Preparation, Characterization and In vitro Evaluation of Novel Drug Delivery System Using As-synthesized, Surface Modified and Regenerated Bacterial Cellulose Cellulose is the most abundant and renewable polymer produced by plants and certain bacterial species, i.e., Agrobacterium, Rhizobium, Sarcina and Gluconacetobacter, and is known as bacterial cellulose (BC). BC is free from pectin, lignin and hemicellulose and chemically identical to plant based cellulose (PC). It is produced as swollen membrane having well organized fibrous network, higher water holding capacity, higher crystallinity and tensile strength, and moldable into desired shape. BC finds various applications in cosmetics, bioelectronics, e-paper preparation, digital displays, biomedical sciences, enzymes immobilization, proteins and drug delivery system. The formulation and manufacture of conventional tablet dosage forms have limitation such as multistep processing, involvement of heavy machinery and labour and time consumption. In addition, numbers of excipients are added in conventional table for design in desired shape and drug delivery, which may have the issue of interactions of excipients with drugs as well other excipients. These interactions may cause processing problems, dosage form instability, poor drug solubility, which lead to poor drug absorption. In the current work, the drug loaded BC matrices for designing novel drug delivery system in as-synthesize, surface modified and regenerated form using famotidine (low water soluble) and tizanidine (highly water soluble) as model drugs. In the first phase of this research studies, BC matrices (12 mm diameter) were prepared and loaded with model drugs. The successful drug loading and uniform distribution into the matrices was confirmed through scanning electron microscopy (SEM) and X-rays diffractometer (XRD). Fourier-transform infrared (FT-IR) spectroscopy and thermogravimetric analysis (TGA) revealed the chemical and thermal stability of the BC-drug matrices, respectively. Percent drug loading of various matrices was in the range of 18.10–67.64%. Similarly, the friability test results were in the range of 0.69–0.83% and 0.14–0.89%, for 20 and 40 mg/ml famotidine loaded matrices respectively, while no weight loss for formulations loaded with 6 mg/ml tizanidine. These weight loss values are below 1%, which is the maximum limit for tablets dosage form as per USP specifications. In-vitro dissolution studies showed more than 80% drug release in the initial 15 min for BC matrices and commercial formulations, following immediate release criteria. In case of in-vitro permeation studies, BC matrices (8 mm diameter) released most of the drug (above 90%) in 10h for famotidine loaded matrices and 8h for tizanidine loaded matrices. In the second phase, BC matrices were surface modified and loaded with model drugs. FT-IR, XRD, SEM and TGA confirmed the successful drug loading, chemical and thermal stability of the drug loaded BC matrices. Percent drug loading was 13.83±1.13% to 50.25±1.82% for famotidine loaded matrices and, 10.16±0.58% to 32.15±4.79% for tizanidine loaded matrices. No drug loss was observed during friability test. The in-vitro dissolution studies using USP type-II dissolution apparatus showed drug release (more than 80%) in 0.5–3h for famotidine loaded matrices and 0.25–0.5h for tizanidine loaded matrices. In the case of permeation studies, all the matrices released most of the drug content in 3h. In the final phase, BC was regenerated using N-methyl-morpholine-oxide (NMMO) solution and loaded with model drug famotidine or tizanidine to prepare regenerated BC (R-BC) matrices. Percent drug loading for R-BC-famotidine matrices was 22.97±0.81% to 27.70±3.24%, and 17.65±1.80% to 28.32±1.00% for R-BC tizanidine matrices, respectively. The friability test data does not show any weight loss from the matrices. Characterization with FT-IR, XRD, SEM and TGA revealed the stability of matrices and successful drug loading. Results of the in-vitro dissolution studies showed drug release (more than 90%) in 0.5h, while Franz cells data revealed that most of the drug (> 90%) was released in 4h for famotidine and 7h for tizanidine, respectively. Various mathematical models including zero order, first order, Higuchi model and Korsmeyer-Peppas model were applied to study the drug release mechanism. The drug release of all the experiment was best fit into first order kinetics model with R2 value greater than 0.997.