diff --git a/mo_gymnasium/utils.py b/mo_gymnasium/utils.py
index 3fc6d0af..def90471 100644
--- a/mo_gymnasium/utils.py
+++ b/mo_gymnasium/utils.py
@@ -41,7 +41,7 @@ def __init__(self, env: gym.Env, weight: np.ndarray = None):
         gym.utils.RecordConstructorArgs.__init__(self, weight=weight)
         gym.Wrapper.__init__(self, env)
         if weight is None:
-            weight = np.ones(shape=env.reward_space.shape)
+            weight = np.ones(shape=env.unwrapped.reward_space.shape)
         self.set_weight(weight)
 
     def set_weight(self, weight: np.ndarray):
@@ -51,7 +51,7 @@ def set_weight(self, weight: np.ndarray):
             weight: new weights to set
         Returns: nothing
         """
-        assert weight.shape == self.env.reward_space.shape, "Reward weight has different shape than reward vector."
+        assert weight.shape == self.env.unwrapped.reward_space.shape, "Reward weight has different shape than reward vector."
         self.w = weight
 
     def step(self, action: ActType) -> Tuple[ObsType, float, bool, bool, dict]:
@@ -169,7 +169,7 @@ def __init__(
         """
         SyncVectorEnv.__init__(self, env_fns, copy=copy)
         # Just overrides the rewards memory to add the number of objectives
-        self.reward_space = self.envs[0].reward_space
+        self.reward_space = self.envs[0].unwrapped.reward_space
         self._rewards = np.zeros(
             (
                 self.num_envs,
@@ -222,7 +222,7 @@ def __init__(self, env: gym.Env, gamma: float = 1.0, deque_size: int = 100):
         RecordEpisodeStatistics.__init__(self, env, deque_size=deque_size)
         # CHANGE: Here we just override the standard implementation to extend to MO
         # We also take care of the case where the env is vectorized
-        self.reward_dim = self.env.reward_space.shape[0]
+        self.reward_dim = self.env.unwrapped.reward_space.shape[0]
         if self.is_vector_env:
             self.rewards_shape = (self.num_envs, self.reward_dim)
         else:
@@ -344,7 +344,7 @@ def step(self, action):
         Returns:
             Max of the last two observations, reward, terminated, truncated, and info from the environment
         """
-        total_reward = np.zeros(self.env.reward_dim, dtype=np.float32)
+        total_reward = np.zeros(self.env.unwrapped.reward_dim, dtype=np.float32)
         terminated = truncated = False
         info = {}
         for i in range(self._skip):
diff --git a/tests/test_envs.py b/tests/test_envs.py
index 5c3a8bf9..4443789c 100644
--- a/tests/test_envs.py
+++ b/tests/test_envs.py
@@ -100,13 +100,13 @@ def test_env_determinism_rollout(env_spec: EnvSpec):
 
 def _test_reward_bounds(env: gym.Env):
     """Test that the reward bounds are respected."""
-    assert env.reward_dim is not None
-    assert env.reward_space is not None
+    assert env.unwrapped.reward_dim is not None
+    assert env.unwrapped.reward_space is not None
     env.reset()
     for _ in range(NUM_STEPS):
         action = env.action_space.sample()
         _, reward, terminated, truncated, _ = env.step(action)
-        assert env.reward_space.contains(reward)
+        assert env.unwrapped.reward_space.contains(reward)
         if terminated or truncated:
             env.reset()