Let
P
(
x
)
=
a
n
x
n
+
a
n
−
1
x
n
−
1
+
a
n
−
2
x
n
−
2
+
.
.
.
+
a
0
{\displaystyle P(x)=a_{n}x^{n}+a_{n-1}x^{n-1}+a_{n-2}x^{n-2}+...+a_{0}}
, where
a
i
∈
Z
{\displaystyle a_{i}\in \mathbb {Z} }
.
Assume
P
(
p
q
)
=
0
{\displaystyle P({\frac {p}{q}})=0}
for coprime
p
,
q
∈
Z
{\displaystyle p,q\in \mathbb {Z} }
. Therefore,
P
(
p
q
)
=
a
n
(
p
q
)
n
+
a
n
−
1
(
p
q
)
n
−
1
+
a
n
−
2
(
p
q
)
n
−
2
+
.
.
.
+
a
1
(
p
q
)
+
a
0
=
0
{\displaystyle P({\frac {p}{q}})=a_{n}({\frac {p}{q}})^{n}+a_{n-1}({\frac {p}{q}})^{n-1}+a_{n-2}({\frac {p}{q}})^{n-2}+...+a_{1}({\frac {p}{q}})+a_{0}=0}
⇒
a
n
p
n
+
a
n
−
1
p
n
−
1
q
+
a
n
−
2
p
n
−
2
q
2
+
.
.
.
+
a
1
p
q
n
−
1
+
a
0
q
n
=
0
{\displaystyle \Rightarrow a_{n}p^{n}+a_{n-1}p^{n-1}q+a_{n-2}p^{n-2}q^{2}+...+a_{1}pq^{n-1}+a_{0}q^{n}=0}
⇒
p
(
a
n
p
n
−
1
+
a
n
−
1
p
n
−
2
q
+
a
n
−
3
p
n
−
2
q
2
+
.
.
.
+
a
1
q
n
−
1
)
=
−
a
0
q
n
{\displaystyle \Rightarrow p(a_{n}p^{n}-1+a_{n-1}p^{n-2}q+a_{n-3}p^{n-2}q^{2}+...+a_{1}q^{n-1})=-a_{0}q^{n}}
Let
w
=
a
n
p
n
−
1
+
a
n
−
1
p
n
−
2
q
+
a
n
−
3
p
n
−
2
q
2
+
.
.
.
+
a
1
q
n
−
1
∈
Z
{\displaystyle w=a_{n}p^{n}-1+a_{n-1}p^{n-2}q+a_{n-3}p^{n-2}q^{2}+...+a_{1}q^{n-1}\in \mathbb {Z} }
Thus,
w
=
−
a
0
q
n
p
{\displaystyle w=-{\frac {a_{0}q^{n}}{p}}}
As
p
{\displaystyle p}
is coprime to
q
{\displaystyle q}
and
w
∈
Z
.
{\displaystyle w\in \mathbb {Z} .}
, thus
a
0
p
∈
Z
{\displaystyle {\frac {a_{0}}{p}}\in \mathbb {Z} }
.
Again,
a
n
p
n
+
a
n
−
1
p
n
−
1
q
+
a
n
−
2
p
n
−
2
q
2
+
.
.
.
+
a
1
p
q
n
−
1
+
a
0
q
n
=
0
{\displaystyle a_{n}p^{n}+a_{n-1}p^{n-1}q+a_{n-2}p^{n-2}q^{2}+...+a_{1}pq^{n-1}+a_{0}q^{n}=0}
⇒
q
(
a
n
−
1
p
n
−
1
+
a
n
−
2
p
n
−
2
q
+
.
.
.
+
a
1
p
q
n
−
2
+
a
0
q
n
−
1
)
=
−
a
n
p
n
{\displaystyle \Rightarrow q(a_{n-1}p^{n-1}+a_{n-2}p^{n-2}q+...+a_{1}pq^{n-2}+a_{0}q^{n-1})=-a_{n}p^{n}}
Let
(
a
n
−
1
p
n
−
1
+
a
n
−
2
p
n
−
2
q
+
.
.
.
+
a
1
p
q
n
−
2
+
a
0
q
n
−
1
)
=
v
∈
Z
{\displaystyle (a_{n-1}p^{n-1}+a_{n-2}p^{n-2}q+...+a_{1}pq^{n-2}+a_{0}q^{n-1})=v\in \mathbb {Z} }
Thus,
q
v
=
−
a
n
p
n
{\displaystyle qv=-a_{n}p^{n}}
⇒
v
=
−
a
n
p
n
q
{\displaystyle \Rightarrow v=-{\frac {a_{n}p^{n}}{q}}}
As
q
{\displaystyle q}
is coprime to
p
{\displaystyle p}
and
v
∈
Z
.
{\displaystyle v\in \mathbb {Z} .}
, thus
a
n
p
∈
Z
{\displaystyle {\frac {a_{n}}{p}}\in \mathbb {Z} }
.
∴
{\displaystyle \therefore }
For
P
(
x
)
=
a
n
x
n
+
a
n
−
1
x
n
−
1
+
a
n
−
2
x
n
−
2
+
.
.
.
+
a
0
{\displaystyle P(x)=a_{n}x^{n}+a_{n-1}x^{n-1}+a_{n-2}x^{n-2}+...+a_{0}}
, if
P
(
p
q
)
=
0
{\displaystyle P({\frac {p}{q}})=0}
, (where
a
i
∈
Z
{\displaystyle a_{i}\in \mathbb {Z} }
and
a
0
,
a
n
≠
0
{\displaystyle a_{0},a_{n}\neq 0}
) then
(
a
0
)
p
∈
Z
{\displaystyle {\frac {(a_{0})}{p}}\in \mathbb {Z} }
and
a
n
q
∈
Z
{\displaystyle {\frac {a_{n}}{q}}\in \mathbb {Z} }
. [Proved]