Copyright © 2009 by ASME 1  The annulus fibrosus (AF) of the intervertebral disc is a multi$lamellar fibrocartilage that, together with the nucleus pulposus, confers mechanical support and flexibility to the spine. Function of the AF is predicated on a high degree of structural organization over multiple length scales: aligned collagen fibers reside within each lamella, and the direction of alignment alternates between adjacent lamellae from +30 o to $30 o with respect to the transverse axis of the spine. Electrospinning permits fabrication of scaffolds consisting of aligned arrays of nanofibers, and has proven effective for directing the alignment of both cells and extracellular matrix (ECM) deposition [1$3]. We recently employed electrospinning to engineer the primary functional unit of the AF, a single lamella [4]. However, it remains a challenge to engineer a multi$lamellar tissue that replicates the cross$ply fiber architecture of the native AF. Moreover, relatively few studies have considered functional properties of engineered AF, and, when measured, tensile properties of these constructs have been inferior to native AF [4]. In this study, mesenchymal stem cells (MSCs) were seeded onto aligned nanofibrous scaffolds organized into bi$lamellar constructs with opposing or parallel fiber orientations, and their functional maturation was evaluated with time. Additionally, we determined a novel role for lamellar ECM in reinforcing the tensile response of bilayers, and confirmed this mechanism by testing acellular bilayers with controllable interface properties.    Poly(ε$caprolactone) (PCL) was electrospun as described [3] to generate an aligned nanofibrous sheet 250 8m thick. Scaffolds (5 x 30 mm 2 ) were excised with the long axis rotated 30 o from the fiber direction (Fig. 1A).     Scaffolds were coated in fibronectin, seeded with bovine bone marrow derived MSCs and cultured for 2 weeks in a chemically !" !" ## ## $!" $!" $!" $!" ## ## %!" %!" $!" $!" # &&’ ( ) *   ’  !" ’& && +(,  &-&&’   +,) defined growth media containing 10 ng/ml TGF$β3 [1]. Samples were then assembled into bilayers with either parallel (+30 o /+30 o ) or opposing (±30 o ) fiber orientations and cultured between pieces of porous polypropylene (PP) wrapped with foil (F) (Fig. 1B) [5]. After 2 weeks, PP and F were removed. . ’ At 2, 4, 6 and 10 weeks samples tested in uniaxial tension (n=5) followed by biochemical analyses to determine glycosaminoglycan (GAG), collagen, and DNA contents [1]. Modulus was computed from the slope of the linear portion of the stress$strain curve. Additional samples (n=3) were cryotomed to 8 Dm across the fiber plane, and stained with Picrosirius Red (PR). Polarized light microscopy was used to visualize and quantify collagen alignment [6]. (/ ’ An acellular study was carried out to relate the tensile behavior of nanofibrous bilayers to the properties of the interlamellar matrix. Aligned PCL scaffolds ~1 mm thick were fabricated (as above) and formed into bilayers by applying molten agarose to the interlamellar space and allowing it to set. Agarose concentration was varied to alter material properties of the interface [7,8]. Parallel and opposing bilayers were tested in uniaxial tension as above, for agarose concentrations of 2, 4, 5, and 6 % in PBS. ’’ Statistical            ! " #"$ % &"% ’"  ( 0""1% 0"23*0 4(5  55  ( 5((   5%(5  (    (5   5) /6 /6   ) 7 8 ) /. 9 )   5) /6 )"* +," !" &"-"* (.*  */."" /"/," 