Epo and Non-hematopoietic Cells: What Do We Know? - Tissue-Protective Cytokines: Methods and Protocols - page 20

TIMP-1 activity and decreased MMP-9 activity in vivo and in vitro,
and can be reversed by inhibition of JAK2 or TIMP-1 ( 103 ).
A couple of studies have recently implicated Epo to be a medi-
ator of the protective effects of nitric oxide (NO) in neurons. Loss
of EpoR coincided with programmed cell death in neurons ( 104 ).
Neuronal NO was induced during hypoxia and correlated with
protection in control cells but not increased in neurons that lacked
the EpoR. However when treated with a neuronal nitric oxide syn-
thase (nNOS) inhibitor the neurons lost their ability to induce
EpoR expression in hypoxia and thus were not protected ( 104 ). In
line with this finding another study demonstrated that nNOS
knockout mice are more susceptible to peripheral neuropathy than
their wild type counterparts due to the absence of NO-mediated
activation of HIF-1 and subsequent downstream neuroprotection
by Epo ( 105 ). Ex vivo experiments showed that protection recov-
ered by using low doses of NOS donors was almost completely
abrogated by Epo siRNA. Thus it appears the neuroprotective
effect of Epo, as well as EpoR expression on neural cells, may also
be regulated by NO.
Intriguingly, what determines the specific pathways activated
by Epo, or the coordination of these multiple cascades, remains till
now unknown.

Different animal models have suggested potential clinical uses of
Epo to combat ischemia or trauma. Cerebroventricular infusion of
Epo was shown to reduce ischemia-induced learning disabilities
and rescue hippocampal CA1 neurons from lethal ischemic damage
in gerbils whereas infusion of EpoR abolished neuroprotection.
In various mouse and rat models of ischemia, intracerebral injection
of Epo also attenuated brain damage by reducing infarct volume
by up to 50% ( 6, 106, 107 ) and improved cognitive function
( 108-110 ). This was further underlined by the fact that cerebral
administration of soluble EpoR reduced the protective effect of
hypoxia preconditioning by up to 80% in other models ( 111, 112 ).
Overall exogenous Epo administration (see Fig. 2 , pathway 6) has
been shown to be protective in multiple cerebral tissue injuries
including neonatal (( 113 ) and reviewed by ref. 114, 115 ) or adult
rodent focal brain ischemia, brain trauma ( 116 ), animal models
of multiple sclerosis ( 117, 118 ) as well as spinal chord injury
( 119, 120 ). Increased oligodendrogenesis and attenuated
proinflammatory cell infiltration was also observed in mouse mod-
els of EAE suggesting Epo positively stimulates oligodendrogen-
esis and reduces the autoimmune response ( 117, 118 ). In the
neonatal brain, Epo significantly reduced white matter damage
during hypoxia/ischemia and increased oligodendrogenesis and
maturation of oligodendrocytes despite being applied in a delayed
manner ( 113 ). Notably, in models of prolonged hypoxia, Epo
secretion from astrocytes was shown to play an important role in

5.3 Neuroprotection
by Epo In Vivo