau) \int_{0}^{\tau }G \biggl( \frac{s(t- \theta)p(t-\theta)}{s_{1}p_{1}} \biggr) \,d\theta \,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{1}s_{1}p_{1} \int_{0}^{h_{2}}\Theta_{2}( \tau) \int_{0}^{\tau }G \biggl( \frac{s(t-\theta)p(t-\theta)}{s_{1}p _{1}} \biggr) \,d\theta \,d\tau \\ &{} +\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{1}y_{1} \int_{0}^{h_{1}}\Theta_{1}(\tau) \int_{0}^{\tau }G \biggl( \frac{s(t- \theta)y(t-\theta)}{s_{1}y_{1}} \biggr) \,d\theta \,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{2}s_{1}y_{1} \int_{0}^{h_{2}}\Theta_{2}( \tau) \int_{0}^{\tau }G \biggl( \frac{s(t-\theta)y(t-\theta)}{s_{1}y _{1}} \biggr) \,d\theta \,d\tau \\ &{} +\tilde{\gamma }\pi_{1}s_{1}p_{1} \frac{1}{\eta_{3}} \int_{0}^{h_{3}} \Theta_{3}(\tau) \int_{0}^{\tau }G \biggl( \frac{y(t-\theta)}{y_{1}} \biggr) \,d\theta \,d\tau. \end{aligned}$$

We have \(V_{1}(s,u,y,p,x)>0\) for all \(s,u,y,p,x>0\) and \(V_{1}(s _{1},u_{1},y_{1},p_{1},0)=0\). Calculating \(\frac{dV_{1}}{dt}\), we obtain

$$\begin{aligned} \frac{dV_{1}}{dt} =&\tilde{\gamma } \biggl( 1-\frac{s_{1}}{s} \biggr) ( \omega -ds-\pi_{1}sp-\pi_{2}sy ) \\ &{}+\frac{\eta_{3}\alpha }{\alpha +\lambda_{u}} \biggl( 1- \frac{u_{1}}{u} \biggr) \biggl\{ \rho \int_{0}^{h_{1}}\Theta_{1}( \tau)s(t-\tau) \bigl[ \pi_{1}p(t-\tau)+\pi_{2}y(t-\tau) \bigr]\,d\tau \\ &{} -(\alpha +\lambda_{u})u \biggr\} \\ &{}+\eta_{3} \biggl( 1-\frac{y_{1}}{y} \biggr) \biggl\{ (1-\rho) \int _{0}^{h_{2}}\Theta_{2}(\tau)s(t-\tau) \bigl[\pi_{1}p(t-\tau)+\pi_{2}y(t- \tau)\bigr]\,d\tau \\ &-\lambda y+\alpha u \biggr\} \\ &{}+\tilde{\gamma }\frac{\pi_{1}s_{1}p_{1}}{n\lambda \eta_{3}y_{1}} \biggl( 1-\frac{p_{1}}{p} \biggr) \biggl( n\lambda \int_{0}^{h_{3}}\Theta _{3}(\tau)y(t-\tau) \,d\tau -cp-apx \biggr) \\ & +\tilde{\gamma }\frac{a\pi _{1}s_{1}p_{1}}{rn\lambda \eta_{3}y_{1}} ( rpx-mx ) \\ &{}+\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{1}p_{1} \int_{0}^{h_{1}}\Theta_{1}(\tau)\biggl[ \frac{sp}{s_{1}p_{1}}-\frac{s(t- \tau)p(t-\tau)}{s_{1}p_{1}} +\ln \biggl( \frac{s(t-\tau)p(t-\tau)}{sp} \biggr) \biggr]\,d\tau \\ &{}+\eta_{3}(1-\rho)\pi_{1}s_{1}p_{1} \int_{0}^{h_{2}}\Theta_{2}( \tau)\biggl[ \frac{sp}{s_{1}p_{1}}-\frac{s(t-\tau)p(t-\tau)}{s_{1}p _{1}} \\ &{}+\ln \biggl( \frac{s(t-\tau)p(t-\tau)}{sp} \biggr) \biggr]\,d\tau \\ &{}+\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{1}y_{1} \int_{0}^{h_{1}}\Theta_{1}(\tau)\biggl[ \frac{sy}{s_{1}y_{1}}-\frac{s(t- \tau)y(t-\tau)}{s_{1}y_{1}} +\ln \biggl( \frac{s(t-\tau)y(t-\tau)}{sy} \biggr) \biggr]\,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{2}s_{1}y_{1} \int_{0}^{h_{2}}\Theta_{2}( \tau)\biggl[ \frac{sy}{s_{1}y_{1}}-\frac{s(t-\tau)y(t-\tau)}{s_{1}y _{1}} \\ &{}+\ln \biggl( \frac{s(t-\tau)y(t-\tau)}{sy} \biggr) \biggr]\,d\tau \\ &{} +\tilde{\gamma }\pi_{1}s_{1}p_{1} \frac{1}{\eta_{3}} \int_{0}^{h_{3}} \Theta_{3}(\tau) \biggl[ \frac{y}{y_{1}}-\frac{y(t-\tau)}{y_{1}}+ \ln \biggl( \frac{y(t-\tau)}{y} \biggr) \biggr]\,d\tau. \end{aligned}$$
(17)

Simplifying Eq. (17) and applying the steady state conditions for \(\Omega_{1}\)

$$\begin{aligned}& \omega =ds_{1}+\pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1}, \\ & \frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ) =\alpha \eta_{3}u_{1}, \\ & \frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ) +\eta_{2} \eta_{3}(1-\rho) ( \pi_{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ) =\lambda \eta_{3}y _{1}, \\ & p_{1}=\frac{n\lambda \eta_{3}}{c}y_{1}, \end{aligned}$$

we get

$$\begin{aligned} \frac{dV_{1}}{dt} =&\tilde{\gamma } \biggl( 1-\frac{s_{1}}{s} \biggr) ( ds_{1}-ds ) +\tilde{\gamma } \biggl( 1-\frac{s_{1}}{s} \biggr) ( \pi_{1}s_{1}p_{1}+\pi_{2}s_{1}y_{1} ) \\ &{} +\frac{\eta_{1}\eta _{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi_{1}s_{1}p_{1}+ \pi _{2}s_{1}y_{1} ) \\ &{} -\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} \int_{0}^{h_{1}} \Theta_{1}(\tau) \biggl( \pi_{1}s_{1}p_{1}\frac{s(t-\tau)p(t-\tau)u _{1}}{s_{1}p_{1}u}+ \pi_{2}s_{1}y_{1}\frac{s(t-\tau)y(t-\tau)u_{1}}{s _{1}y_{1}u} \biggr)\,d \tau \\ &{} -\eta_{3}(1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau) \biggl( \pi_{1}s _{1}p_{1}\frac{s(t-\tau)p(t-\tau)y_{1}}{s_{1}p_{1}y}+ \pi_{2}s_{1}y _{1}\frac{s(t-\tau)y(t-\tau)}{s_{1}y} \biggr)\,d \tau \\ &{} +\frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ) +\eta_{2} \eta_{3}(1-\rho) (\pi _{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1}) \\ &{} -\frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{1}p_{1}+ \pi_{2}s_{1}y_{1} ) \frac{uy_{1}}{u_{1}y}+ \tilde{ \gamma }\pi_{1}s_{1}p_{1} \biggl( 1- \frac{1}{\eta_{3}} \int_{0} ^{h_{3}}\Theta_{3}(\tau) \frac{y(t-\tau)p_{1}}{y_{1}p}\,d\tau \biggr) \\ &{} +\tilde{\gamma }\frac{a\pi_{1}s_{1}}{c} ( p_{1}-p_{2} ) x+\frac{ \eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{1}p_{1} \int_{0} ^{h_{1}}\Theta_{1}(\tau)\ln \biggl( {\frac{s(t-\tau)p(t-\tau)}{sp}} \biggr)\,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{1}s_{1}p_{1} \int_{0}^{h_{2}}\Theta_{2}( \tau)\ln \biggl( {\frac{s(t-\tau)p(t-\tau)}{sp}} \biggr)\,d\tau \\ &{} +\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{1}y_{1} \int_{0}^{h_{1}}\Theta_{1}(\tau)\ln \biggl( {\frac{s(t-\tau)y(t- \tau)}{sy}} \biggr)\,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{2}s_{1}y_{1} \int_{0}^{h_{2}}\Theta_{2}( \tau)\ln \biggl( {\frac{s(t-\tau)y(t-\tau)}{sy}} \biggr)\,d\tau \\ &{} +\tilde{\gamma }\pi_{1}s_{1}p_{1} \frac{1}{\eta_{3}} \int_{0}^{h_{3}} \Theta_{3}(\tau)\ln \biggl( {\frac{y(t-\tau)}{y}} \biggr)\,d\tau. \end{aligned}$$

Consider the following equalities with (\(i=1\)):

$$\begin{aligned}& \ln \biggl( {\frac{s(t-\tau)p(t-\tau)}{sp}} \biggr) =\ln \biggl( {\frac{s(t-\tau)p(t-\tau)u_{i}}{s_{i}p_{i}u}} \biggr) +\ln \biggl( { \frac{uy_{i}}{u_{i}y}} \biggr) +\ln \biggl( {\frac{yp_{i}}{y_{i}p}} \biggr) +\ln \biggl( {\frac{s_{i}}{s}} \biggr), \\& \ln \biggl( {\frac{s(t-\tau)p(t-\tau)}{sp}} \biggr) =\ln \biggl( {\frac{s(t-\tau)p(t-\tau)y_{i}}{s_{i}p_{i}y}} \biggr) +\ln \biggl( {\frac{yp_{i}}{y_{i}p}} \biggr) +\ln \biggl( { \frac{s_{i}}{s}} \biggr), \\& \ln \biggl( {\frac{s(t-\tau)y(t-\tau)}{sy}} \biggr) =\ln \biggl( {\frac{s(t-\tau)y(t-\tau)u_{i}}{s_{i}y_{i}u}} \biggr) +\ln \biggl( {\frac{uy_{i}}{u_{i}y}} \biggr) +\ln \biggl( { \frac{s_{i}}{s}} \biggr), \\& \ln \biggl( {\frac{s(t-\tau)y(t-\tau)}{sy}} \biggr) =\ln \biggl( {\frac{s(t-\tau)y(t-\tau)}{s_{i}y}} \biggr) +\ln \biggl( {\frac{s_{i}}{s}} \biggr), \\& \ln \biggl( {\frac{y(t-\tau)}{y}} \biggr) =\ln \biggl( {\frac{y(t- \tau)p_{i}}{y_{i}p}} \biggr) +\ln \biggl( {\frac{y_{i}p}{yp_{i}}} \biggr), \end{aligned}$$
(18)

we obtain

$$\begin{aligned} \frac{dV_{1}}{dt} =&-\tilde{\gamma }\frac{d(s-s_{1})^{2}}{s}-\tilde{\gamma }( \pi_{1}s_{1}p_{1}+\pi_{2}s_{1}y_{1})G \biggl( \frac{s_{1}}{s} \biggr) \\ &{}-\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{1}p_{1} \int_{0}^{h_{1}}\Theta_{1}(\tau)G \biggl( \frac{s(t-\tau)p(t-\tau)u _{1}}{s_{1}p_{1}u} \biggr)\,d\tau \\ &{}-\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{1}y_{1} \int_{0}^{h_{1}}\Theta_{1}(\tau)G \biggl( \frac{s(t-\tau)y(t-\tau)u _{1}}{s_{1}y_{1}u} \biggr)\,d\tau \\ &{} -\eta_{3}(1-\rho)\pi_{1}s_{1}p_{1} \int_{0}^{h_{2}}\Theta_{2}( \tau)G \biggl( \frac{s(t-\tau)p(t-\tau)y_{1}}{s_{1}p_{1}y} \biggr)\,d\tau \\ &{} -\eta_{3}(1-\rho)\pi_{2}s_{1}y_{1} \int_{0}^{h_{2}}\Theta_{2}( \tau)G \biggl( \frac{s(t-\tau)y(t-\tau)}{s_{1}y} \biggr)\,d\tau \\ &{} -\frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}(\pi_{1}s _{1}p_{1}+ \pi_{2}s_{1}y_{1})G \biggl( \frac{y_{1}u}{yu_{1}} \biggr) \\ &{} -\tilde{\gamma }\pi_{1}s_{1}p_{1} \frac{1}{\eta_{3}} \int_{0}^{h_{3}} \Theta_{3}(\tau)G \biggl( \frac{y(t-\tau)p_{1}}{y_{1}p} \biggr)\,d\tau +\tilde{\gamma }\frac{a\pi_{1}s_{1}}{c} ( p_{1}-p_{2} ) x. \end{aligned}$$

From Lemma 4, we have if \(\tilde{\mathcal{R}}_{1}\leq 1\) then \(p_{1} \leq p_{2}\). Thus \(\frac{dV_{1}}{dt}\leq 0\) and \(\frac{dV_{1}}{dt}=0\) occur at the infected steady state without antibodies \(\Omega_{1}\). Thus, \(\Omega_{1}\) is globally asymptotically stable when \(\tilde{\mathcal{R}}_{1}\leq 1\) and \(\tilde{\mathcal{R}}_{0}>1\). □

Theorem 6

For system (12), suppose that \(\tilde{\mathcal{R}}_{1}>1\), then \(\Omega_{2}\) is globally asymptotically stable.

Proof

Define \(V_{2} ( s,u,y,p,x ) \) as follows:

$$\begin{aligned} V_{2} =&\tilde{\gamma }s_{2}G \biggl( \frac{s}{s_{2}} \biggr) +\frac{ \alpha \eta_{3}}{\alpha +\lambda_{u}}u_{2}G \biggl( \frac{u}{u_{2}} \biggr) +\eta_{3}y_{2}G \biggl( \frac{y}{y_{2}} \biggr) \\ &{}+\tilde{\gamma }\frac{\pi_{1}s_{2}p_{2}}{n\lambda \eta_{3}y_{2}}p _{2}G \biggl( \frac{p}{p_{2}} \biggr) +\tilde{\gamma }\frac{a\pi_{1}s _{2}p_{2}}{rn\lambda \eta_{3}y_{2}}x_{2}G \biggl( \frac{x}{x_{2}} \biggr) \\ &{}+\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{2}p_{2} \int_{0}^{h_{1}}\Theta_{1}(\tau) \int_{0}^{\tau }G \biggl( \frac{s(t- \theta)p(t-\theta)}{s_{2}p_{2}} \biggr) \,d\theta \,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{1}s_{2}p_{2} \int_{0}^{h_{2}}\Theta_{2}( \tau) \int_{0}^{\tau }G \biggl( \frac{s(t-\theta)p(t-\theta)}{s_{2}p _{2}} \biggr) \,d\theta \,d\tau \\ &{}+\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{2}y_{2} \int_{0}^{h_{1}}\Theta_{1}(\tau) \int_{0}^{\tau }G \biggl( \frac{s(t- \theta)y(t-\theta)}{s_{2}y_{2}} \biggr) \,d\theta \,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{2}s_{2}y_{2} \int_{0}^{h_{2}}\Theta_{2}( \tau) \int_{0}^{\tau }G \biggl( \frac{s(t-\theta)y(t-\theta)}{s_{2}y _{2}} \biggr) \,d\theta \,d\tau \\ &{} +\tilde{\gamma }\pi_{1}s_{2}p_{2} \frac{1}{\eta_{3}} \int_{0}^{h_{3}} \Theta_{3}(\tau) \int_{0}^{\tau }G \biggl( \frac{y(t-\theta)}{y_{2}} \biggr) \,d\theta \,d\tau. \end{aligned}$$

We have \(V_{2}(s,u,y,p,x)>0\) for all \(s,u,y,p,x>0\), while \(V_{2}(s,u,y,p,x)\) reaches its global minimum at \(\Omega_{2}\). Calculate \(\frac{dV_{2}}{dt}\) as follows:

$$\begin{aligned} \frac{dV_{2}}{dt} =&\tilde{\gamma } \biggl( 1-\frac{s_{2}}{s} \biggr) ( \omega -ds-\pi_{1}sp-\pi_{2}sy ) +\frac{\alpha \eta_{3}}{\alpha +\lambda_{u}} \biggl( 1- \frac{u_{2}}{u} \biggr) \\ &{}\times \biggl\{ \rho \int_{0}^{h_{1}}\Theta_{1}( \tau)s(t-\tau) \bigl[ \pi_{1}p(t-\tau)+\pi_{2}y(t-\tau) \bigr]\,d\tau -( \alpha +\lambda_{u})u \biggr\} \\ &{} +\eta_{3} \biggl( 1-\frac{y_{2}}{y} \biggr) \\ &{}\times \biggl\{ (1-\rho) \int _{0}^{h_{2}}\Theta_{2}(\tau)s(t-\tau) \bigl[\pi_{1}p(t-\tau)+\pi_{2}y(t- \tau)\bigr]\,d\tau - \lambda y+\alpha u \biggr\} \\ &{} +\tilde{\gamma }\frac{\pi_{1}s_{2}p_{2}}{n\lambda \eta_{3}y_{2}} \biggl( 1-\frac{p_{2}}{p} \biggr) \biggl( n\lambda \int_{0}^{h_{3}}\Theta _{3}(\tau)y(t-\tau) \,d\tau -cp-apx \biggr) \\ &{} +\tilde{\gamma }\frac{a\pi_{1}s_{2}p_{2}}{rn\lambda \eta_{3}y _{2}} \biggl( 1-\frac{x_{2}}{x} \biggr) ( rpx-mx ) +\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{2}p_{2} \\ &{}\times \int_{0}^{h_{1}}\Theta_{1}(\tau) \biggl[ \frac{sp}{s_{2}p_{2}}-\frac{s(t- \tau)p(t-\tau)}{s_{2}p_{2}}+\ln \biggl( \frac{s(t-\tau)p(t-\tau)}{sp} \biggr) \biggr]\,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{1}s_{2}p_{2} \\ &{}\times \int_{0}^{h_{2}}\Theta_{2}( \tau) \biggl[ \frac{sp}{s_{2}p_{2}}-\frac{s(t-\tau)p(t-\tau)}{s_{2}p _{2}}+\ln \biggl( \frac{s(t-\tau)p(t-\tau)}{sp} \biggr) \biggr]\,d\tau \\ &{} +\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{2}y_{2} \int_{0}^{h_{1}}\Theta_{1}(\tau) \biggl[ \frac{sy}{s_{2}y_{2}}-\frac{s(t- \tau)y(t-\tau)}{s_{2}y_{2}}+\ln \biggl( \frac{s(t-\tau)y(t-\tau)}{sy} \biggr) \biggr]\,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{2}s_{2}y_{2} \\ &{}\times \int_{0}^{h_{2}}\Theta_{2}( \tau) \biggl[ \frac{sy}{s_{2}y_{2}}-\frac{s(t-\tau)y(t-\tau)}{s_{2}y _{2}}+\ln \biggl( \frac{s(t-\tau)y(t-\tau)}{sy} \biggr) \biggr]\,d\tau \\ &{} +\tilde{\gamma }\pi_{1}s_{2}p_{2} \frac{1}{\eta_{3}} \int_{0}^{h_{3}} \Theta_{3}(\tau) \biggl[ \frac{y}{y_{2}}-\frac{y(t-\tau)}{y_{2}}+ \ln \biggl( \frac{y(t-\tau)}{y} \biggr) \biggr]\,d\tau. \end{aligned}$$
(19)

Simplifying Eq. (19) and applying the steady state conditions for \(\Omega_{2}\)

$$\begin{aligned}& \omega =ds_{2}+\pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2}, \\& \frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) =\alpha \eta_{3}u_{2}, \\& \frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) +\eta_{2} \eta_{3}(1-\rho) ( \pi_{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) =\lambda \eta_{3}y _{2}, \\& p_{2}=\frac{m}{r},\qquad n\lambda \eta_{3}y_{2}=cp_{2}+ap_{2}x_{2}, \end{aligned}$$

we get

$$\begin{aligned} \frac{dV_{2}}{dt} =&\tilde{\gamma } \biggl( 1-\frac{s_{2}}{s} \biggr) ( ds_{2}-ds ) +\tilde{\gamma } \biggl( 1-\frac{s_{2}}{s} \biggr) ( \pi_{1}s_{2}p_{2}+\pi_{2}s_{2}y_{2} ) \\ &{} -\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} \int_{0}^{h_{1}} \Theta_{1}(\tau) \biggl( \pi_{1}s_{2}p_{2}\frac{s(t-\tau)p(t-\tau)u _{2}}{s_{2}p_{2}u}+ \pi_{2}s_{2}y_{2}\frac{s(t-\tau)y(t-\tau)u_{2}}{s _{2}y_{2}u} \biggr)\,d \tau \\ &{} +\frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) \\ &{} -\eta_{3}(1-\rho) \int_{0}^{h_{2}}\Theta_{2}(\tau) \biggl( \pi_{1}s _{2}p_{2}\frac{s(t-\tau)p(t-\tau)y_{2}}{s_{2}p_{2}y}+ \pi_{2}s_{2}y _{2}\frac{s(t-\tau)y(t-\tau)}{s_{2}y} \biggr)\,d \tau \\ &{} +\frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}(\pi_{1}s _{2}p_{2}+ \pi_{2}s_{2}y_{2})+\eta_{1} \eta_{3}(1-\rho) (\pi_{1}s_{2}p _{2}+ \pi_{2}s_{2}y_{2}) \\ &{} -\frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) \frac{uy_{2}}{u_{2}y}+ \tilde{ \gamma }\pi_{1}s_{2}p_{2} \biggl( 1- \frac{1}{\eta_{3}} \int_{0} ^{h_{3}}\Theta_{3}(\tau) \frac{y(t-\tau)p_{2}}{y_{2}p}\,d\tau \biggr) \\ &{} +\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{2}p_{2} \int_{0}^{h_{1}}\Theta_{1}(\tau)\ln \biggl( {\frac{s(t-\tau)p(t- \tau)}{sp}} \biggr)\,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{1}s_{2}p_{2} \int_{0}^{h_{2}}\Theta_{2}( \tau)\ln \biggl( {\frac{s(t-\tau)p(t-\tau)}{sp}} \biggr)\,d\tau \\ &{} +\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{2}s_{2}y_{2} \int_{0}^{h_{1}}\Theta_{1}(\tau)\ln \biggl( {\frac{s(t-\tau)y(t- \tau)}{sy}} \biggr)\,d\tau \\ &{} +\eta_{3}(1-\rho)\pi_{2}s_{2}y_{2} \int_{0}^{h_{2}}\Theta_{2}( \tau)\ln \biggl( {\frac{s(t-\tau)y(t-\tau)}{sy}} \biggr)\,d\tau \\ &{} +\tilde{\gamma }\pi_{1}s_{2}p_{2} \frac{1}{\eta_{3}} \int_{0}^{h_{3}} \Theta_{3}(\tau)\ln \biggl( {\frac{y(t-\tau)}{y}} \biggr)\,d\tau. \end{aligned}$$

Considering equalities (18) with (\(i=2 \)), we obtain

$$\begin{aligned} \frac{dV_{2}}{dt} =&-\tilde{\gamma }\frac{d(s-s_{2})^{2}}{s}-\tilde{\gamma }( \pi_{1}s_{2}p_{2}+\pi_{2}s_{2}y_{2})G \biggl( \frac{s_{2}}{s} \biggr) \\ &{}-\frac{\eta_{3}\alpha \rho }{\alpha +\lambda_{u}}\pi_{1}s_{2}p_{2} \int_{0}^{h_{1}}\Theta_{1}(\tau)G \biggl( \frac{s(t-\tau)p(t-\tau)u _{2}}{s_{2}p_{2}u} \biggr)\,d\tau \\ &{} -\frac{\eta_{3}\alpha \rho }{(\alpha +\lambda_{u})}\pi_{2}s_{2}y _{2} \int_{0}^{h_{1}}\Theta_{1}(\tau)G \biggl( \frac{s(t-\tau)y(t- \tau)u_{2}}{s_{2}y_{2}u} \biggr)\,d\tau \\ &{} -\eta_{3}(1-\rho)\pi_{1}s_{2}p_{2} \int_{0}^{h_{2}}\Theta_{2}( \tau)G \biggl( \frac{s(t-\tau)p(t-\tau)y_{2}}{s_{2}p_{2}y} \biggr)\,d\tau \\ &{} -\eta_{3}(1-\rho)\pi_{2}s_{2}y_{2} \int_{0}^{h_{2}}\Theta_{2}( \tau)G \biggl( \frac{s(t-\tau)y(t-\tau)}{s_{2}y} \biggr)\,d\tau \\ &{} -\frac{\eta_{1}\eta_{3}\alpha \rho }{\alpha +\lambda_{u}} ( \pi _{1}s_{2}p_{2}+ \pi_{2}s_{2}y_{2} ) G \biggl( \frac{y_{2}u}{yu_{2}} \biggr) \\ &{} -\tilde{\gamma }\pi_{1}s_{2}p_{2} \frac{1}{\eta_{3}} \int_{0}^{h_{3}} \Theta_{3}(\tau)G \biggl( \frac{y(t-\tau)p_{2}}{y_{2}p} \biggr)\,d\tau. \end{aligned}$$

Since \(\tilde{\mathcal{R}}_{1}>1\), then \(s_{2}\), \(u_{2}\), \(y_{2}\), \(p_{2}\), and \(x_{2}>0\). We have \(\frac{dV_{2}}{dt}\leq 0\), then following the proof of Theorem 3, one can show that \(\Omega_{2}\) is globally asymptotically stable. □

4 Numerical simulations

We present some numerical simulations to approve our theoretical results of system (2) with parameter values given in Table 1. We consider different initial values:

IV1::

\(\phi_{1}(\theta)=800\), \(\phi_{2}(\theta)=2\), \(\phi_{3}(\theta)=2\), \(\phi_{4}(\theta)=6\), \(\phi_{5}(\theta)=2\),

IV2::

\(\phi_{1}(\theta)=600\), \(\phi_{2}(\theta)=4\), \(\phi_{3}(\theta)=6\), \(\phi_{4}(\theta)=9\), \(\phi_{5}(\theta)=4\),

IV3::

\(\phi_{1}(\theta)=400\), \(\phi_{2}(\theta)=6\), \(\phi_{3}(\theta)=9\), \(\phi_{4}(\theta)=15\), \(\phi_{5}(\theta)=6\),

IV4::

\(\phi_{1}(\theta)=700\), \(\phi_{2}(\theta)=1\), \(\phi_{3}(\theta)=1\), \(\phi_{4}(\theta)=9\), \(\phi_{5}(\theta)=4\), \(\theta \in {}[ -\max \{ \tau_{1},\tau_{2},\tau_{3}\},0]\).

The stability of the steady states will be investigated by varying six parameters r, \(\pi_{1}\), \(\pi_{2}\), \(\tau_{1}\), \(\tau_{2}\), and \(\tau_{3}\) and fixing the other parameters.

Table 1 The data of model (2)

Case (I) Effect of the parameters \(\pi_{1}\), \(\pi_{2}\), and r:

We choose \(\tau_{1}=\tau_{2}=\tau_{3}=0\) and \(\pi_{1}\), \(\pi_{2}\), and r are varied.

Set(1) \(\pi_{1}=0.0001\), \(\pi_{2}=0.0001\), and \(r=0.001\). This yields \(\mathcal{R}_{0}=0.3818<1\) and \(\mathcal{R}_{1}=0.1401<1\). Figure 1 shows that the concentration of susceptible cells increases and tends to the value \(\omega /d=1000\). In addition, the concentrations of infected cells, free pathogens, and antibodies are decreased and tend to zero for IV1–IV3. Therefore, there exists only one steady state that is \(\Omega_{0}\) and it is globally asymptotically stable. This shows the validity of Theorem 1.

Figure 1
figure 1

The simulation of trajectories of system (2) for Case (I)

Set(2) \(\pi_{1}=0.0006\), \(\pi_{2}=0.0006\), and \(r=0.001\). With these values we obtain \(\mathcal{R}_{0}= 2.2909>1\) and \(\mathcal{R} _{1}=0.2367<1\). Figure 1 shows that the solutions of the system tend to the steady state \(\Omega_{1}=(436.5079,5.1227,6.1472,15.368,0)\) for all the three initial values IV1–IV3. Therefore, \(\Omega_{1}\) exists and it is globally asymptotically stable. Hence, the result of Theorem 2 is confirmed.

Set(3) \(\pi_{1}=0.0006\), \(\pi_{2}=0.0006\), and \(r=0.04\) and then \(\mathcal{R}_{0}=2.2909>1\) and \(\mathcal{R}_{1}=2.5285>1\). Figure 1 shows that the solutions of the system approach the steady state \(\Omega_{2}=(768.22,2.1071,2.5285,2.5,30.5702)\) for all the initial values IV1–IV3. Thus, \(\Omega_{2}\) exists and it is globally asymptotically stable. This validates the result of Theorem 3.

Case (II) Effect of time delay parameters:

For this case, we take IV4 and choose the values \(\pi_{1}=0.0006\), \(\pi_{2}=0.0006\), and \(r=0.04\). Let us consider the case \(\tau =\tau_{1}=\tau_{2}=\tau_{3}\). We compute the values of \(\mathcal{R}_{0}\), \(\mathcal{R}_{1}\) and the steady states of system (2) as a function of τ (see Table 2).

Table 2 The values of \(\mathcal{R}_{0}\) and \(\mathcal{R}_{1}\) for system (2) with different values of τ

Table 2 shows that the values of \(\mathcal{R}_{0}\) and \(\mathcal{R} _{1}\) are decreased as τ is increased. Moreover, we have the following cases:

  1. (i)

    \(\Omega_{2}\) exists and it is globally asymptotically stable when \(0\leq \tau <0.368091\);

  2. (ii)

    \(\Omega_{1}\) exists and it is globally asymptotically stable when \(0.368091\leq \tau <0.499367\);

  3. (iii)

    \(\Omega_{0}\) is globally asymptotically stable when \(\tau \geq 0.499367\).

Figure 2 depicts that the numerical results are also compatible with the results of Theorems 13.

Figure 2
figure 2

The simulation of trajectories of system (2) with different values of τ for Case (II)

This means that the time delay can play the role of controller which can be designed to stabilize the system around the pathogen-free steady state \(\Omega_{0}\).

5 Conclusion

In this paper, we have studied two pathogen dynamics models with antibody immune response. Both pathogen-to-susceptible and infected-to-susceptible transmissions have been considered. We have considered two types of infected cells, latently infected cells, and actively infected cells. We have incorporated three types of discrete-time delays and distributed-time delays in the first and second models, respectively. We have shown that the solutions of the system are nonnegative and ultimately bounded, which ensures the well-posedness of the models. For each model, we have derived two threshold parameters \(\mathcal{R}_{0}\) (the basic reproduction number) and \(\mathcal{R} _{1}\) (the antibody response activation number), which fully determine the existence and stability of the three steady states of the model. We have investigated the global stability of all steady states of the model by using Lyapunov method and LaSalle’s invariance principle. We have proven that (i) if \(\mathcal{R}_{0}\leq 1\), then the pathogen-free steady state \(\Omega_{0}\) is globally asymptotically stable; (ii) if \(\mathcal{R}_{1}\leq 1<\mathcal{R}_{0}\), then the infected steady state without antibodies \(\Omega_{1}\) is globally asymptotically stable; and (iii) if \(\mathcal{R}_{1}>1\), then the infected steady state with antibodies \(\Omega_{2}\) is globally asymptotically stable. We have conducted numerical simulations and have shown that both the theoretical and numerical results are consistent.

References

  1. Nowak, M.A., May, R.M.: Virus Dynamics: Mathematical Principles of Immunology and Virology. Oxford University Press, Oxford (2000)

    MATH  Google Scholar 

  2. Shu, H., Wang, L., Watmough, J.: Global stability of a nonlinear viral infection model with infinitely distributed intracellular delays and CTL immune responses. SIAM J. Appl. Math. 73(3), 1280–1302 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  3. Huang, G., Takeuchi, Y., Ma, W.: Lyapunov functionals for delay differential equations model of viral infections. SIAM J. Appl. Math. 70(7), 2693–2708 (2010)

    Article  MathSciNet  MATH  Google Scholar 

  4. Hattaf, K., Yousfi, N.: A generalized virus dynamics model with cell-to-cell transmission and cure rate. Adv. Differ. Equ. 2016, 174 (2016)

    Article  MathSciNet  Google Scholar 

  5. Wang, J., Teng, Z., Miao, H.: Global dynamics for discrete-time analog of viral infection model with nonlinear incidence and CTL immune response. Adv. Differ. Equ. 2016, 143 (2016)

    Article  MathSciNet  Google Scholar 

  6. Elaiw, A.M., Raezah, A.A.: Stability of general virus dynamics models with both cellular and viral infections and delays. Math. Methods Appl. Sci. 40(16), 5863–5880 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  7. Elaiw, A.M., Elnahary, E.K., Raezah, A.A.: Effect of cellular reservoirs and delays on the global dynamics of HIV. Adv. Differ. Equ. 2018, 85 (2018)

    Article  MathSciNet  Google Scholar 

  8. Elaiw, A.M., Hassanien, I.A., Azoz, S.A.: Global stability of HIV infection models with intracellular delays. J. Korean Math. Soc. 49(4), 779–794 (2012)

    Article  MathSciNet  MATH  Google Scholar 

  9. Elaiw, A.M.: Global dynamics of an HIV infection model with two classes of target cells and distributed delays. Discrete Dyn. Nat. Soc. 2012, Article ID 253703 (2012)

    MathSciNet  MATH  Google Scholar 

  10. Elaiw, A.M., AlShamrani, N.H., Hattaf, K.: Dynamical behaviors of a general humoral immunity viral infection model with distributed invasion and production. Int. J. Biomath. 10(3), Article ID 1750035 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  11. Elaiw, A.M., Raezah, A.A., Hattaf, K.: Stability of HIV-1 infection with saturated virus-target and infected-target incidences and CTL immune response. Int. J. Biomath. 10(5), Article ID 1750070 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  12. Elaiw, A.M., AlShamrani, N.H.: Stability of latent pathogen infection model with adaptive immunity and delays. J. Integr. Neurosci. https://doi.org/10.3233/JIN-180087

  13. Kang, C., Miao, H., Chen, X., Xu, J., Huang, D.: Global stability of a diffusive and delayed virus dynamics model with Crowley–Martin incidence function and CTL immune response. Adv. Differ. Equ. 2017, 324 (2017)

    Article  MathSciNet  Google Scholar 

  14. Callaway, D.S., Perelson, A.S.: HIV-1 infection and low steady state viral loads. Bull. Math. Biol. 64, 29–64 (2002)

    Article  MATH  Google Scholar 

  15. Elaiw, A.M., AlShamrani, N.H., Alofi, A.S.: Stability of CTL immunity pathogen dynamics model with capsids and distributed delay. AIP Adv. 7, Article ID 125111 (2017)

    Article  Google Scholar 

  16. Elaiw, A.M.: Global properties of a class of HIV models. Nonlinear Anal., Real World Appl. 11, 2253–2263 (2010)

    Article  MathSciNet  MATH  Google Scholar 

  17. Elaiw, A.M.: Global properties of a class of virus infection models with multitarget cells. Nonlinear Dyn. 69(1–2), 423–435 (2012)

    Article  MathSciNet  MATH  Google Scholar 

  18. Elaiw, A.M., Azoz, S.A.: Global properties of a class of HIV infection models with Beddington–DeAngelis functional response. Math. Methods Appl. Sci. 36, 383–394 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  19. Elaiw, A.M., Almuallem, N.A.: Global dynamics of delay-distributed HIV infection models with differential drug efficacy in cocirculating target cells. Math. Methods Appl. Sci. 39, 4–31 (2016)

    Article  MathSciNet  MATH  Google Scholar 

  20. Li, B., Chen, Y., Lu, X., Liu, S.: A delayed HIV-1 model with virus waning term. Math. Biosci. Eng. 13, 135–157 (2016)

    Article  MathSciNet  MATH  Google Scholar 

  21. Elaiw, A.M., Raezah, A.A., Alofi, B.S.: Dynamics of delayed pathogen infection models with pathogenic and cellular infections and immune impairment. AIP Adv. 8, Article ID 025323 (2018)

    Article  Google Scholar 

  22. Zhang, F., Li, J., Zheng, C., Wang, L.: Dynamics of an HBV/HCV infection model with intracellular delay and cell proliferation. Commun. Nonlinear Sci. Numer. Simul. 42, 464–476 (2017)

    Article  MathSciNet  Google Scholar 

  23. Gómez-Acevedo, H., Li, M.Y.: Backward bifurcation in a model for HTLV-I infection of CD4+ T cells. Bull. Math. Biol. 67(1), 101–114 (2005)

    Article  MathSciNet  MATH  Google Scholar 

  24. Murase, A., Sasaki, T., Kajiwara, T.: Stability analysis of pathogen-immune interaction dynamics. J. Math. Biol. 51, 247–267 (2005)

    Article  MathSciNet  MATH  Google Scholar 

  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. 411, 63–74 (2014)

    Article  MathSciNet  MATH  Google Scholar 

  26. Elaiw, A.M., AlShamrani, N.H.: Global stability of humoral immunity virus dynamics models with nonlinear infection rate and removal. Nonlinear Anal., Real World Appl. 26, 161–190 (2015)

    Article  MathSciNet  MATH  Google Scholar 

  27. Elaiw, A.M., AlShamrani, N.H.: Global properties of nonlinear humoral immunity viral infection models. Int. J. Biomath. 8(5), Article ID 1550058 (2015)

    Article  MathSciNet  MATH  Google Scholar 

  28. Elaiw, A.M., AlShamrani, N.H.: Stability of a general delay-distributed virus dynamics model with multi-staged infected progression and immune response. Math. Methods Appl. Sci. 40(3), 699–719 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  29. Xu, J., Zhou, Y., Li, Y., Yang, Y.: Global dynamics of a intracellular infection model with delays and humoral immunity. Math. Methods Appl. Sci. 39(18), 5427–5435 (2016)

    Article  MathSciNet  MATH  Google Scholar 

  30. Culshaw, R.V., Ruan, S., Webb, G.: A mathematical model of cell-to-cell spread of HIV-1 that includes a time delay. J. Math. Biol. 46, 425–444 (2003)

    Article  MathSciNet  MATH  Google Scholar 

  31. Chen, S.-S., Cheng, C.-Y., Takeuchi, Y.: Stability analysis in delayed within-host viral dynamics with both viral and cellular infections. J. Math. Anal. Appl. 442, 642–672 (2016)

    Article  MathSciNet  MATH  Google Scholar 

  32. Lai, X., Zou, X.: Modeling cell-to-cell spread of HIV-1 with logistic target cell growth. J. Math. Anal. Appl. 426, 563–584 (2015)

    Article  MathSciNet  MATH  Google Scholar 

  33. Yang, Y., Zou, L., Ruan, S.: Global dynamics of a delayed within-host viral infection model with both virus-to-cell and cell-to-cell transmissions. Math. Biosci. 270, 183–191 (2015)

    Article  MathSciNet  MATH  Google Scholar 

  34. Elaiw, A.M., Raezah, A., Alofi, A.: Stability of a general delayed virus dynamics model with humoral immunity and cellular infection. AIP Adv. 7(6), Article ID 065210 (2017)

    Article  Google Scholar 

  35. Elaiw, A.M., Raezah, A., Alofi, A.S.: Effect of humoral immunity on HIV-1 dynamics with virus-to-target and infected-to-target infections. AIP Adv. 6(8), Article ID 085204 (2016)

    Article  Google Scholar 

  36. 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. 315, 516–530 (2017)

    MathSciNet  Google Scholar 

  37. Gibelli, L., Elaiw, A., Alghamdi, M.A., Althiabi, A.M.: Heterogeneous population dynamics of active particles: progression, mutations, and selection dynamics. Math. Models Methods Appl. Sci. 27, 617–640 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  38. Hale, J.K., Lunel, S.M.V.: Introduction to Functional Differential Equations. Springer, New York (1993)

    Book  MATH  Google Scholar 

Download references

Funding

This work was supported by the Deanship of Scientific Research (DSR) at King Abdulaziz University, Jeddah, under grant No. (D-127-130-1438). The authors, therefore, gratefully acknowledge the DSR for technical and financial support.

Author information

Authors and Affiliations

Authors

Contributions

All authors contributed equally to the writing of this paper. All authors read and approved the final manuscript.

Corresponding author

Correspondence to A. D. Hobiny.

Ethics declarations

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hobiny, A.D., Elaiw, A.M. & Almatrafi, A.A. Stability of delayed pathogen dynamics models with latency and two routes of infection. Adv Differ Equ 2018, 276 (2018). https://doi.org/10.1186/s13662-018-1720-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13662-018-1720-x

Keywords